{
  "status": "completed",
  "completed_count": 57,
  "applicable_count": 57,
  "failed_count": 0,
  "schema_version": 8,
  "config_hash": "f12296572b3e973d",
  "data": {
    "scenes": [
      {
        "scene_index": 1,
        "heading": "S#1. 숲속 / 해 질 무렵",
        "beat_title": "어둠 속의 습격과 공포",
        "representative_moment": "겁에 질려 위를 올려다보는 C06의 얼굴",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Low angle observer framing from just below the chin, shallow depth of field on the overhead canopy, dim blue-green dusk light with humid low contrast",
            "t2i_prompt": "Photorealistic cinematic still. Low angle observer framing from just below the chin, with rain-damp foliage brushing the left and right foreground edges. C06O11 in a bloody hiking jacket, a Korean man in his 20s with a young face, black hair, and dark eyes, his neck and shoulder edge occupying the lower frame. He is positioned at a three-quarter angle from the left, his jaw lifted upward, dark eyes fixed on the darkness above in terror. Focus on his trembling lower eyelid and muddy cheek. Dim blue-green dusk light seeps through the overhead canopy, washing his skin with humid low contrast, while faint red-brown blood tones pool on his cheek. The background consists of the immediate darkening tree canopy dissolving into shallow depth of field in the upper negative space, with damp torn grass visible slightly behind him.",
            "outfit_assignments": [
              {
                "character_id": "C06",
                "outlook_id": "O11"
              }
            ],
            "source_facts": [
              "Scene heading states: 숲속 / 해 질 무렵",
              "Shot 2 mentions: 잔뜩 겁에 질린 남자 표정(T.S)"
            ],
            "visual_inferences": [
              "fixed_element dusk_forest_environment: damp torn grass, mud, and dim blue-green dusk light",
              "Low angle framing emphasizes his vulnerability to an unseen threat above"
            ],
            "creative_decisions": [
              "Rain-damp leaves brushing the foreground edges to create a claustrophobic, through-object view",
              "Shallow depth of field on the upper canopy directs all visual attention to the terror in his expression"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "c9921b96ac0eb720",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "1d13fa52042f2e56",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "Not dependent on prior visual scenes; establishes the initial moment of terror",
        "_shot_index": 2,
        "visible_entities": [
          "C06"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 1,
            "shot_index": 2
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A tighter-than-stage CU low angle through-object frame holds less than a foot from the 20대 한국인 남자’s face, crouched just under his chin as wet leaves cut across the left and right edges. The man’s face occupies the center-left 70% of frame, his muddy cheek and trembling lower eyelid sharp while the foliage and dusk canopy above dissolve into shallow DOF; this is an observer view, not his POV, watching him look upward in terror.",
            "lighting_mood": "Dim blue-green dusk light seeps through the canopy, with humid low contrast and faint red-brown blood tones warming the man’s skin.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C06"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C06",
                "outlook_id": "O11"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": null,
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "dusk_forest_environment",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A dense, humid forest setting at dusk. The ground and immediate surroundings are covered in damp, crushed leaves, torn grass, and patches of dark mud. The ambient lighting is dim and filtered, casting deep blue-green and damp green-brown tones across the vegetation.",
                "applies_to_shots": [
                  2,
                  4
                ],
                "source_facts": [
                  "Scene heading states: 숲속 / 해 질 무렵 (Forest / Dusk)",
                  "Shot 2 mentions wet leaves, foliage, dusk canopy, and dim blue-green dusk light.",
                  "Shot 4 mentions torn grass, mud, crushed leaves, and damp green-brown forest tones."
                ],
                "visual_inferences": [
                  "The ground remains consistently muddy and disturbed from the chase and the ensuing physical struggle."
                ],
                "creative_decisions": [
                  "Emphasized the damp, blue-green and muddy brown color palette to visually unify the background across different camera framing distances."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C06O11",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "77c410345b7119ea"
      },
      {
        "scene_index": 1,
        "heading": "S#1. 숲속 / 해 질 무렵",
        "beat_title": "검은 인영의 기습과 짓눌림",
        "representative_moment": "검은 인영(C10)이 바닥에 쓰러진 C06의 상체를 강하게 덮치며 짓누르는 찰나의 순간.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Ground-level handheld medium framing, dim blue-green dusk lighting",
            "t2i_prompt": "Photorealistic cinematic still. A ground-level medium framing on the frontal axis, capturing a mid-impact struggle. C10O14 in a dark tattered cloak, a humanoid figure, with back to camera, lunging down hard from above and pinning the other figure down. The humanoid silhouette appears as a dark mass partially blocking the view, body volume preserved with faint rim light tracing the shoulders, its face fully swallowed in shadow with no discernible features. Beneath it, C06O11 in a bloody hiking jacket, an East Asian man in his 20s, is pinned on his back in the lower half of the frame, facing the camera with eyes fixed upward. Both figures share the damp muddy forest ground as their physical anchor. Focus on the forceful contact point where the dark figure's weight presses the fallen man's shoulders into the mud. A mud-smeared shoe rests near the frame edge, toe angled away from the camera, heel turned toward the impact area. The background is a dense, humid forest setting with damp, crushed leaves and torn grass. Lit by dying dusk filtered through leaves, casting deep blue-green and damp green-brown tones across the vegetation.",
            "outfit_assignments": [
              {
                "character_id": "C06",
                "outlook_id": "O11"
              },
              {
                "character_id": "C10",
                "outlook_id": "O14"
              }
            ],
            "source_facts": [
              "S#1 action: '어디선가 쏜살같이 달려온 검은 인영이 남자를 덮친다.'",
              "shot description: A ground-level MWS eye-level-to-the-fallen-body handheld frame lands on the frontal axis",
              "shot description: The 검은 인영 crashes into the center foreground, its dark mass partially blocking the man beneath"
            ],
            "visual_inferences": [
              "fixed_element dusk_forest_environment: dense, humid forest setting at dusk with deep blue-green and green-brown tones",
              "mid-impact physical struggle with C10 lunging down (back to camera) and C06 pinned on his back (facing camera)",
              "mud-smeared shoe positioned near the frame edge to indicate the aftermath of the chase"
            ],
            "creative_decisions": [
              "Used ground-level framing to anchor both characters to the muddy impact surface and emphasize the power imbalance",
              "Applied faint rim light to C10's shoulders to maintain body volume and avoid a flat cutout, while strictly adhering to the face-obscured trait"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "361bddd90992e48b",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "09cbdb5ddc8f21e8",
              "validator": "scene_detail_owned_objects.v1",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 1,
        "dependency_reason": "Shot 2의 숲속 배경(dusk_forest_environment) 연속성 유지 (exact_background)",
        "_shot_index": 4,
        "visible_entities": [
          "C06",
          "C10",
          "L01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 1,
            "shot_index": 4
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A ground-level MWS eye-level-to-the-fallen-body handheld frame lands on the frontal axis about three feet from the 20대 한국인 남자’s upper body at the exact impact. The 검은 인영 crashes into the center foreground, its dark mass partially blocking the man beneath; the man’s shoulders and face are pinned in the lower half while torn grass, mud, and crushed leaves shake around the frame.",
            "lighting_mood": "The handheld impact is lit by dying dusk filtered through leaves, with the silhouette nearly black against damp green-brown forest tones.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C06",
              "C10",
              "L01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C06",
                "outlook_id": "O11"
              },
              {
                "character_id": "C10",
                "outlook_id": "O14"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "not_applicable",
            "bg_id": null,
            "reference_usage": "none",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": false,
            "constraints": [
              "no chain_bg is bound to this shot — describe the location freely without referencing a background image",
              "do not use 'from the reference' / 'the existing room' / 'preserving the room' style phrasing — there is no reference image to refer to"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "dusk_forest_environment",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A dense, humid forest setting at dusk. The ground and immediate surroundings are covered in damp, crushed leaves, torn grass, and patches of dark mud. The ambient lighting is dim and filtered, casting deep blue-green and damp green-brown tones across the vegetation.",
                "applies_to_shots": [
                  2,
                  4
                ],
                "source_facts": [
                  "Scene heading states: 숲속 / 해 질 무렵 (Forest / Dusk)",
                  "Shot 2 mentions wet leaves, foliage, dusk canopy, and dim blue-green dusk light.",
                  "Shot 4 mentions torn grass, mud, crushed leaves, and damp green-brown forest tones."
                ],
                "visual_inferences": [
                  "The ground remains consistently muddy and disturbed from the chase and the ensuing physical struggle."
                ],
                "creative_decisions": [
                  "Emphasized the damp, blue-green and muddy brown color palette to visually unify the background across different camera framing distances."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 1,
                "shot_index": 2,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C06O11",
                "policy": "required"
              },
              {
                "kind": "character_outlook",
                "id": "C10O14",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "0a533ce80c7d1e6b"
      },
      {
        "scene_index": 2,
        "heading": "S#2. 도시 뒷골목 / N",
        "beat_title": "가로등 아래 괴물로 변한 남자의 공격",
        "representative_moment": "어두운 골목길에서 붉은 눈을 번뜩이며 C07의 목을 강하게 물어뜯는 C06의 클로즈업.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Eye-level head-and-shoulder CU OTS composition, claustrophobic and humid atmosphere with dim sodium streetlight and ominous red eye-glow.",
            "t2i_prompt": "Photorealistic cinematic still. An eye-level head-and-shoulder CU OTS composition. In the foreground left, C07O13 in an ivory silk blouse, Korean, 20대 여성, 젊은 여성 얼굴, 검은 머리, 어두운 눈동자, with back to camera forming a blurred foreground wedge, her face partially visible with eyes closed, neck exposed limp as the figure receiving the force. The shared physical contact point is at her neck. C06O12 in a black hood zip-up, Korean, 20대 남성, 젊은 남성 얼굴, 검은 머리, his typically 어두운 눈동자 transformed into glowing red eyes, positioned at three-quarter angle from the right, leaning forward as the figure exerting force. His teeth are pressed firmly and forcefully into the exposed side of her neck at the lower right, his expression intense, eyes fixed on the void. Focus on his glowing red eyes and the decisive physical contact point. [: close frame edge bounded by a dark weathered utility pole behind the bodies, its cylindrical side catching a thin rim of dim practical sodium streetlight, tangled cables climbing upward. A wet alley surface implied at the lower edge reflects broken amber light and a faint red glow pooling on the slick texture.] Black shadows swallow the alley edges, creating a humid, claustrophobic atmosphere. Cool red ambient light reflects smoothly on his cheek and jawline.",
            "outfit_assignments": [
              {
                "character_id": "C06",
                "outlook_id": "O12"
              },
              {
                "character_id": "C07",
                "outlook_id": "O13"
              }
            ],
            "source_facts": [
              "S#2 action: 남자의 붉은 눈동자, 목의 살점을 뜯는 이빨… 푹 힘없이 쓰러지는 여자.",
              "Shot description: Close-up of the Korean man's face with glowing red eyes, biting the woman's neck rigidly.",
              "DP notes: observer POV, OTS-like frame just off the woman's shoulder line."
            ],
            "visual_inferences": [
              "The woman's shoulder forms a blurred foreground wedge, ensuring she is anchored in the same space but keeping focus on the man.",
              "Red glow from his eyes casts a faint red reflection on his cheek and the wet pavement below.",
              "A weathered utility pole forms the background boundary directly behind them to maintain close framing without wide room references."
            ],
            "creative_decisions": [
              "Applied 'figure exerting force' and 'figure receiving the force' to accurately reflect the violent physical constraint without triggering intimate/romantic misinterpretations.",
              "Used [L02] instead of [L01] to omit any prior broad background references, focusing solely on the immediate frame-edge textures (utility pole and slick ground).",
              "Retained exact stable_traits tokens ('검은 머리', '어두운 눈동자' etc.) while detailing the glowing red eye transformation."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "8714c6b95842b9df",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "25467122504194fc",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "none",
        "_shot_index": 2,
        "visible_entities": [
          "C06",
          "C07"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 2,
            "shot_index": 2
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "An eye-level CU OTS-like frame sits just off the 20대 한국인 여자’s shoulder line, isolating the 20대 한국인 남자’s glowing red eyes, cheek, mouth, and the exposed side of her neck. The man’s face fills the upper center, his teeth pressed into the neck at lower right, while the woman’s shoulder and hair form a blurred foreground wedge on the left; the camera is almost motionless at the peak of the approach.",
            "lighting_mood": "Sodium streetlight and sickly red eye-glow create a humid, claustrophobic palette with black shadows swallowing the alley edges.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C06",
              "C07"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C06",
                "outlook_id": "O12"
              },
              {
                "character_id": "C07",
                "outlook_id": "O13"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": null,
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C06O12",
                "policy": "required"
              },
              {
                "kind": "character_outlook",
                "id": "C07O13",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "77f0b3deb1bd1cde"
      },
      {
        "scene_index": 3,
        "heading": "S#3. 현장 / D",
        "beat_title": "뉴스 화면 속 폴리스 라인과 수사 지휘",
        "representative_moment": "폴리스 라인 너머로 흰색 덮개(P10)가 덮인 시신을 둘러싸고 혜수(C02)와 형사들이 사건 현장을 감식하는 뉴스 보도 화면 정지컷.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Static eye-level news-camera composition, muted news-feed gray-green grading with low contrast and humid haze",
            "t2i_prompt": "Photorealistic cinematic still. A static eye-level news-camera composition holds about 15 meters outside P09, the taut rain-damp police line tape acting as a sharp foreground boundary across the lower third, its printed warning side facing the camera. Beyond the tape, focus on P10, a white body cover stained with moisture, forming a low mound on the ground and lying center frame in [L01: a damp, leaf-covered forest crime scene marked by footprints]. Gathering around the shared spatial anchor of P10, C02O05 in a charcoal gray suit jacket, a middle-aged Asian woman with black hair and dark eyes, stands at a three-quarter angle from the left, bending over a clipboard, eyes fixed on the covered body. Across the same tarp from her, an Asian man in a detective coat is in right profile, crouching near P10, eyes fixed on the covered body. They form a frozen broadcast-like tableau among the investigators. Overcast daylight is flattened into a muted news-feed gray-green, with humid haze and low contrast making P10 feel unnaturally stark. Slight digital compression artifacts and a faint broadcast border surround the image frame.",
            "outfit_assignments": [
              {
                "character_id": "C02",
                "outlook_id": "O05"
              }
            ],
            "source_facts": [
              "S#3 action: '현장을 감식하는 경찰, 폴리스 라인을 치고 시신을 수거한다. 형사무리 속에 사건을 지휘하는 혜수의 모습이 보인다.'",
              "shot description: A static EWS eye-level through-device news-camera composition holds about 15 meters outside the police line... Beyond it, the covered body lies center frame",
              "DP notes: police line tape taut, rain-damp; news image frame edge indicating a TV news feed"
            ],
            "visual_inferences": [
              "fixed_element exact_background: damp forest crime scene from the prior shot is preserved as the base location",
              "C02 bends over a clipboard as a gesture of leading the investigation",
              "The unregistered detective crouches near the tarp, maintaining visual focus on the body to reinforce the crime scene context"
            ],
            "creative_decisions": [
              "Applied muted news-feed gray-green grading and compression artifacts to simulate the broadcast look directly on the image surface without declaring an artificial TV screen inside a room.",
              "Placed P09 across the lower third to act as a sharp visual barrier and depth anchor for the eye-level composition.",
              "Used explicit shared spatial anchor phrasing ('Gathering around the shared spatial anchor of P10', 'Across the same tarp') to firmly establish foreground/background depth consistency."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "9f028ee2b8dd3f24",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "b49efe6cdb275c94",
              "validator": "scene_detail_owned_objects.v1",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 1,
        "dependency_reason": "scene_consistency fixed_elements: exact_background from Scene 1 Shot 4 (forest crime scene) requires continuity of the established background environment.",
        "_shot_index": 1,
        "visible_entities": [
          "C02",
          "P09",
          "P10",
          "L01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 3,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A static EWS eye-level through-device news-camera composition holds about 15 meters outside the police line, using the tape as a sharp foreground boundary across the lower third. Beyond it, the covered body lies center frame on the forest floor while 혜수 and 김형사 are small figures among investigators, arranged in a frozen broadcast-like tableau with no reporter visible.",
            "lighting_mood": "Overcast daylight is flattened into a muted news-feed gray-green, with humid haze and low contrast making the white cover feel unnaturally stark.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C02",
              "P09",
              "P10",
              "L01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C02",
                "outlook_id": "O05"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "not_applicable",
            "bg_id": null,
            "reference_usage": "none",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": false,
            "constraints": [
              "no chain_bg is bound to this shot — describe the location freely without referencing a background image",
              "do not use 'from the reference' / 'the existing room' / 'preserving the room' style phrasing — there is no reference image to refer to"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [
              {
                "scene_index": 1,
                "shot_index": 4,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C02O05",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "3656f096cfd4aab5"
      },
      {
        "scene_index": 4,
        "heading": "S#4. 옥탑방 / D",
        "beat_title": "불안감이 감도는 인천 변두리 옥탑방 전경",
        "representative_moment": "맑은 하늘 아래 낡은 빌라 옥탑방과 주변 골목이 내려다보이는 부감 쇼트.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "high angle EWS bird's eye establishing shot, 45-degree slant, muted coastal haze",
            "t2i_prompt": "Photorealistic cinematic still. A high angle bird's eye establishing shot looks down at a 45-degree slant over the coastal outskirts. The framing matches the wide aerial reference exactly, capturing the rooftop deck from the reference positioned slightly right of center. The neighboring roofs from the reference and the narrow alley from the reference form staggered blocks beneath the camera, occupying the lower and middle frame. The rooftop unit from the reference and the stair bulkhead from the reference sit quietly on the flat surface. Clear daylight is tempered into a muted coastal haze, casting pale blues and dim concrete grays across the structures. The sky occupies the upper frame, emphasizing scale and an unsettling stillness. No characters are present.",
            "outfit_assignments": [],
            "source_facts": [
              "S#4 action: 인천 변두리 부감, 오래된 빌라 옥탑방 전경.",
              "shot description: 맑은 하늘 아래 인천 변두리의 낡은 빌라 옥탑방과 주변 전경이 내려다보이는 부감 쇼트."
            ],
            "visual_inferences": [
              "No characters are present in this establishing exterior shot",
              "Coastal humidity softens the daylight into pale blues and grays"
            ],
            "creative_decisions": [
              "Explicitly anchored owned objects (rooftop deck, neighboring roofs, alley, rooftop unit, stair bulkhead) with 'from the reference' to prevent double generation",
              "Emphasized the 45-degree high angle bird's eye slant to establish spatial scale and isolation"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "8febff83d2054535",
              "owned_hash": "72d8385eab7418f1",
              "camera_direction_hash": "a107c71c016f43e0",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "rooftop deck",
                  "violating_phrase": "the rooftop deck from the reference",
                  "reason": "The owned object is mentioned with an explicit reference anchor, which indicates the existing background object rather than a new redraw instruction.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "neighboring roof",
                  "violating_phrase": "The neighboring roofs from the reference",
                  "reason": "The owned object appears in a reference-anchored description, so it is treated as an existing element to retain.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "alley",
                  "violating_phrase": "the narrow alley from the reference",
                  "reason": "The owned object is only used as a reference-anchored spatial element, not as the direct object of a redraw verb.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "rooftop unit",
                  "violating_phrase": "The rooftop unit from the reference",
                  "reason": "This is a reference-anchored mention of an existing background object, not a command to create a new one.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "stair bulkhead",
                  "violating_phrase": "the stair bulkhead from the reference",
                  "reason": "The phrase explicitly ties the object to the reference, so it is an anchor reference rather than a redraw request.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "장소와 분위기를 보여주는 독립적인 인서트/장소 제시 샷",
        "_shot_index": 1,
        "visible_entities": [
          "L03"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 4,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A high angle EWS bird's eye establishing shot looks down from roughly 90 meters at a 45-degree slant over the Incheon outskirts, holding before any descent begins. The old rooftop room sits slightly right of center among low-rise villas and narrow lanes, occupying only a small portion of the frame so the surrounding neighborhood and clear sky establish scale and unease.",
            "lighting_mood": "Clear daylight is tempered into a muted coastal haze, bright but not cheerful, with pale blues and concrete grays dominating.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "L03"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L03B05",
            "reference_usage": "exact_background",
            "owned_objects": [
              "AC condenser",
              "alley",
              "antenna",
              "building",
              "circle mark",
              "door",
              "drain",
              "entrance canopy",
              "laundry line",
              "neighboring roof",
              "parapet",
              "parking strip",
              "pipe",
              "rooftop deck",
              "rooftop unit",
              "satellite dish",
              "stair bulkhead",
              "water tank",
              "window"
            ],
            "camera_reference": {
              "camera_position": "above the site centered over number 2 (rooftop deck), facing down to include number 1 (old villa footprint), number 3 (rooftop unit), and surrounding numbers 14, 15, and 16",
              "camera_height": "high aerial overview, map-like reference height",
              "lens_hint": "wide aerial framing, equivalent 28mm",
              "framing_notes": "Use as a layout reference for a day aerial view; keep the rooftop unit and stair bulkhead readable while showing the surrounding alley, parking strip, and neighboring roofs."
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "when focusing on an owned object, anchor it explicitly: 'the door from the reference'",
              "camera_reference specifies the framing under which the background was painted — match camera_position, camera_height, and framing",
              "if deviating from the reference framing (push-in, angle shift, ECU), state the deviation explicitly so the compositor can account for the scale difference"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "background",
                "id": "L03B05",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "1794552f83451123"
      },
      {
        "scene_index": 5,
        "heading": "S#5. 옥탑방 안 / 실내",
        "beat_title": "황급히 TV를 끄는 민숙",
        "representative_moment": "딸이 나오자 놀란 민숙이 황급히 TV 리모컨을 든 손을 뻗은 채 멈칫하는 순간.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Eye-level medium close-up, through-object push-in resolving into a hold on the subject's face, with the foreground hand slightly out of focus. Cold TV flicker mixed with muted daylight.",
            "t2i_prompt": "Photorealistic cinematic still. C03O06 in a gray homewear and kitchen apron, an East Asian woman in her early 40s, occupies the center-right in an eye-level medium close-up, her shoulders and upper body visible. She is seen at a three-quarter angle from the right, her startled eyes snapping away from the lens toward an off-frame doorway. Her outstretched arm extends into the lower-left foreground, hand gripping a television remote. A gray homewear sleeve cuff borders her exposed wrist, where a distinct, small, circular reddish mark is clearly visible on the skin. Focus is firmly on her tense facial expression. Cold, flickering blue light from the off-camera television washes across the left side of her face and outstretched arm, contrasting with dim ambient light. [L04: The soft, out-of-focus background features a dim kitchenette counter, the metal spout of a running sink faucet, and a steaming pot on the stove, illuminated by a hazy closed window filtering pale daylight.]",
            "outfit_assignments": [
              {
                "character_id": "C03",
                "outlook_id": "O06"
              }
            ],
            "source_facts": [
              "Scene action: '딸, 수리영이 방문을 열고... 나오자 후다닥 TV를 끄는 민숙.'",
              "DP instruction: 'looking past 민숙’s outstretched remote hand in the near foreground toward her startled face'",
              "DP instruction: 'eyes snapping toward 수리영’s doorway rather than the lens'"
            ],
            "visual_inferences": [
              "fixed_element: Minsuk wears casual domestic clothes with a kitchen apron, and a distinct circular reddish mark is present on her wrist.",
              "forward_zoom_target: The gray homewear sleeve cuff and the circular mark on the wrist must be clearly visible on the outstretched arm to set up a subsequent close-up."
            ],
            "creative_decisions": [
              "Placed the hand gripping the remote in the lower-left foreground to establish depth and emphasize the sudden action of turning off the TV.",
              "Used cold, flickering blue light to simulate the TV screen's illumination, contrasting with the hazy, humid daylight of the background room."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "2c24eb65278ba2f8",
              "owned_hash": "336090adaa16cf9a",
              "camera_direction_hash": "0eecdac105ac324f",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "N/A",
        "_shot_index": 3,
        "visible_entities": [
          "C03",
          "L04"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 5,
            "shot_index": 3
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "From the television side of the room, an MCU eye-level through-object push-in resolves into a hold, looking past 민숙’s outstretched remote hand in the near foreground toward her startled face. The remote and fingers sit in the lower-left foreground at natural scale, while 민숙 occupies the center-right with the kitchenette soft behind her, her eyes snapping toward 수리영’s doorway rather than the lens.",
            "lighting_mood": "Thin daylight from the small window mixes with cold TV flicker, keeping the rooftop room muted and humid with a faint domestic warmth under the alarm.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C03",
              "L04"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C03",
                "outlook_id": "O06"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B01",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "minsuk_attire_and_mark",
                "element_type": "character_state",
                "character_name": "Minsuk",
                "description": "A 40-year-old East Asian woman wearing casual domestic clothes with a kitchen apron tied around her body. A distinct, small, circular reddish mark is permanently present on her wrist skin.",
                "applies_to_shots": [
                  3,
                  7
                ],
                "source_facts": [
                  "Scene text action: '앞치마를 두른 채' (wearing an apron)",
                  "Scene text action: '민숙 손목에 보이는 작은 원(이하 표식)' (small circular mark on Minsuk's wrist)",
                  "Shot 3: MCU push-in framing of Minsuk",
                  "Shot 7: CU profile framing showing Minsuk's upper body in the shallow plane"
                ],
                "visual_inferences": [
                  "The apron and casual clothes are consistently worn from the living room area to the entryway.",
                  "The circular mark on the wrist is an inherent, unmoving physical trait that belongs to her overall character model."
                ],
                "creative_decisions": [
                  "Merged the persistent attire and physical mark into a single 'full/upper body' character state for Shots 3 and 7.",
                  "Excluded Shot 9 from fixed elements because it is a solitary extreme close-up, adhering to the rule to prevent conflicting zoom descriptions."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [
              {
                "scene_index": 5,
                "shot_index": 9,
                "description": "민숙의 손목에 있는 또렷한 둥근 표식 클로즈업.",
                "keep_elements": [
                  {
                    "label": "woman's wrist with a distinct circular reddish mark",
                    "kind": "immobilized_character"
                  },
                  {
                    "label": "gray homewear sleeve cuff bordering the wrist",
                    "kind": "static_prop"
                  }
                ]
              }
            ],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C03O06",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "fca6816b2f9141e7"
      },
      {
        "scene_index": 5,
        "heading": "S#5. 옥탑방 안 / 실내",
        "beat_title": "외출 전 엄마의 온기를 느끼는 수리영",
        "representative_moment": "외출 전 수리영이 눈을 감고 민숙의 앞치마 옷자락을 쥐며 기대어 선다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Slightly high angle head-and-shoulder composition, shallow depth of field, warm gray-beige tones",
            "t2i_prompt": "Photorealistic cinematic still. A slightly high angle head-and-shoulder composition from an observer POV. C01O02 in a beige hoodie, an East Asian woman in her early 20s, in left profile with eyes closed, leans her face gently against the soft, lightly wrinkled fabric of C03O06's apron. C01O02's hand is visible in the lower center, clutching the fabric. C03O06 in a kitchen apron over homewear, an East Asian woman in her early 40s, stands just beyond her at a three-quarter angle from the left in the shallow depth of field, her upper body and shoulder partially visible. C03O06 wears a soft smile paused on her face, eyes fixed on C01O02. Soft room light and faint daylight flatten into warm gray-beige tones, illuminating the intimate gesture. Focus on C01O02's closed-eyed expression resting against the tactile fabric. [: the inner face of a scuffed closed door and a dim narrow wall, with a shoe rack partially visible at the lower frame edge, all falling into a soft background depth].",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              },
              {
                "character_id": "C03",
                "outlook_id": "O06"
              }
            ],
            "source_facts": [
              "Scene text action: 수리영이 엄마의 냄새를 맡으며 '아, 좋다'라고 말함.",
              "Shot description: 수리영 leaning into 민숙's side, hand clutching 민숙's clothing, eyes closed.",
              "Shot description: 민숙 remaining just beyond her in the same shallow plane, soft smile paused."
            ],
            "visual_inferences": [
              "fixed_element minsuk_attire_and_mark: Minsuk is wearing casual domestic clothes with a kitchen apron tied around her body.",
              "Suryeong is wearing a casual beige hoodie, fitting the 'light outfit' described for her part-time job outing.",
              "The background is restricted to the immediate frame edges (entry door and shoe rack) due to the close framing nature of the shot."
            ],
            "creative_decisions": [
              "Used a head-and-shoulder composition to appropriately frame the close-up profile without violating the face-filling-the-frame restriction.",
              "Placed C03O06 in the shallow background, partially visible, keeping the visual hierarchy focused on C01O02's action while maintaining spatial continuity.",
              "Omitted the specific background reference ID as requested and per the skipped_close_framing policy, describing only the frame-edge surfaces directly in an L00 block."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "eda65a10381c4a7b",
              "owned_hash": "336090adaa16cf9a",
              "camera_direction_hash": "ad505378fee2f4bf",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 5,
        "dependency_reason": "동일한 공간에서 이전 샷(Shot 3)에 연이어 일어나는 상황이므로 배경 톤과 복장의 연속성이 유지되어야 하나, 클로즈 프레이밍으로 인해 배경 묘사는 인접 피사체로 제한됨.",
        "_shot_index": 7,
        "visible_entities": [
          "C01",
          "C03"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 5,
            "shot_index": 7
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Inside the narrow entry, a slightly high angle CU profile frame holds at shoulder height about 0.8 meters from 수리영 as she leans into 민숙’s side. 수리영’s closed-eyed side profile fills the left-center, her hand clutching 민숙’s clothing near the lower center, while 민숙 remains just beyond her in the same shallow plane, softened but present.",
            "lighting_mood": "Soft room light and faint daylight flatten into warm gray-beige tones, making the intimate gesture feel tender but still shadowed by the earlier news anxiety.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01",
              "C03"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O01"
              },
              {
                "character_id": "C01",
                "outlook_id": "O02"
              },
              {
                "character_id": "C03",
                "outlook_id": "O06"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B01",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "minsuk_attire_and_mark",
                "element_type": "character_state",
                "character_name": "Minsuk",
                "description": "A 40-year-old East Asian woman wearing casual domestic clothes with a kitchen apron tied around her body. A distinct, small, circular reddish mark is permanently present on her wrist skin.",
                "applies_to_shots": [
                  3,
                  7
                ],
                "source_facts": [
                  "Scene text action: '앞치마를 두른 채' (wearing an apron)",
                  "Scene text action: '민숙 손목에 보이는 작은 원(이하 표식)' (small circular mark on Minsuk's wrist)",
                  "Shot 3: MCU push-in framing of Minsuk",
                  "Shot 7: CU profile framing showing Minsuk's upper body in the shallow plane"
                ],
                "visual_inferences": [
                  "The apron and casual clothes are consistently worn from the living room area to the entryway.",
                  "The circular mark on the wrist is an inherent, unmoving physical trait that belongs to her overall character model."
                ],
                "creative_decisions": [
                  "Merged the persistent attire and physical mark into a single 'full/upper body' character state for Shots 3 and 7.",
                  "Excluded Shot 9 from fixed elements because it is a solitary extreme close-up, adhering to the rule to prevent conflicting zoom descriptions."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 5,
                "shot_index": 3,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              },
              {
                "kind": "character_outlook",
                "id": "C03O06",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "e138489cb9a88016"
      },
      {
        "scene_index": 5,
        "heading": "S#5. 옥탑방 안 / 실내",
        "beat_title": "손목의 붉은 표식 클로즈업",
        "representative_moment": "수리영의 시선으로, 민숙의 손목에 있는 또렷하고 둥근 표식(P08)이 극단적으로 가깝게 보이는 정지 컷.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Subjective ECU Dutch angle macro, soft indirect daylight with faint ominous red-brown undertone, extreme shallow depth of field.",
            "t2i_prompt": "Photorealistic cinematic still. Focus on the same wrist of an East Asian woman in her 40s wearing a grey homewear sleeve, captured in an extreme close-up Dutch angle macro. The forearm extends diagonally across the middle third of the frame. P08, a distinct dark circular mark, is held sharply in focus on the slightly damp outer skin surface facing the lens. Soft indirect daylight washes the skin, with a faint ominous red-brown ambient undertone pooling on P08. The wrinkled fabric of the sleeve cuff partially shadows the wrist's edge, while [: the out-of-focus background blurs entirely into abstract dim beige-gray tones].",
            "outfit_assignments": [],
            "source_facts": [
              "S#5 action: '힐끗, 민숙 손목에 보이는 작은 원(이하 표식)을 본다. 오늘따라 유난히 또렷해 보이지만...'",
              "shot description: subjective POV ECU Dutch angle macro from 수리영's glance, circular wrist mark on the skin."
            ],
            "visual_inferences": [
              "fixed_element: isolated wrist view omitting full bodies or faces",
              "P08 mark highlighted by a faint ominous red-brown undertone reflecting on the skin",
              "background elements reduced to extreme out-of-focus soft blur due to macro lens"
            ],
            "creative_decisions": [
              "Used 'Focus on' trigger to comply with body_part_focus_rule, replacing character ID with a demographic descriptor.",
              "Placed P08 diagonally across the frame to enhance the Dutch angle feeling without specifying camera metrics.",
              "Rendered the background as an abstract [L01: ...] block to maintain close framing constraints while omitting prior background references."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "2ba6859d752d4b31",
              "owned_hash": "336090adaa16cf9a",
              "camera_direction_hash": "df94279b5c816c7b",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 5,
        "dependency_reason": "zoom_in_detail from previous shot in the same space, focusing tightly on the wrist.",
        "_shot_index": 9,
        "visible_entities": [
          "C03",
          "P08"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 5,
            "shot_index": 9
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A subjective POV ECU Dutch angle macro from 수리영’s glance sits less than 20 centimeters from 민숙’s wrist, the lens nearly parallel to the skin surface. Both faces are excluded; the wrist cuts diagonally across the middle third, and the clear circular mark is held sharp while the entry wall and fabric blur into abstract muted color behind it.",
            "lighting_mood": "The macro is lit by soft indirect daylight with a faint ominous red-brown undertone in the mark, turning a domestic detail into an occult warning.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C03",
              "P08"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C03",
                "outlook_id": "O06"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B01",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [
              {
                "scene_index": 5,
                "shot_index": 3,
                "ref_usage": "zoom_in_detail",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "eb68536f60271488"
      },
      {
        "scene_index": 6,
        "heading": "S#6. 해안가 도로 / D",
        "beat_title": "바닷바람을 느끼는 수리영",
        "representative_moment": "C01이 이어폰을 끼고 눈을 감은 채 미소 지으며 자전거로 도로를 달리는 얼굴 클로즈업.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Tracking close-up eye-level shot, humid pastel coastal daylight with motion-blurred background",
            "t2i_prompt": "Photorealistic cinematic still. A tracking close-up eye-level composition from the front-right. C01O02 in a beige hoodie, a young Asian woman, at a three-quarter angle from the right, her face occupying the center-right of the frame with a relaxed smile and eyes closed. Her dark hair is blown back by the wind in the direction opposite to her forward motion. She is wearing earphones with wires trailing down her neck. A sliver of her shoulder, the curved bicycle handlebar, and a slice of the front frame are visible at the lower edge, anchoring the mid-ride motion. Focus on her relaxed expression and closed eyes. Lit by soft, pale blue coastal daylight bringing out warm skin tones against a humid pastel ambient. The background is a soft, motion-blurred coastal backdrop of a hazy bright sea horizon and sunlit road texture.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "S#6 action: '수리영 자전거를 타고 해안가 도로를 달린다. 이어폰을 끼고 바다냄새를 실컷 마신다. 밝고 활기차 보인다.'",
              "shot description: 눈을 감고 바닷바람을 맞는 수리영의 미소 띤 얼굴 클로즈업.",
              "DP instruction: A front-right tracking CU eye-level shot moves backward at bicycle speed"
            ],
            "visual_inferences": [
              "hair blowing back and motion-blurred background used to convey forward motion",
              "handlebar visible at lower edge to anchor the bicycle context in a close frame"
            ],
            "creative_decisions": [
              "humid pastel color palette to reflect the bright, energetic mood and coastal setting",
              "face positioned center-right to create a sense of moving into the frame space"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "555fb5ff9ad42f8b",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "9ccd15073d6e3bb2",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "",
        "_shot_index": 2,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 6,
            "shot_index": 2
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A front-right tracking CU eye-level shot moves backward at bicycle speed, holding 수리영’s face close while the coast streaks softly behind her. Her closed eyes and relaxed smile occupy the center-right 70% of frame, with a sliver of handlebar and shoulder movement at the lower edge to preserve the sensation of motion.",
            "lighting_mood": "Bright coastal daylight is softened into a humid pastel palette, with pale blue sea light and warm skin tones giving her a brief sense of freedom.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": null,
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "f61d5418f68dc299"
      },
      {
        "scene_index": 7,
        "heading": "S#7. 유기견 보호원 / D",
        "beat_title": "사납게 짖는 맹견 로트와일러",
        "representative_moment": "C08이 철창을 붙잡고 이빨을 드러낸 채 사납게 카메라를 노려보고 있다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Handheld MS low angle through-object frame, cold fluorescent lighting with damp reflections, grim coastal-gray cast.",
            "t2i_prompt": "Photorealistic cinematic still. Handheld medium shot from a low angle, the wet concrete floor from the reference visible along the lower edge. Thick, vertical metal cage bars of the animal enclosure from the reference cut the foreground into hard strips. Focus on the bared teeth of the animal. C08O00, a large dog with dark brown short fur, a broad head, strong jaw, and dark eyes, facing the camera behind the foreground barrier. C08O00's raised paws are hooked on the damp metal bars at mid-frame, strong jaw open to reveal bared teeth just above the center, dark eyes fixed directly on the camera. The dim kennel aisle from the reference recedes in the background. Cold fluorescent lighting mixes with damp concrete reflections in the humid room, creating pale highlights on the wet metallic surfaces and giving the space a grim coastal-gray cast, highlighting the dark, violent accent of the dog's mouth.",
            "outfit_assignments": [
              {
                "character_id": "C08",
                "outlook_id": "O00"
              }
            ],
            "source_facts": [
              "Shot 1 action: 'A vicious Rottweiler inside a cage, standing on its hind legs, baring its teeth and gripping the cage bars.'",
              "Shot 1 environment: 'damp, humid room illuminated by cold fluorescent lighting', 'pale reflections on the wet metallic and concrete surfaces'."
            ],
            "visual_inferences": [
              "fixed_element damp_kennel_bars: used as a foreground barrier to create a through-object composition.",
              "Low camera angle places the dog's bared teeth just above center frame, emphasizing dominance and threat."
            ],
            "creative_decisions": [
              "Handheld framing slightly offset to maintain tension, with focus locked squarely on the aggressive display of teeth and grip on the metal bars.",
              "Coastal-gray cast and fluorescent reflections enhance the gritty, institutional atmosphere of the shelter."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "9e1ec62b5be7754e",
              "owned_hash": "456d19eb30dd6a5c",
              "camera_direction_hash": "3c07a1918b3a748c",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "cage",
                  "violating_phrase": "the animal enclosure from the reference",
                  "reason": "The cage is mentioned only as an existing referenced enclosure element, not as something newly created or added.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "kennel",
                  "violating_phrase": "The dim kennel aisle from the reference",
                  "reason": "The kennel is referenced as part of the existing background context from the reference, with no explicit redraw instruction.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "No direct prior scene specified.",
        "_shot_index": 1,
        "visible_entities": [
          "C08"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 7,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "From inside the kennel, a handheld MS low angle through-object frame sits about 40 centimeters above the floor and one meter from 로트와일러, looking outward through thick cage bars. The bars cut the foreground into hard vertical strips while the dog’s raised paws grip the metal at mid-frame and its bared teeth sit just above center, confronting the lens directly.",
            "lighting_mood": "Cold fluorescent light mixes with damp concrete reflections, giving the shelter a grim coastal-gray cast while the dog’s mouth reads as a dark, violent accent.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C08"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C08",
                "outlook_id": "O00"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L06B01",
            "reference_usage": "exact_background",
            "owned_objects": [
              "CCTV camera",
              "bathtub",
              "cabinet",
              "cage",
              "cart",
              "counter",
              "crate",
              "door",
              "dryer",
              "exam table",
              "floor drain",
              "grooming table",
              "hamper",
              "hose sprayer",
              "kennel",
              "light fixture",
              "partition",
              "rack",
              "refrigerator",
              "shelving",
              "sink",
              "trolley",
              "ventilation unit",
              "window"
            ],
            "camera_reference": {
              "camera_position": "near number 6 (main interior door) on the south side, facing diagonally north toward number 8 (west dog kennel bank), number 9 (east dog kennel bank), and number 17 (grooming table), with number 20 (bathing tub) visible deeper to the northeast",
              "camera_height": "eye-level standing ~1.6m",
              "lens_hint": "28mm wide angle",
              "framing_notes": "Include the busy kennel aisle and at least one wet-care activity zone; keep number 12 and number 13 readable in the mid-background for vaccination prep context."
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "when focusing on an owned object, anchor it explicitly: 'the door from the reference'",
              "camera_reference specifies the framing under which the background was painted — match camera_position, camera_height, and framing",
              "if deviating from the reference framing (push-in, angle shift, ECU), state the deviation explicitly so the compositor can account for the scale difference"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "damp_kennel_bars",
                "element_type": "environment_state",
                "character_name": "",
                "description": "Thick, vertical metal cage bars of an animal enclosure. The structure is set in a damp, humid room illuminated by cold fluorescent lighting, creating pale reflections on the wet metallic and concrete surfaces.",
                "applies_to_shots": [
                  1,
                  5
                ],
                "source_facts": [
                  "Shot 1 camera: 'thick cage bars', 'hard vertical strips'",
                  "Shot 1 lighting: 'Cold fluorescent light mixes with damp concrete reflections'",
                  "Shot 5 camera: 'blurred kennel bars remain as damp vertical streaks behind her'",
                  "Shot 5 lighting: 'muted by humidity', 'dim practical spill'"
                ],
                "visual_inferences": [
                  "The metal cages are uniform throughout the holding area, consistently reflecting the bleak, humid lighting of the shelter room."
                ],
                "creative_decisions": [
                  "Using the damp, vertical metal bars as a recurring environmental motif to ground the location's gritty, institutional, and enclosed atmosphere across different camera angles."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "background",
                "id": "L06B01",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "c769645ed67efdcd"
      },
      {
        "scene_index": 7,
        "heading": "S#7. 유기견 보호원 / D",
        "beat_title": "늦어지는 일정과 받지 않는 전화",
        "representative_moment": "C01 stares down at her phone with a hardened expression as the unanswered call casts a pale glow under her chin.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Eye-level close-up, shallow depth of field, cool humid daylight contrasted with pale green-blue screen underlighting.",
            "t2i_prompt": "Photorealistic cinematic still. Eye-level close-up composition. C01O02 in a beige hoodie, a Korean female in her early 20s with a youthful face and dark hair, at three-quarter angle from the right, head bent forward with eyes looking down toward the bottom edge. Her face and shoulders occupy the upper two-thirds of the frame, capturing her lowered gaze and hardened mouth. A pale green-blue glow from an unseen phone held just below the frame reflects upward, pooling on her chin, jawline, and the front of her hoodie against the dim ambient fill. Shallow depth of field blurs the immediate background into soft vertical streaks, showing thick metal cage bars catching pale reflections from cold fluorescent lighting in the damp, humid air. Focus on her tense facial features illuminated by the upward screen glow.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "Shot 5 description: 휴대폰 액정을 바라보는 수리영의 굳은 얼굴 클로즈업.",
              "S#7 action: 민숙에게 연락을 취해보지만, 받지 않는다.",
              "DP Notes: 'with her face occupying the upper two-thirds and the phone held just below frame casting light up under her chin'"
            ],
            "visual_inferences": [
              "fixed_element damp_kennel_bars: Thick, vertical metal cage bars... creating pale reflections on the wet metallic and concrete surfaces.",
              "The phone itself remains mostly out of frame to maintain focus on her reaction, acting as a practical light source from below."
            ],
            "creative_decisions": [
              "Close framing limits the background to blurred vertical cage bars, maintaining spatial continuity without requiring a direct background reference image.",
              "Using the green-blue glow of the unanswered call to carve her features from below, emphasizing her tension and the cold environment."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "5dfeefa86a7a48ea",
              "owned_hash": "456d19eb30dd6a5c",
              "camera_direction_hash": "dde9c0538a830e86",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "Relies on prior shot within the same scene for continuity of environment, but close framing omits direct background reference requirements.",
        "_shot_index": 5,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 7,
            "shot_index": 5
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "From the kennel aisle flow, the camera settles into a CU, eye-level shallow-DOF push-in about 50–60 cm from 수리영, with her face occupying the upper two-thirds and the phone held just below frame casting light up under her chin. This is neutral observer vision, not her phone POV; the frame compresses her lowered gaze and hardened mouth while blurred kennel bars remain as damp vertical streaks behind her.",
            "lighting_mood": "Cool kennel daylight is muted by humidity, with the phone’s pale green-blue glow carving her tense features against dim practical spill.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L06B01",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "damp_kennel_bars",
                "element_type": "environment_state",
                "character_name": "",
                "description": "Thick, vertical metal cage bars of an animal enclosure. The structure is set in a damp, humid room illuminated by cold fluorescent lighting, creating pale reflections on the wet metallic and concrete surfaces.",
                "applies_to_shots": [
                  1,
                  5
                ],
                "source_facts": [
                  "Shot 1 camera: 'thick cage bars', 'hard vertical strips'",
                  "Shot 1 lighting: 'Cold fluorescent light mixes with damp concrete reflections'",
                  "Shot 5 camera: 'blurred kennel bars remain as damp vertical streaks behind her'",
                  "Shot 5 lighting: 'muted by humidity', 'dim practical spill'"
                ],
                "visual_inferences": [
                  "The metal cages are uniform throughout the holding area, consistently reflecting the bleak, humid lighting of the shelter room."
                ],
                "creative_decisions": [
                  "Using the damp, vertical metal bars as a recurring environmental motif to ground the location's gritty, institutional, and enclosed atmosphere across different camera angles."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 7,
                "shot_index": 1,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "82b782e34fe35093"
      },
      {
        "scene_index": 7,
        "heading": "S#7. 유기견 보호원 / D",
        "beat_title": "밀린 강아지 목욕 업무 돌입",
        "representative_moment": "수리영(C01)이 목욕 테이블에 기대어 비누 거품 묻은 강아지의 등에 손을 얹고 있다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "MCU from a slight high angle with shallow DOF, flat fluorescent light softened by steam, subdued coastal blue tone",
            "t2i_prompt": "Photorealistic cinematic still. Cropped to the upper body from a slight high angle beside the wash table, C01O02 in a beige hoodie, an East Asian woman in her early 20s with a young face, black hair, and dark eyes, at three-quarter angle from the left, leaning over the wash table, eyes fixed on the dog below. Her soapy hand rests on the dog's wet back covered with white foam, the dog occupying the lower foreground in partial visibility. The wet, scratched stainless wash table has its long side facing the camera at a shallow diagonal, its rim cutting across the lower frame edge. Behind her shoulder, a fogged, water-spotted plastic splash guard blurs grooming tools and a gray wall, next to a streaked wall-mounted soap dispenser. The scene is illuminated by flat fluorescent wash-room light softened by rising steam and humidity, creating dull steel reflections and a subdued coastal blue ambient tone. Shallow depth of field focuses on her upper body and the soapy foreground interaction.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "S#7 action: '네', 하고 달려가 털깍고, 연고 바르고, 목욕시키고 등",
              "shot description: 강아지 몸에 비누 거품을 묻힌 손을 얹고 있는 수리영의 상체 컷"
            ],
            "visual_inferences": [
              "dog's back as partial foreground element consistent with close framing rule",
              "stainless wash table acts as frame-edge surface without invoking skipped background references"
            ],
            "creative_decisions": [
              "shallow depth of field through rising steam to emphasize the intimate but workmanlike motion",
              "subdued coastal blue tone and flat fluorescent light carried from S7_Shot5 atmosphere reference"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "a05b824c6e0363e4",
              "owned_hash": "456d19eb30dd6a5c",
              "camera_direction_hash": "4a0fe92d51de4890",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 7,
        "dependency_reason": "Shot 5의 조명/분위기 톤 연속성 (atmosphere_reference)",
        "_shot_index": 7,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 7,
            "shot_index": 7
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Beside the wash table, the tracking flow becomes a MCU with a slight high angle, cropped to 수리영’s upper body while her soapy hand and the dog’s wet back sit in the lower foreground. The camera stays chest-height and about one meter away, using shallow DOF through rising steam so her careful motion feels intimate but still workmanlike.",
            "lighting_mood": "Flat fluorescent wash-room light is softened by steam and humidity, with dull steel reflections and subdued coastal blues.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L06B01",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [
              {
                "scene_index": 7,
                "shot_index": 5,
                "ref_usage": "atmosphere_reference",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "5287cc750c9b473f"
      },
      {
        "scene_index": 8,
        "heading": "S#8. 번화가·대형 마트 / 해질 무렵",
        "beat_title": "엇갈린 동선에 혼란스러운 수리영",
        "representative_moment": "C01이 빈 박스를 든 여직원 어깨 너머로 눈을 동그랗게 뜨고 바라보는 순간",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Eye-level over-the-shoulder MCU, greenish fluorescent and damp neon lighting",
            "t2i_prompt": "Photorealistic cinematic still. Eye-level over-the-shoulder composition. In the right foreground, the shoulder and back of a middle-aged Korean woman in a staff vest appear in partial visibility with her back to the camera, an empty cardboard box tucked to her ribs cutting into the lower-right frame at chest height. Across the shared mart aisle space, C01O02 in a beige hoodie, an early 20s Korean woman with black hair and dark eyes, stands at a three-quarter angle from the right, her weight on her back foot, her eyes fixed on the employee. Focus on C01O02's widened eyes in a head-and-shoulder composition, occupying the center-left. Greenish fluorescent ceiling light washes C01O02's hair, mixing with a continuous lighting condition of early evening dusk and colorful damp neon spilling from the entrance. The neon glow reflects faintly on C01O02's jaw and the cardboard box edge. In the soft background behind C01O02, partly stocked mart shelving runs diagonally along the aisle edge.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "S#8 action: 수리영이 마트 직원에게 민숙의 행방을 묻자 조퇴했다는 대답을 듣는다.",
              "shot description: 빈 박스를 든 40대 한국인 여직원 앞에 서서 눈을 동그랗게 뜬 수리영의 표정."
            ],
            "visual_inferences": [
              "fixed_element damp_dusk_neon_lighting: continuous lighting condition of early evening dusk mixed with colorful neon lights reflecting off damp surfaces",
              "The empty cardboard box acts as a physical barrier between the two characters in the foreground, anchored by the shared aisle space."
            ],
            "creative_decisions": [
              "Over-the-shoulder framing to emphasize C01O02's reaction while keeping the unregistered employee anchored in the right foreground without showing her face.",
              "Applied a head-and-shoulder composition for C01O02 to prevent 'floating face' artifacts in close-up generation.",
              "Omitted L08 background ID per modification request, directly describing the frame-edge mart shelving with soft focus to comply with close framing background skip rules."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "3badc2905ac87b09",
              "owned_hash": "7c29ab00671f32cf",
              "camera_direction_hash": "b06a4d780e6784e6",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "N/A",
        "_shot_index": 2,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 8,
            "shot_index": 2
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Inside the mart front aisle, the dolly-in settles into a near-static MCU, eye-level OTS from the 40대 한국인 여직원’s box-side shoulder toward 수리영’s widened eyes. 수리영 holds the center-left of frame while the empty cardboard box cuts into the right foreground at chest height, keeping the employee’s answer physically between them.",
            "lighting_mood": "Greenish fluorescent mart light mixes with damp dusk neon leaking through the entrance, muting skin tones and adding unease.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L08B01",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "damp_dusk_neon_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A continuous lighting condition of early evening dusk mixed with colorful neon lights. The neon light reflects off damp surfaces, casting a moody, slightly muted glow that visually bridges the interior entrance and the exterior street environment.",
                "applies_to_shots": [
                  2,
                  5
                ],
                "source_facts": [
                  "Shot 2 lighting: 'mixes with damp dusk neon leaking through the entrance'",
                  "Shot 5 lighting: 'Dusk has fallen into blue-gray street darkness, punctured by damp neon'"
                ],
                "visual_inferences": [
                  "The ambient damp neon lighting spills from the wet street into the mart's entrance, maintaining consistent environmental color tones on the character's face across both locations."
                ],
                "creative_decisions": [
                  "Using wet surface reflections and neon color spills to visually connect the interior and exterior shots, maintaining a consistent somber and uneasy mood."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "1eff32da3900b9c4"
      },
      {
        "scene_index": 8,
        "heading": "S#8. 번화가·대형 마트 / 해질 무렵",
        "beat_title": "사라진 엄마와 불길한 예감",
        "representative_moment": "어두운 마트 밖에서 C01이 자전거 핸들을 꽉 쥔 채 굳은 표정으로 생각에 잠겨 있다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Medium close-up at low handlebar height, moody damp neon lighting, shallow depth of field",
            "t2i_prompt": "Photorealistic cinematic still. Medium close-up, low camera positioned at handlebar height. C01O02 in a beige hoodie, an early 20s Korean woman with a young face, black hair, and dark eyes, at three-quarter angle from the left with her eyes looking distant. She is gripping the wet curved front bicycle handlebar that crosses the lower third of the frame, her clenched hands sitting sharp in the lower center, with the bicycle's front tube extending below her grip. Her hardened face and chest remain clearly visible in the upper frame above the hands. Lit by a continuous early evening dusk mixed with colorful damp neon lights, casting a moody glow with a faint ominous red spill washing her cheek and the wet metal. The close background behind her consists of soft, out-of-focus darkened glass doors streaked with moisture and a neon-glowing mart sign pointing toward the camera softened by drizzle, while the rain-slick pavement reflects broken streaks of red and green neon in the blue-gray street darkness. The composition centers on her rigid posture and hardened expression against the urban blur.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "S#8 action: '수리영 마트 밖을 나오면서 연락을 취해보지만 받지않는다. 일단 집으로-'",
              "shot description: Outside the mart after dusk, the frame drops into a MCU, low angle at handlebar height, using the bicycle handlebar as a through-object foreground line."
            ],
            "visual_inferences": [
              "fixed_element: damp_dusk_neon_lighting applied to the wet street exterior",
              "C01 is positioned at her bicycle, pausing in a tense state as realization of her mother's unexplained absence sets in",
              "The background is rendered soft and out-of-focus to emphasize the character's internal anxiety and physical tension"
            ],
            "creative_decisions": [
              "Positioned the camera low to anchor the clenched hands on the handlebar in the foreground, projecting a strong line of tension",
              "Used a faint ominous red spill washing her cheek to subtly foreshadow the upcoming darker events",
              "Skipped full background reference generation and tightly framed the immediate surroundings with moisture-streaked glass and wet pavement reflections"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "3605fed04c4bdad6",
              "owned_hash": "82729a7331412f24",
              "camera_direction_hash": "3852248132c20b7d",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 8,
        "dependency_reason": "previous_shot_refs indicates atmosphere_reference dependency on scene 8 shot 2 for consistent damp dusk neon lighting",
        "_shot_index": 5,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 8,
            "shot_index": 5
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Outside the mart after dusk, the frame drops into a MCU, low angle at handlebar height, using the bicycle handlebar as a through-object foreground line across the lower third. 수리영’s clenched hands sit sharp in the lower center while her hardened face remains visible above them, slightly compressed against wet neon reflections and passing street blur.",
            "lighting_mood": "Dusk has fallen into blue-gray street darkness, punctured by damp neon and a faint ominous red spill on the wet pavement.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L08B02",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "damp_dusk_neon_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A continuous lighting condition of early evening dusk mixed with colorful neon lights. The neon light reflects off damp surfaces, casting a moody, slightly muted glow that visually bridges the interior entrance and the exterior street environment.",
                "applies_to_shots": [
                  2,
                  5
                ],
                "source_facts": [
                  "Shot 2 lighting: 'mixes with damp dusk neon leaking through the entrance'",
                  "Shot 5 lighting: 'Dusk has fallen into blue-gray street darkness, punctured by damp neon'"
                ],
                "visual_inferences": [
                  "The ambient damp neon lighting spills from the wet street into the mart's entrance, maintaining consistent environmental color tones on the character's face across both locations."
                ],
                "creative_decisions": [
                  "Using wet surface reflections and neon color spills to visually connect the interior and exterior shots, maintaining a consistent somber and uneasy mood."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 8,
                "shot_index": 2,
                "ref_usage": "atmosphere_reference",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "a12ce821dece4b74"
      },
      {
        "scene_index": 9,
        "heading": "S#9. 해안가 / N",
        "beat_title": "불안한 밤길 귀가",
        "representative_moment": "어둠이 깔린 해안도로에서 입술을 꽉 깨문 채 자전거 페달을 힘껏 밟는 수리영(C01)의 모습.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Parallel tracking medium wide shot from waist height, slight high angle, cold coastal night tones with sodium streetlight glints",
            "t2i_prompt": "Photorealistic cinematic still. C01O02 in a beige hoodie, a Korean woman in her early 20s with dark eyes and black hair, seated on the saddle of the bicycle in left profile, mid-pedal stroke with one foot driving down, hair flowing back in the direction opposite to motion. Her mouth is firmly clenched, eyes fixed on the distant road ahead. Her hands tightly grip the bicycle front handlebar, with the bicycle frame and front wheel clearly visible below in the lower right. Parallel to the bicycle on its left side, a tracking waist-height medium wide shot with a slight high angle, her upper body occupying the upper-left frame. [L05: dark coastal road, wet from rain, streaking laterally behind her, the humid foggy sea edge barely visible]. Cold coastal night tones dominate, sodium streetlight glints and a faint red roadside warning sign reflection catching on the wet metal and her cheek.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "S#9 action: '해안도로를 자전거를 타고 바삐 집으로 달려가는 수리영.'",
              "shot description: mid-pedal on the dark coastal road, clenched mouth, waist height MWS tracking profile",
              "forward zoom targets: woman riding bicycle in left profile, bicycle front handlebar, wet dark coastal road"
            ],
            "visual_inferences": [
              "motion direction preservation: hair flowing back in the direction opposite to motion",
              "prop visibility rule: bicycle frame and front wheel clearly visible below the rider"
            ],
            "creative_decisions": [
              "slight high angle waist-height composition to capture both her clenched mouth and driving foot",
              "red reflections catching on the wet metal and cheek to visually heighten the suspense without cutting into the body"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "b0b68a838aecb6f3",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "c63788b3db5875d9",
              "validator": "scene_detail_owned_objects.v1",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 6,
        "dependency_reason": "continuity_elements_used.previous_shot_refs specifies Scene 6 Shot 2 exact_background",
        "_shot_index": 1,
        "visible_entities": [
          "C01",
          "L05"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 9,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Parallel to the bicycle on its left side, the camera tracks in a MWS with a slight high angle from waist height, widened enough to catch 수리영’s clenched mouth near the upper left and her foot driving the pedal down in the lower right. The road streaks laterally behind her in an unsteady tracking profile, preserving the flow’s speed while making one mid-pedal strike feel like a held breath.",
            "lighting_mood": "Cold coastal night tones flatten the road, with sodium streetlight glints and faint red roadside reflections flashing across wet metal.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01",
              "L05"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "not_applicable",
            "bg_id": null,
            "reference_usage": "none",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": false,
            "constraints": [
              "no chain_bg is bound to this shot — describe the location freely without referencing a background image",
              "do not use 'from the reference' / 'the existing room' / 'preserving the room' style phrasing — there is no reference image to refer to"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [
              {
                "scene_index": 6,
                "shot_index": 2,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [
              {
                "scene_index": 22,
                "shot_index": 4,
                "description": "뺨을 타고 내리던 빗방울이 피부에 맺힌 채, 핏발 선 눈으로 정면을 고정하여 응시하는 수리영의 옆얼굴 클로즈업.",
                "keep_elements": [
                  {
                    "label": "woman riding bicycle in left profile",
                    "kind": "immobilized_character"
                  },
                  {
                    "label": "bicycle front handlebar and grip",
                    "kind": "static_prop"
                  },
                  {
                    "label": "wet dark coastal road behind her",
                    "kind": "environment"
                  }
                ]
              }
            ],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "4d88776ec19d28ed"
      },
      {
        "scene_index": 10,
        "heading": "S#10. 옥탑방 / N",
        "beat_title": "문 앞의 긴 그림자와 전단지",
        "representative_moment": "프레임 밖 낯선 남자의 긴 그림자가 옥상 바닥을 가로질러 문틈에 전단지가 꽂힌 철문까지 뻗어 있는 순간.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "medium wide shot, low angle at ankle height, weak practical lighting, cool blue-gray tone with rust-red accents",
            "t2i_prompt": "Photorealistic cinematic still. Deviating from the reference's eye-level framing to a medium wide shot at a low ankle-height perspective, the floor occupying the lower frame. Focus on an elongated, dark human shadow stretching diagonally across the damp, wind-scuffed concrete floor from the reference. The figure casting the shadow remains completely off-frame. A weak rooftop practical light shapes this dominant diagonal silhouette leading the eye toward the threshold. In the upper frame, the weathered iron door from the reference stands closed, its bottom edge creating a black vertical seam. At the narrow gap of the door, a creased advertising flyer is wedged edge-first, its printed face partly visible to the camera showing blurred ad graphics. The blue-gray concrete is marked with rust-red stains near the threshold, evoking a dim, humid coastal night atmosphere.",
            "outfit_assignments": [],
            "source_facts": [
              "S10 action: '남자의 긴 그림자. (얼굴은 보이지 않는다)'",
              "S10 action: '문틈에 낀 광고 전단지.'",
              "shot description: '철문 현관 앞 바닥에 길게 드리워진 낯선 남자의 그림자.'"
            ],
            "visual_inferences": [
              "The figure must be entirely off-frame to maximize the ominous presence of the shadow alone.",
              "A low angle composition naturally places the concrete surface in the foreground and the door in the upper frame."
            ],
            "creative_decisions": [
              "Explicitly stated the deviation from the background reference's eye-level framing to an ankle-height perspective to guide the compositor on scale differences.",
              "Anchored the 'concrete floor' and 'iron door' with 'from the reference' to satisfy the background_ref_attached constraint and prevent double-drawing.",
              "Introduced rust-red stains on the blue-gray concrete to subtly support the occult thriller tone."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "fb1dfe784cd019c9",
              "owned_hash": "434f35347ebb15a4",
              "camera_direction_hash": "cb4ceaa85a4e9753",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "shadow",
                  "violating_phrase": "elongated, dark human shadow",
                  "reason": "'shadow' is explicitly mentioned, but it is only used as a compositional element with the figure off-frame and no redraw verb is applied to the owned object.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "door",
                  "violating_phrase": "the weathered iron door from the reference stands closed",
                  "reason": "The phrase explicitly anchors the door to the reference, so it indicates reuse of an existing background object rather than a new redraw.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 4,
        "dependency_reason": "S4_Shot1의 옥탑방 배경(L03B04)을 그대로 공유하되, 바닥 앵글로 프레이밍 편차를 두어 그림자와 문턱의 디테일을 강조함",
        "_shot_index": 3,
        "visible_entities": [
          "L03"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 10,
            "shot_index": 3
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "On the rooftop floor directly before the iron door, the camera holds a MWS, low angle at ankle height, tracking to a near-stop as the 낯선 남자’s elongated shadow stretches from off-frame across the concrete. His body and face remain outside the image; the shadow forms the dominant diagonal shape leading the eye to the threshold and the narrow door crack.",
            "lighting_mood": "A weak rooftop practical throws a long black shadow across blue-gray concrete, with rust-red stains near the threshold adding occult unease.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "L03"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L03B04",
            "reference_usage": "exact_background",
            "owned_objects": [
              "clothesline",
              "door",
              "drainage grate",
              "light fixture",
              "parapet",
              "platform",
              "post",
              "rooftop deck",
              "shadow",
              "stairwell headhouse",
              "utility pipe",
              "water tank"
            ],
            "camera_reference": {
              "camera_position": "near number 2 (iron door), facing diagonally across number 1 (rooftop deck) toward number 5 (water tank) and number 6 (clothesline posts)",
              "camera_height": "eye-level standing ~1.6m",
              "lens_hint": "35mm wide angle",
              "framing_notes": "include number 2 prominently as the intrusion point, keep number 5 and number 6 readable in the background, and place number 9 along the floor path for the long intruder silhouette trace"
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "when focusing on an owned object, anchor it explicitly: 'the door from the reference'",
              "camera_reference specifies the framing under which the background was painted — match camera_position, camera_height, and framing",
              "if deviating from the reference framing (push-in, angle shift, ECU), state the deviation explicitly so the compositor can account for the scale difference"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [
              {
                "scene_index": 4,
                "shot_index": 1,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "background",
                "id": "L03B04",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "8e24b347b906e449"
      },
      {
        "scene_index": 10,
        "heading": "S#10. 옥탑방 / N",
        "beat_title": "어깨를 물어뜯기는 민숙",
        "representative_moment": "어깨를 붙잡힌 채 입을 크게 벌리고 비명을 지르는 민숙(C03)의 일그러진 얼굴 클로즈업.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Eye-level handheld extreme close-up, Dutch angle, dim practical amber and cold blue spill",
            "t2i_prompt": "Photorealistic cinematic still. Eye-level handheld extreme close-up with a subtle Dutch angle. C03O06 in a grey homewear top and apron, an early 40s Korean woman with a middle-aged face, black hair, and dark eyes, at three-quarter angle from the left. Tight framing on her face, her neck and shoulder visible at the edge, the shoulder pulled out of alignment by the firm grip of a shadowy hand. She leans backward in recoil, eyes fixed on the void, her mouth wide open, cheek muscles visibly strained. Focus on her open mouth and strained cheek muscles. Dim practical amber light washes her skin, fighting a cold blue spill from the window to create a sickly, panicked tone. [: blurred frame-edge surfaces collapsing into smeared darkness — a half-shadowed interior doorframe cutting diagonally, a weak soft glow from a small room lamp, out-of-focus yellowed stained wallpaper, and a dark window glass catching a warped smear of lamp light and a vague moving darkness.]",
            "outfit_assignments": [
              {
                "character_id": "C03",
                "outlook_id": "O06"
              }
            ],
            "source_facts": [
              "S#10 action: '창문에 비친 그림자. 민숙을 덮쳐 어깨를 물어뜯는 듯한 모습 아악-'",
              "Shot description: 어깨를 붙잡힌 채 입을 크게 벌린 민숙의 일그러진 얼굴 클로즈업."
            ],
            "visual_inferences": [
              "Handheld camera shake and Dutch angle imply the sudden physical impact and panic",
              "The attacker's unseen grip is visualized as a shadowy hand firmly gripping her shoulder at the frame edge"
            ],
            "creative_decisions": [
              "Applied tight framing with the neck and shoulder visible to avoid the 'floating face' artifact in T2I models",
              "Used 'Dim practical amber light washes her skin' to ensure lighting interacts safely with the surface without 'cutting' the body",
              "Replaced the background reference with an [L03] frame-edge description matching the 'skipped_close_framing' constraints"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "544d73eac996b7a9",
              "owned_hash": "434f35347ebb15a4",
              "camera_direction_hash": "80edeb07725f8caa",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 10,
        "dependency_reason": "S10_Shot3에서 확립된 옥탑방 내부 공간 (close framing 룰에 따라 직접적인 bg 합성 대신 프레임 엣지 묘사로 축소하여 일관성 유지)",
        "_shot_index": 8,
        "visible_entities": [
          "C03"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 10,
            "shot_index": 8
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Inside the room, the camera snaps into an ECU, eye-level handheld close observation less than half a meter from 민숙’s face, tightening until her open mouth and strained cheek muscles occupy most of the frame. The angle is subtly Dutch, with the unseen grip jerking her shoulder out of alignment while the background collapses into smeared darkness.",
            "lighting_mood": "Dim practical amber fights cold blue window spill, while the handheld Dutch instability makes the skin tones sickly and panicked.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C03"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C03",
                "outlook_id": "O06"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L03B04",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [
              {
                "scene_index": 10,
                "shot_index": 3,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C03O06",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "e4c2651d05178b14"
      },
      {
        "scene_index": 11,
        "heading": "S#12. 옥탑방 안 / N",
        "beat_title": "핏자국의 발견",
        "representative_moment": "옥탑방 바닥에 흥건하게 고인 피와 커튼 쪽으로 이어진 핏자국 클로즈업.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Low Dutch angle close-up, lens just above the ground, strong cool blue-gray tone with highly saturated red contrast",
            "t2i_prompt": "Photorealistic cinematic still. Close-up from a low Dutch angle, lens just above the ground near the doorway, focus on the thick, wet pool of deep red blood. [L04: Saturated red liquid heavily spread across dark, dirty, damp-glossed wooden floorboards]. Smeared, uneven partial bloody footprints recede away from the camera along the long board seams toward the distant curtain side. In the soft background, a blurred long mundane fabric window curtain hangs vertically, its bottom edge touching the floorboards, framing the upper left. Submerged in oppressive, humid shadows with a strong cool, blue-gray tone, a weak muted spill of light catches the glossy surface of the fresh blood, creating a striking contrast against the grim, shadowy textures of the floor.",
            "outfit_assignments": [],
            "source_facts": [
              "Shot 11 action: '방바닥에 흥건하게 고인 붉은 핏자국 클로즈업.'",
              "Scene text: '바닥에 흥건한 혈흔. 피묻은 발자국은 커튼으로 이어져 있다.'",
              "shot description: floor-tracking flow nearly stops in a CU near the doorway, footprint trail leading deeper into the bedroom"
            ],
            "visual_inferences": [
              "fixed_element pooled_blood_on_floor: wet, thick, deep red blood forming a striking contrast on dark wooden boards",
              "fixed_element dim_blue_gray_lighting: weak light casting humid gray shadow over the floor",
              "fixed_element hanging_window_curtain: blurred curtain edge serving as a distant scale reference"
            ],
            "creative_decisions": [
              "Using a floor-level Dutch angle to emphasize the unsettling and gruesome discovery of the blood trail",
              "Keeping the curtain completely out of focus in the soft background to maintain the strict close-up framing on the floor",
              "Enhancing the glossy reflection on the blood to contrast with the damp, dirty wood textures"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "01aa027ed01c1c31",
              "owned_hash": "930b9286d35f0313",
              "camera_direction_hash": "caceee152e91bc8a",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 5,
        "dependency_reason": "atmosphere_reference for the dim, oppressive blue-gray interior lighting tone",
        "_shot_index": 11,
        "visible_entities": [
          "L04"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 11,
            "shot_index": 11
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Inside 수리영’s bedroom, the floor-tracking flow nearly stops in a CU, Dutch angle, with the lens only about 20 cm above the ground near the doorway. The pooled red blood occupies the lower center and stains the eye path deeper into the room, while floorboards and a blurred curtain line remain visible as scale references so the stain feels real rather than abstract.",
            "lighting_mood": "Weak TV spill and a dim room lamp leave the floor in humid gray shadow, broken by the unnaturally saturated red of the blood.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "L04"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B06",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "dim_blue_gray_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The interior space is submerged in oppressive, humid shadows characterized by a strong cool, blue-gray tone. A weak, muted spill of light from an off-camera source barely illuminates the space, maintaining a grim, dark, and gloomy atmosphere.",
                "applies_to_shots": [
                  11,
                  14,
                  20
                ],
                "source_facts": [
                  "Shot 11 lighting: 'dim room lamp leave the floor in humid gray shadow'",
                  "Shot 14 lighting: 'Muted lamp spill... carve silhouette into a gray-blue silhouette'",
                  "Shot 20 lighting: 'The room stays dim and blue-gray'"
                ],
                "visual_inferences": [
                  "The ambient light is extremely low, relying on a single weak artificial light source that casts cool, unsettling shadows."
                ],
                "creative_decisions": [
                  "Unifying the visual tone across wide and close-up shots by locking in the specific cool blue-gray color grading to enhance the horror atmosphere."
                ],
                "confidence": "high"
              },
              {
                "element_id": "pooled_blood_on_floor",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A thick, wet pool of deep red blood is heavily spread across the dark wooden floorboards. The saturated red liquid creates a striking, gruesome contrast against the muted, shadowy textures of the room's floor.",
                "applies_to_shots": [
                  11,
                  14
                ],
                "source_facts": [
                  "Scene text: '바닥에 흥건한 혈흔.'",
                  "Shot 11 action: '방바닥에 흥건하게 고인 붉은 핏자국'",
                  "Shot 11 camera: 'The pooled red blood occupies the lower center'",
                  "Shot 14 lighting: 'sick red stains pulling the eye downward'"
                ],
                "visual_inferences": [
                  "The blood is fresh enough to appear wet and highly saturated, starkly contrasting with the wooden floor."
                ],
                "creative_decisions": [
                  "Using the blood pool as a persistent environmental anchor between the close-up floor tracking and the wide reveal of the body."
                ],
                "confidence": "high"
              },
              {
                "element_id": "hanging_window_curtain",
                "element_type": "persistent_prop",
                "character_name": "",
                "description": "A long, mundane fabric window curtain hangs vertically, its bottom edge touching the wooden floorboards. It is partially pulled aside, serving as a soft, shadowy framing device at the edge of the visual field.",
                "applies_to_shots": [
                  11,
                  14
                ],
                "source_facts": [
                  "Scene text: '피묻은 발자국은 커튼으로 이어져 있다. 호흡을 가다듬고 커튼을 젖히자...'",
                  "Shot 11 camera: 'blurred curtain line remain visible as scale references'",
                  "Shot 14 camera: 'behind the pulled-aside fabric. The curtain edge frames the left foreground'"
                ],
                "visual_inferences": [
                  "The curtain is likely made of a heavy, neutral-colored fabric that absorbs the dim room light."
                ],
                "creative_decisions": [
                  "Maintaining the curtain as a constant foreground or background element to emphasize the cramped, hidden nature of the space."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 5,
                "shot_index": 9,
                "ref_usage": "atmosphere_reference",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "d791fbece6aa221b"
      },
      {
        "scene_index": 11,
        "heading": "S#12. 옥탑방 안 / N",
        "beat_title": "엄마의 참혹한 죽음 목격",
        "representative_moment": "커튼 뒤에서 피투성이로 죽어 있는 민숙을 발견한 순간",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Low angle medium framing at 1m height, dim blue-gray lighting with muted lamp spill",
            "t2i_prompt": "Photorealistic cinematic still. Deviating from the exact background reference camera height to a lower observer angle at about one meter high, medium framing. The folded front edge of the curtain from the reference faces the camera in the left foreground, opening toward the center. C03O06 in grey homewear and an apron, a Korean middle-aged woman with black hair and dark eyes, looking away, slumped seated on the floor from the reference. Her head is dropped forward and limbs are heavy, eyes dead and closed, body volume preserved against the dim ambient fill. Focus on her slumped figure pressed by the cramped room depth. The rumpled bed from the reference faces the camera from the right, cutting into the frame. Her limp hand and exposed wrist rest motionless on the floor beside a crumpled, old yellowed photo from the reference. An uneven red circular mark is visible on the dim damp-stained wall from the reference behind the bed. A thick wet bloodstain from the reference is spread across the dark wooden floor, catching the weak light. The dark glass of the closed window from the reference forms the dim back plane. The space is submerged in oppressive dim blue-gray lighting; muted lamp spill washes across her shoulders and the floor, shaping her into a gray-blue silhouette against the shadows.",
            "outfit_assignments": [
              {
                "character_id": "C03",
                "outlook_id": "O06"
              }
            ],
            "source_facts": [
              "S11 action: '앉은 채로 고개를 떨군 채 죽어있는 엄마- 민숙'",
              "S11 action: '엄마 손안에 구겨진 사진 한 장이 보인다'",
              "DP framing: 'WS, low angle from about one meter high, held just far enough back for 민숙’s entire slumped body to read behind the pulled-aside fabric'"
            ],
            "visual_inferences": [
              "fixed_element dim_blue_gray_lighting",
              "fixed_element pooled_blood_on_floor",
              "fixed_element hanging_window_curtain",
              "Visible exposed wrist and nearby floor photo to satisfy forward zoom targets"
            ],
            "creative_decisions": [
              "Deviating from the exact background reference height (~1.55m) to ~1m high to capture the slumped body and floor clues dynamically",
              "Preserving body volume in silhouette to avoid a flat cutout effect against the dim window background"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "a398067d7ff92a2c",
              "owned_hash": "930b9286d35f0313",
              "camera_direction_hash": "ed2ca7ae2d27e9bc",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "curtain",
                  "violating_phrase": "The folded front edge of the curtain from the reference",
                  "reason": "The curtain is explicitly referenced as already present from the reference, not newly created or modified.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "bed",
                  "violating_phrase": "The rumpled bed from the reference",
                  "reason": "The bed is described as an existing referenced object, which anchors the composition rather than redrawing it.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "photo",
                  "violating_phrase": "a crumpled, old yellowed photo from the reference",
                  "reason": "The photo is explicitly mentioned as part of the reference background, so this is an anchor reference.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "wall",
                  "violating_phrase": "An uneven red circular mark is visible on the dim damp-stained wall from the reference",
                  "reason": "The wall is treated as an existing background element from the reference, not something to add or redraw.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "bloodstain",
                  "violating_phrase": "A thick wet bloodstain from the reference",
                  "reason": "The bloodstain is explicitly referenced as already present, so it is an anchor reference.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "window",
                  "violating_phrase": "The dark glass of the closed window from the reference",
                  "reason": "The window is identified as a pre-existing referenced object, not a newly generated one.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 11,
        "dependency_reason": "exact_background (same room, different moment/angle)",
        "_shot_index": 14,
        "visible_entities": [
          "C03",
          "L04"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 11,
            "shot_index": 14
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Near the window curtain, the completed tilt resolves into a WS, low angle from about one meter high, held just far enough back for 민숙’s entire slumped body to read behind the pulled-aside fabric. The curtain edge frames the left foreground like a torn veil, while 민숙 sits center-right with her head dropped and limbs heavy, the room’s cramped depth pressing in around her.",
            "lighting_mood": "Muted lamp spill and cold window darkness carve 민숙 into a gray-blue silhouette with sick red stains pulling the eye downward.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C03",
              "L04"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C03",
                "outlook_id": "O06"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L04B06",
            "reference_usage": "exact_background",
            "owned_objects": [
              "backpack",
              "bed",
              "bloodstain",
              "curtain",
              "doll",
              "door",
              "floor",
              "photo",
              "wall",
              "wardrobe",
              "window"
            ],
            "camera_reference": {
              "camera_position": "just inside number 20 (main bedroom door), facing toward number 23 (curtain partition), number 24 (concealed body position), number 25 (blood stain symbol), and number 26 (red circular tattoo clue)",
              "camera_height": "eye-level standing ~1.55m",
              "lens_hint": "35mm wide angle",
              "framing_notes": "Keep the curtain as the main blocker; show the floor clue positions without exposing graphic detail."
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "when focusing on an owned object, anchor it explicitly: 'the door from the reference'",
              "camera_reference specifies the framing under which the background was painted — match camera_position, camera_height, and framing",
              "if deviating from the reference framing (push-in, angle shift, ECU), state the deviation explicitly so the compositor can account for the scale difference"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "dim_blue_gray_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The interior space is submerged in oppressive, humid shadows characterized by a strong cool, blue-gray tone. A weak, muted spill of light from an off-camera source barely illuminates the space, maintaining a grim, dark, and gloomy atmosphere.",
                "applies_to_shots": [
                  11,
                  14,
                  20
                ],
                "source_facts": [
                  "Shot 11 lighting: 'dim room lamp leave the floor in humid gray shadow'",
                  "Shot 14 lighting: 'Muted lamp spill... carve silhouette into a gray-blue silhouette'",
                  "Shot 20 lighting: 'The room stays dim and blue-gray'"
                ],
                "visual_inferences": [
                  "The ambient light is extremely low, relying on a single weak artificial light source that casts cool, unsettling shadows."
                ],
                "creative_decisions": [
                  "Unifying the visual tone across wide and close-up shots by locking in the specific cool blue-gray color grading to enhance the horror atmosphere."
                ],
                "confidence": "high"
              },
              {
                "element_id": "pooled_blood_on_floor",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A thick, wet pool of deep red blood is heavily spread across the dark wooden floorboards. The saturated red liquid creates a striking, gruesome contrast against the muted, shadowy textures of the room's floor.",
                "applies_to_shots": [
                  11,
                  14
                ],
                "source_facts": [
                  "Scene text: '바닥에 흥건한 혈흔.'",
                  "Shot 11 action: '방바닥에 흥건하게 고인 붉은 핏자국'",
                  "Shot 11 camera: 'The pooled red blood occupies the lower center'",
                  "Shot 14 lighting: 'sick red stains pulling the eye downward'"
                ],
                "visual_inferences": [
                  "The blood is fresh enough to appear wet and highly saturated, starkly contrasting with the wooden floor."
                ],
                "creative_decisions": [
                  "Using the blood pool as a persistent environmental anchor between the close-up floor tracking and the wide reveal of the body."
                ],
                "confidence": "high"
              },
              {
                "element_id": "hanging_window_curtain",
                "element_type": "persistent_prop",
                "character_name": "",
                "description": "A long, mundane fabric window curtain hangs vertically, its bottom edge touching the wooden floorboards. It is partially pulled aside, serving as a soft, shadowy framing device at the edge of the visual field.",
                "applies_to_shots": [
                  11,
                  14
                ],
                "source_facts": [
                  "Scene text: '피묻은 발자국은 커튼으로 이어져 있다. 호흡을 가다듬고 커튼을 젖히자...'",
                  "Shot 11 camera: 'blurred curtain line remain visible as scale references'",
                  "Shot 14 camera: 'behind the pulled-aside fabric. The curtain edge frames the left foreground'"
                ],
                "visual_inferences": [
                  "The curtain is likely made of a heavy, neutral-colored fabric that absorbs the dim room light."
                ],
                "creative_decisions": [
                  "Maintaining the curtain as a constant foreground or background element to emphasize the cramped, hidden nature of the space."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 11,
                "shot_index": 11,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [
              {
                "scene_index": 11,
                "shot_index": 20,
                "description": "민숙의 손목에 있는 둥근 표식이 붉은색을 띤 클로즈업.",
                "keep_elements": [
                  {
                    "label": "limp hand and exposed wrist of the motionless middle-aged woman",
                    "kind": "immobilized_character"
                  },
                  {
                    "label": "crumpled photograph beside the hand",
                    "kind": "static_prop"
                  },
                  {
                    "label": "blood-spotted floor with dark smears",
                    "kind": "environment"
                  },
                  {
                    "label": "dark bed base shadow at the upper edge",
                    "kind": "environment"
                  }
                ]
              },
              {
                "scene_index": 23,
                "shot_index": 6,
                "description": "붉은 원 정중앙에 날카로운 핀으로 꽂힌, 핏자국이 묻은 낡은 사진 한 장 클로즈업.",
                "keep_elements": [
                  {
                    "label": "old yellowed blood-stained photograph pierced by a small metal pin",
                    "kind": "static_prop"
                  },
                  {
                    "label": "uneven red circular mark painted on the wall",
                    "kind": "static_prop"
                  },
                  {
                    "label": "dim damp-stained wall surface",
                    "kind": "environment"
                  }
                ]
              }
            ],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "background",
                "id": "L04B06",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "dad733019e0c1d7c"
      },
      {
        "scene_index": 11,
        "heading": "S#12. 옥탑방 안",
        "beat_title": "붉게 변하는 손목의 문신",
        "representative_moment": "구겨진 사진을 쥔 민숙의 늘어진 손목 위에서, 붉은색으로 변해가는 원형 표식을 익스트림 클로즈업으로 비춘다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Extreme close-up, overhead bird's-eye angle at hand level, shallow rack focus on the wrist mark, dim blue-gray lighting with wet red emphasis.",
            "t2i_prompt": "Photorealistic cinematic still. Extreme close-up from an overhead bird's-eye angle at hand level, tight focus on the exposed wrist. The limp hand of a Korean, early 40s woman in a grey homewear sleeve crosses the lower third of the frame, clutching a crumpled photograph kept partially visible at frame left. The creased printed face of the photograph is partly turned toward the camera, its image content obscured by folds and blood smears. Dead center on the upward-facing skin of the wrist, P08, a circular mark, is visibly reddening with a subtle wet red emphasis. The environment is submerged in oppressive, dim blue-gray lighting with a weak muted spill, maintaining a grim tone. [: A blood-spotted floor with sticky dark red smears sits directly below the hand, while the dim side of a bed base forms a dark horizontal shadow beyond the wrist.]",
            "outfit_assignments": [],
            "source_facts": [
              "S12 action: '엄마 손안에 구겨진 사진 한 장이 보인다. ... 엄마 손목에 있던 문신이 붉은색으로 변한다.'",
              "shot description: ECU, overhead / bird's eye angle at hand level, focus on the circular wrist mark."
            ],
            "visual_inferences": [
              "fixed_element dim_blue_gray_lighting: The space is submerged in oppressive, humid shadows with a cool, blue-gray tone",
              "since this is a tight body-part focus on the wrist, the character's face is excluded and represented purely by the hand",
              "the crumpled photograph and blood-spotted floor anchor the scene physically while the mark turns red"
            ],
            "creative_decisions": [
              "Used a common noun with demographic descriptor ('limp hand of a Korean, early 40s woman') per the isolated body-part focus rule, preventing detached face synthesis.",
              "Emphasized the wet red glow of P08 against the cool blue-gray ambient shadows to highlight the supernatural shift within gritty realism.",
              "Kept the floor and bed base shadow as the background [L01] to maintain geometrically consistent overhead framing."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "ef6640e65235f5d1",
              "owned_hash": "930b9286d35f0313",
              "camera_direction_hash": "954f06c972415b82",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 11,
        "dependency_reason": "zoom_in_detail from previous shot's established room and dim blue-gray lighting.",
        "_shot_index": 20,
        "visible_entities": [
          "C03",
          "P08"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 11,
            "shot_index": 20
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Beside 민숙 on the bedroom floor, the camera completes the evidence-detail push into an ECU, overhead / bird's eye angle at hand level, with a shallow rack focus from the crumpled photograph past her limp fingers to the circular wrist mark. The reddening mark sits dead center, her hand crossing the lower third and the photo kept partially visible at frame left so the supernatural change stays tied to physical evidence.",
            "lighting_mood": "The room stays dim and blue-gray, but the wrist symbol emits a subtle wet red emphasis that feels occult without breaking gritty realism.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C03",
              "P08"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C03",
                "outlook_id": "O06"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B06",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "dim_blue_gray_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The interior space is submerged in oppressive, humid shadows characterized by a strong cool, blue-gray tone. A weak, muted spill of light from an off-camera source barely illuminates the space, maintaining a grim, dark, and gloomy atmosphere.",
                "applies_to_shots": [
                  11,
                  14,
                  20
                ],
                "source_facts": [
                  "Shot 11 lighting: 'dim room lamp leave the floor in humid gray shadow'",
                  "Shot 14 lighting: 'Muted lamp spill... carve silhouette into a gray-blue silhouette'",
                  "Shot 20 lighting: 'The room stays dim and blue-gray'"
                ],
                "visual_inferences": [
                  "The ambient light is extremely low, relying on a single weak artificial light source that casts cool, unsettling shadows."
                ],
                "creative_decisions": [
                  "Unifying the visual tone across wide and close-up shots by locking in the specific cool blue-gray color grading to enhance the horror atmosphere."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 11,
                "shot_index": 14,
                "ref_usage": "zoom_in_detail",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "eef27edaf99cf221"
      },
      {
        "scene_index": 12,
        "heading": "S#13. 옥탑방 외곽 / N",
        "beat_title": "충격에 빠져 혜수를 올려다보는 수리영",
        "representative_moment": "눈물을 머금은 채 위쪽의 혜수를 향해 고개를 들어 올리는 C01의 얼굴을 잡은 클로즈업.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "low angle head-and-shoulder composition, shallow depth of field, muted coastal night fog with red siren accents",
            "t2i_prompt": "Photorealistic cinematic still. A low-angle head-and-shoulder composition. The focal point is the tear-wet expression of C01O02 in a beige hoodie, a Korean young woman in her early 20s, capturing her immediate shock and distress. She is positioned at three-quarter angle from the right, chin tilted upward, eyes looking up toward an off-camera figure, with her neck and upper chest visible in the lower frame. In the shallow depth of field behind her cheek, a damp rain-slick stair rail runs diagonally away into the darkness. Muted coastal night light with fog-softened contrast washes the scene, while a faint pulsing red emergency siren light reflects on her wet cheek and the dim railing.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "S#13 action: '비명을 지르며 옥탑방을 빠져나오는 수리영... (올라다보며)'",
              "Shot 4 description: '눈물을 머금은 채 혜수 쪽으로 고개를 든 수리영의 얼굴 클로즈업'",
              "Shot 4 DP instructions: 'A tight CU, slightly low angle, from the foot of the rooftop stairs... lifts her tear-wet face toward off-camera 혜수'"
            ],
            "visual_inferences": [
              "fixed_element environment_state: Faint, pulsing red emergency siren lights from outside",
              "Shallow depth of field isolates her face and blurs the damp stair railing to emphasize her emotional distress"
            ],
            "creative_decisions": [
              "Used a low-angle head-and-shoulder composition to frame her upward gaze while avoiding the 'floating head' body-part focus trap",
              "Contrasted the cool, fog-softened coastal night lighting with the ominous red reflection of the police siren to heighten the thriller tension"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "e2af5ba2c38430e8",
              "owned_hash": "13f8b28591f9882d",
              "camera_direction_hash": "f28cfac769a6e57c",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 11,
        "dependency_reason": "atmosphere_reference: 옥탑방 외부 계단의 안개 낀 차가운 해안가 밤 분위기(Muted coastal night light) 톤 유지",
        "_shot_index": 4,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 12,
            "shot_index": 4
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A tight CU, slightly low angle, from the foot of the rooftop stairs, holding just below 수리영’s eyeline as she lifts her tear-wet face toward off-camera 혜수. The framing uses shallow DOF: 수리영 occupies the center-left 70% of frame, with the damp stair rail and a faint red siren smear compressed behind her; 혜수 is kept fully off-frame and exists only through 수리영’s upward gaze.",
            "lighting_mood": "Muted coastal night light with fog-softened contrast, a cold practical spill from the rooftop room, and a faint red emergency flicker catching her tears.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B07",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "emergency_siren_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "Faint, pulsing red emergency siren lights from outside, casting subtle red accents and reflections across the space to create an uneasy and tense atmosphere.",
                "applies_to_shots": [
                  4,
                  13
                ],
                "source_facts": [
                  "S#13 action: '동시에 사이렌이 울리고'",
                  "Shot 4 lighting: 'a faint red emergency flicker catching her tears'",
                  "Shot 13 lighting: 'with the red accents now reduced to a barely perceptible unease'"
                ],
                "visual_inferences": [
                  "The pulsing red light from the arrived police car reaches both the exterior stairs and the interior living area."
                ],
                "creative_decisions": [
                  "Maintained the red siren light as a subtle, unifying atmospheric element across the active shots to emphasize the lingering tension of the crime scene."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 11,
                "shot_index": 14,
                "ref_usage": "atmosphere_reference",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "da7896f82d4e56fb"
      },
      {
        "scene_index": 12,
        "heading": "S#14. 옥탑방 안 / N",
        "beat_title": "사라진 시신과 깨끗해진 방",
        "representative_moment": "A static wide shot of the unnervingly spotless bedroom, with only the slightly shifted backpack and doll left behind.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "A static wide-angle shot at eye-level, cold grayish-blue dawn light mixing with dim practical lighting",
            "t2i_prompt": "Photorealistic cinematic still. A static wide-angle shot at eye-level from just outside the door from the reference, looking into the bedroom. Focus on the unnervingly spotless and dry floor from the reference in the center, an empty zone leaving no trace of blood. The mattress from the reference sits mid-right, neatly arranged, its long side facing the camera. On the lower-left floor, the backpack from the reference rests with its front pocket facing partly toward the camera, its straps collapsed. Nearby in the lower center, the motionless doll from the reference lies with its face and torso angled up toward the camera. In the background, the curtain from the reference hangs closed, veiling the window from the reference. The wardrobe from the reference, the wall mark from the reference, and the baseboard from the reference complete the background structure. In the immediate foreground outside the doorway, a dark wood dining table with a humid sheen holds a half-full clear water glass with condensation and the framed photograph from the reference displaying a warped middle-aged woman's portrait. A cold, grayish-blue ambient dawn light seeps in, mixing with dim practical light from the light fixture from the reference, rendering the space in a sterile, cool tone.",
            "outfit_assignments": [],
            "source_facts": [
              "S14 text: 지나칠 정도로 깨끗하게 정돈된 방안. 핏자국도 보이지 않고, 침대 위에 붉은 표식은 물론 커튼 뒤의 시신도 감쪽 같이 사라졌다.",
              "S14 text: 수리영 방안에 놓여진 백팩, 인형. 약간씩 위치가 달라진.",
              "DP notes: A static WS at eye-level from just outside 수리영’s bedroom doorway, looking across the spotless bed and floor area like a held evidentiary frame."
            ],
            "visual_inferences": [
              "fixed_element cool_interior_ambient_light: A cold, grayish-blue ambient dawn light seeping into the interior of the house, rendering the surrounding walls and empty space in a sterile, cool tone.",
              "The dark wood dining table, water glass, and framed photo are positioned in the immediate foreground to seamlessly set up the forward zoom target for a subsequent shot.",
              "The backpack and doll remain on the floor, slightly shifted, highlighting the eerie absence of the body."
            ],
            "creative_decisions": [
              "Used 'from the reference' anchoring for all background objects to strictly enforce the exact_background binding policy.",
              "Emphasized the negative space (the spotless floor) as the primary focus to convey the sudden disappearance of the crime scene.",
              "Contrasted the sterile, cool dawn light with dim practical lighting to amplify the unsettling psychological horror of the moment."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "d8f9e87aef9bfbd7",
              "owned_hash": "13f8b28591f9882d",
              "camera_direction_hash": "a647b1f4086409a6",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "door",
                  "violating_phrase": "from just outside the door from the reference",
                  "reason": "The door is only referenced as an existing anchor point with an explicit reference mention, not newly created or modified.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "floor",
                  "violating_phrase": "the spotless and dry floor from the reference",
                  "reason": "The floor is described with state/quality and reference anchoring, not drawn or added anew.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "mattress",
                  "violating_phrase": "The mattress from the reference sits mid-right",
                  "reason": "The mattress is positioned as an existing reference object, with no redraw verb directed at it.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "backpack",
                  "violating_phrase": "the backpack from the reference rests",
                  "reason": "The backpack is mentioned as an existing reference object in a spatial description, not as a newly created object.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "doll",
                  "violating_phrase": "the motionless doll from the reference lies",
                  "reason": "The doll is referenced as an existing object and only described by pose and state.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "curtain",
                  "violating_phrase": "the curtain from the reference hangs closed",
                  "reason": "The curtain is an anchored existing object, described with state and reference language.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "window",
                  "violating_phrase": "the window from the reference",
                  "reason": "The window is explicitly tied to the reference and used as background anchoring, not redraw instruction.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "wardrobe",
                  "violating_phrase": "The wardrobe from the reference",
                  "reason": "The wardrobe is only referenced as an existing background element.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "wall mark",
                  "violating_phrase": "the wall mark from the reference",
                  "reason": "The wall mark is referenced as an existing element with explicit reference anchoring.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "baseboard",
                  "violating_phrase": "the baseboard from the reference",
                  "reason": "The baseboard is mentioned as an existing structural detail and not as a new addition.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "light fixture",
                  "violating_phrase": "the light fixture from the reference",
                  "reason": "The light fixture is an existing referenced object used for ambient lighting context.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 12,
        "dependency_reason": "exact_background mode applied to preserve the bedroom geography established in the previous night scene, ensuring spatial consistency.",
        "_shot_index": 8,
        "visible_entities": [
          "P04",
          "L04"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 12,
            "shot_index": 8
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A static WS at eye-level from just outside 수리영’s bedroom doorway, looking across the spotless bed and floor area like a held evidentiary frame. The bed sits mid-right, the backpack rests low-left, and the doll is slightly separated near the lower center, leaving an unnerving empty zone where blood and the body should have been.",
            "lighting_mood": "Dim interior practical light mixes with bluish coastal dawn seepage, making the clean room feel unnaturally sterile and humid.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "P04",
              "L04"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L04B07",
            "reference_usage": "exact_background",
            "owned_objects": [
              "backpack",
              "baseboard",
              "curtain",
              "doll",
              "door",
              "floor",
              "light fixture",
              "mattress",
              "photo",
              "wall mark",
              "wardrobe",
              "window"
            ],
            "camera_reference": {
              "camera_position": "near number 20 (main bedroom door), facing toward number 21 (low bed), number 27 (backpack), number 28 (doll), and number 23 (curtain partition)",
              "camera_height": "eye-level standing ~1.55m",
              "lens_hint": "35mm wide angle",
              "framing_notes": "Use the same bedroom geography as the prior night scene, but emphasize the now-clean floor and shifted personal items."
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "when focusing on an owned object, anchor it explicitly: 'the door from the reference'",
              "camera_reference specifies the framing under which the background was painted — match camera_position, camera_height, and framing",
              "if deviating from the reference framing (push-in, angle shift, ECU), state the deviation explicitly so the compositor can account for the scale difference"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "cool_interior_ambient_light",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A cold, grayish-blue ambient dawn light seeping into the interior of the house, rendering the surrounding walls and empty space in a sterile, cool tone.",
                "applies_to_shots": [
                  8,
                  13
                ],
                "source_facts": [
                  "Shot 8 lighting: 'bluish coastal dawn seepage, making the clean room feel unnaturally sterile'",
                  "Shot 13 lighting: 'the surrounding room stays cool and gray'"
                ],
                "visual_inferences": [
                  "The early dawn light from the coastal area uniformly affects the interior color palette of the rooftop house, contrasting with localized practical lights."
                ],
                "creative_decisions": [
                  "Using a cool, sterile color temperature for the background interior to highlight the unnatural emptiness where the body used to be."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 12,
                "shot_index": 4,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [
              {
                "scene_index": 16,
                "shot_index": 14,
                "description": "액자 속 사진 표면에 찍힌 중년 여자의 웃는 얼굴이 기괴한 형태로 일그러진 정지 이미지 클로즈업.",
                "keep_elements": [
                  {
                    "label": "framed photograph with a warped middle-aged woman's portrait",
                    "kind": "static_prop"
                  },
                  {
                    "label": "dark wood dining table with a humid sheen",
                    "kind": "environment"
                  },
                  {
                    "label": "half-full clear water glass with condensation",
                    "kind": "static_prop"
                  }
                ]
              }
            ],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "background",
                "id": "L04B07",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "a22ce83db697fd02"
      },
      {
        "scene_index": 12,
        "heading": "S#14. 옥탑방 안 / N",
        "beat_title": "혜수의 다정한 위로와 보호",
        "representative_moment": "혜수가 수리영의 어깨를 양손으로 감싸며 다정하게 안심시킨다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "An MCU at eye-level, frontal composition with an intimate OTS-like foreground connection, shallow depth of field.",
            "t2i_prompt": "Photorealistic cinematic still. An MCU at eye-level, frontal composition. C02O04 in a detective coat, East Asian middle-aged woman, facing the camera, upper body in frame, hands gripping the shoulders of C01O02, eyes fixed on C01O02. C01O02 in a beige hoodie, East Asian young woman, with back to camera, shoulders held tense anchoring the lower foreground, eyes fixed on C02O04. The shared space between them forms an intimate OTS connection. Camera focuses on C02O04's central position. [: out-of-focus open bedroom doorway frame and a clean living-area table edge receding in the soft background, small fogged window showing a dull gray exterior glow]. Soft practical light warms C02O04's face, while a cold, grayish-blue ambient dawn light renders the surrounding frame edges in a sterile, cool tone. Faint, pulsing red emergency siren lights reflect across C01O02's hoodie shoulder.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              },
              {
                "character_id": "C02",
                "outlook_id": "O04"
              }
            ],
            "source_facts": [
              "S#14 action: '수리영 어깨를 어루만지며 영아.'",
              "Shot 13 DP: '혜수 is centered squarely while 수리영’s shoulders anchor the lower foreground. 혜수’s hands grip both of 수리영’s shoulders near the bottom third'"
            ],
            "visual_inferences": [
              "fixed_element cool_interior_ambient_light: A cold, grayish-blue ambient dawn light rendering the surrounding soft background in a sterile tone",
              "fixed_element emergency_siren_lighting: Faint red emergency siren lights casting subtle reflections on foreground textures"
            ],
            "creative_decisions": [
              "Used skipped_close_framing strategy with an [L01] block instead of L04 to loosely describe the out-of-focus background frame edges without importing the previous exact room layout.",
              "Positioned C01O02 explicitly 'with back to camera' to enforce the OTS foreground anchor while avoiding a second sharp face in the close frame."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "e86153a65165756b",
              "owned_hash": "13f8b28591f9882d",
              "camera_direction_hash": "3921866889853abc",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 12,
        "dependency_reason": "Maintains the exact environmental lighting state from the prior shots in the same space, focusing tightly on the character interaction.",
        "_shot_index": 13,
        "visible_entities": [
          "C01",
          "C02"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 12,
            "shot_index": 13
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "An MCU at eye-level, frontal and gently pushing in, positioned in the living area near the bedroom side so 혜수 is centered squarely while 수리영’s shoulders anchor the lower foreground. The composition uses an intimate OTS-like foreground connection: 혜수’s hands grip both of 수리영’s shoulders near the bottom third, and 혜수’s face occupies the central upper half with the softened room receding behind her.",
            "lighting_mood": "Soft practical light warms 혜수’s face while the surrounding room stays cool and gray, with the red accents now reduced to a barely perceptible unease.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01",
              "C02"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              },
              {
                "character_id": "C02",
                "outlook_id": "O04"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B07",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "emergency_siren_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "Faint, pulsing red emergency siren lights from outside, casting subtle red accents and reflections across the space to create an uneasy and tense atmosphere.",
                "applies_to_shots": [
                  4,
                  13
                ],
                "source_facts": [
                  "S#13 action: '동시에 사이렌이 울리고'",
                  "Shot 4 lighting: 'a faint red emergency flicker catching her tears'",
                  "Shot 13 lighting: 'with the red accents now reduced to a barely perceptible unease'"
                ],
                "visual_inferences": [
                  "The pulsing red light from the arrived police car reaches both the exterior stairs and the interior living area."
                ],
                "creative_decisions": [
                  "Maintained the red siren light as a subtle, unifying atmospheric element across the active shots to emphasize the lingering tension of the crime scene."
                ],
                "confidence": "high"
              },
              {
                "element_id": "cool_interior_ambient_light",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A cold, grayish-blue ambient dawn light seeping into the interior of the house, rendering the surrounding walls and empty space in a sterile, cool tone.",
                "applies_to_shots": [
                  8,
                  13
                ],
                "source_facts": [
                  "Shot 8 lighting: 'bluish coastal dawn seepage, making the clean room feel unnaturally sterile'",
                  "Shot 13 lighting: 'the surrounding room stays cool and gray'"
                ],
                "visual_inferences": [
                  "The early dawn light from the coastal area uniformly affects the interior color palette of the rooftop house, contrasting with localized practical lights."
                ],
                "creative_decisions": [
                  "Using a cool, sterile color temperature for the background interior to highlight the unnatural emptiness where the body used to be."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 12,
                "shot_index": 8,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              },
              {
                "kind": "character_outlook",
                "id": "C02O04",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "9a35c639852c52c1"
      },
      {
        "scene_index": 13,
        "heading": "S#15. 해안가. 버스정류장 부근 / N",
        "beat_title": "자신의 목격에 대한 의심과 직접 확인하려는 결심",
        "representative_moment": "벤치에서 몸을 반쯤 일으킨 채 굳은 입술로 정면(어둠)을 응시하는 C01의 결연한 모습",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "A slightly low angle, shallow depth of field head-and-shoulder composition with a cold blue-gray coastal night mood.",
            "t2i_prompt": "Photorealistic cinematic still. Head-and-shoulder composition at a slightly low angle. C01O02 in a beige hoodie, a young Korean woman in her early 20s with dark hair, at three-quarter angle from the left, half-risen from a rain-damp bench. Her torso leans forward center-right, eyes fixed on the distant darkness, firm lips tightened in resolve. The wet front edge of the empty bench cuts across the lower frame, receding toward the left beneath her. A weak fluorescent light reflects on her face, while cold blue-gray coastal night ambient washes the figure. Just behind her, a dim practical glow from a damp bus timetable board catches her shoulder, its printed schedule face turned obliquely toward the camera with unreadable text. Streaked wet glass with condensation forms a soft background, blurring the dark sea road into deep negative space. Focus on her resolute expression.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "Shot 3: '벤치에서 몸을 반쯤 일으킨 채 입술을 굳게 다문 수리영의 결연한 표정'",
              "Shot description: 'caught half-risen, her firm lips and resolved eyes tightened in shallow focus'"
            ],
            "visual_inferences": [
              "fixed_element damp_coastal_bus_stop: cold fluorescent glow, damp bench reflecting faint artificial light",
              "half-risen posture implies mid-action forward momentum from the bench"
            ],
            "creative_decisions": [
              "Slightly low angle to emphasize her sudden decision and resolve",
              "Shallow focus on her expression with blurred streaked glass behind her to isolate her within the cold coastal environment"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "4e4d12aaccfd5052",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "ea4b5928fb6c1906",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "N/A",
        "_shot_index": 3,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 13,
            "shot_index": 3
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A tight MCU, slightly low angle, from within arm’s length of the bus-stop bench as 수리영 is caught half-risen, her firm lips and resolved eyes tightened in shallow focus. The bench cuts across the lower frame, her torso leans forward center-right, and the dark sea road falls into blurred negative space behind her.",
            "lighting_mood": "Cold blue-gray coastal night wraps the shelter, with a weak fluorescent practical flattening her face and moisture glinting on the bench.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": null,
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "damp_coastal_bus_stop",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A dark, damp coastal roadside at night, featuring a bus stop shelter that casts a weak, cold fluorescent glow. The paved ground is wet, reflecting the faint artificial light, while the surrounding background is swallowed by deep, blue-black coastal darkness.",
                "applies_to_shots": [
                  3,
                  4,
                  5
                ],
                "source_facts": [
                  "Shot 3: 'weak fluorescent practical flattening her face and moisture glinting on the bench'",
                  "Shot 4: 'wet road', 'roadside sodium spill', 'darkness ahead'",
                  "Shot 5: 'Weak shelter light', 'damp blue-black roadside'"
                ],
                "visual_inferences": [
                  "The asphalt remains uniformly wet across the scene, creating a consistent cold reflection of the nearby artificial lights."
                ],
                "creative_decisions": [
                  "Emphasizing a cold blue-gray color palette for the night environment to amplify the sense of isolation and uncertainty."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "38da50a348595721"
      },
      {
        "scene_index": 13,
        "heading": "S#15. 해안가. 버스정류장 부근 / N",
        "beat_title": "어둠 속으로 달려가는 수리영",
        "representative_moment": "C01 runs away from the bus stop into the darkness.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Kinetic medium wide shot, low angle tracking at waist height, cold blue-gray palette with roadside sodium highlights.",
            "t2i_prompt": "Photorealistic cinematic still. C01O02 in a beige hoodie, a Korean woman in her 20s, with her back largely to the camera and head turned slightly revealing her left profile, eyes fixed on the distant darkness, captured in a mid-stride posture. At waist height tracking behind and slightly to her left, her back and pumping arms in frame, legs implied beyond the lower frame edge. Her body leans forward, the hoodie collar flaring backward in the direction opposite to her forward motion. Focus on her tense posture and the side of her face. Behind her on the left edge, a damp coastal bus stop shelter casts a weak, cold fluorescent glow. The wet paved road reflects a thin roadside sodium spill, while foggy, deep blue-black coastal darkness swallows the open space ahead.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "S#15 action: '수리영, 응. 나 갈게 (달려간다)'",
              "shot description: tracking behind and slightly to the side of 수리영 at waist-to-shoulder height as she runs away",
              "fixed_element: damp_coastal_bus_stop with weak fluorescent glow"
            ],
            "visual_inferences": [
              "C01 is in a mid-stride freeze posture with collar flaring backward opposite to motion to preserve momentum",
              "Waist-height camera places her torso and arms in frame while lower legs are implied beyond the edge, complying with camera_frame_rule",
              "Head turned slightly to show left profile allows C01O02 face identifiability while keeping the back-to-camera tracking angle"
            ],
            "creative_decisions": [
              "Emphasized the foggy blue-black coastal darkness in the background to contrast with the warm sodium spill on the wet road",
              "Focused on the trailing hoodie fabric to convey kinetic forward momentum in a single static frame"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "746c5b35b315771b",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "b538df2e22f299df",
              "validator": "scene_detail_owned_objects.v1",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 13,
        "dependency_reason": "Maintains the exact background of the damp coastal bus stop established in the prior shot.",
        "_shot_index": 4,
        "visible_entities": [
          "C01",
          "L09"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 13,
            "shot_index": 4
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A kinetic MWS, low angle, tracking behind and slightly to the side of 수리영 at waist-to-shoulder height as she runs away from the bus stop. Her back-left side occupies the right third, one foot just lifted from the wet road and her collar flaring backward, while the bus stop shrinks behind her on the left edge.",
            "lighting_mood": "The palette stays cold and humid, with thin roadside sodium spill skimming her moving silhouette and the darkness ahead swallowing detail.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01",
              "L09"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "not_applicable",
            "bg_id": null,
            "reference_usage": "none",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": false,
            "constraints": [
              "no chain_bg is bound to this shot — describe the location freely without referencing a background image",
              "do not use 'from the reference' / 'the existing room' / 'preserving the room' style phrasing — there is no reference image to refer to"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "damp_coastal_bus_stop",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A dark, damp coastal roadside at night, featuring a bus stop shelter that casts a weak, cold fluorescent glow. The paved ground is wet, reflecting the faint artificial light, while the surrounding background is swallowed by deep, blue-black coastal darkness.",
                "applies_to_shots": [
                  3,
                  4,
                  5
                ],
                "source_facts": [
                  "Shot 3: 'weak fluorescent practical flattening her face and moisture glinting on the bench'",
                  "Shot 4: 'wet road', 'roadside sodium spill', 'darkness ahead'",
                  "Shot 5: 'Weak shelter light', 'damp blue-black roadside'"
                ],
                "visual_inferences": [
                  "The asphalt remains uniformly wet across the scene, creating a consistent cold reflection of the nearby artificial lights."
                ],
                "creative_decisions": [
                  "Emphasizing a cold blue-gray color palette for the night environment to amplify the sense of isolation and uncertainty."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 13,
                "shot_index": 3,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "7d553398a64df43b"
      },
      {
        "scene_index": 13,
        "heading": "S#15. 해안가. 버스정류장 부근 / N",
        "beat_title": "수리영이 집으로 달려가고 혜수가 소리침",
        "representative_moment": "혜수가 멀어지는 수리영을 향해 한 손을 입가에 대고 소리친다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "A static MS at eye-level near the edge of the bus stop area, looking back across the roadside toward 혜수 as she calls after 수리영. Weak shelter light isolates 혜수 against a damp blue-black roadside, with the background darkness left ominously unfilled.",
            "t2i_prompt": "Photorealistic cinematic still. C02O04 in investigator field wear, an East Asian woman in her early 40s with black hair and dark eyes, stands at three-quarter angle from the right in the left-center upper body area, one hand cupped to her mouth as she shouts, eyes fixed on the distant figure. In the right half of the frame, held open as dark negative space, an East Asian figure, a young woman, is seen with back to camera, looking away and receding into the darkness as a small distant figure. The characters occupy the same dark, damp coastal roadside at night, the wet paved ground reflecting the weak, cold fluorescent glow from the bus stop shelter roof light above C02O04. The rain-streaked glass panel of the shelter catches C02O04's faint outline behind her. Focus on C02O04's expressive face calling out, framed against the deep, blue-black coastal darkness swallowing the misty, underlit road ahead.",
            "outfit_assignments": [
              {
                "character_id": "C02",
                "outlook_id": "O04"
              }
            ],
            "source_facts": [
              "S#15 action: '혜수 (등 뒤에 대고 소리친다) 엄마 보면 나한테 전화 하라고 해!'",
              "shot description: 멀어지는 수리영을 향해 한 손을 입가에 댄 채 소리치는 혜수의 상체 컷."
            ],
            "visual_inferences": [
              "fixed_element damp_coastal_bus_stop: dark, damp coastal roadside at night, weak cold fluorescent glow, wet paved ground, deep blue-black coastal darkness.",
              "DP instruction: static MS at eye-level, 혜수 fills left-center upper body area with one hand cupped to mouth, right half is dark negative space where 수리영 recedes as a small distant figure.",
              "DP instruction: 혜수 three_quarter_right, eyes→수리영; 수리영 looking_away, eyes→distant.",
              "C02 demographic: East Asian, early 40s woman.",
              "수리영 (unregistered extra in this shot context but visible as small figure): East Asian young woman, back to camera."
            ],
            "creative_decisions": [
              "Applied medium shot (MS) framing, ensuring C02O04's upper body is visible without cutting off awkwardly, while keeping the distant figure small and in the background to emphasize the negative space and isolation.",
              "Used shared spatial anchor ('the characters occupy the same dark, damp coastal roadside') to legally bind the foreground subject and background receding figure.",
              "Integrated the damp coastal environment and weak fluorescent practical lighting to match the exact_background ref constraint from Shot 4."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "1363c3abf0902cca",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "6cdff6c1b1efa72e",
              "validator": "scene_detail_owned_objects.v1",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "",
        "_shot_index": 5,
        "visible_entities": [
          "C02"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 13,
            "shot_index": 5
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A static MS at eye-level near the edge of the bus stop area, looking back across the roadside toward 혜수 as she calls after 수리영. 혜수 fills the left-center upper body area with one hand cupped to her mouth, while the right half of the frame is held open as dark negative space where 수리영 recedes as a small distant figure.",
            "lighting_mood": "Weak shelter light isolates 혜수 against a damp blue-black roadside, with the background darkness left ominously unfilled.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C02"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C02",
                "outlook_id": "O04"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "not_applicable",
            "bg_id": null,
            "reference_usage": "none",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": false,
            "constraints": [
              "no chain_bg is bound to this shot — describe the location freely without referencing a background image",
              "do not use 'from the reference' / 'the existing room' / 'preserving the room' style phrasing — there is no reference image to refer to"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "damp_coastal_bus_stop",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A dark, damp coastal roadside at night, featuring a bus stop shelter that casts a weak, cold fluorescent glow. The paved ground is wet, reflecting the faint artificial light, while the surrounding background is swallowed by deep, blue-black coastal darkness.",
                "applies_to_shots": [
                  3,
                  4,
                  5
                ],
                "source_facts": [
                  "Shot 3: 'weak fluorescent practical flattening her face and moisture glinting on the bench'",
                  "Shot 4: 'wet road', 'roadside sodium spill', 'darkness ahead'",
                  "Shot 5: 'Weak shelter light', 'damp blue-black roadside'"
                ],
                "visual_inferences": [
                  "The asphalt remains uniformly wet across the scene, creating a consistent cold reflection of the nearby artificial lights."
                ],
                "creative_decisions": [
                  "Emphasizing a cold blue-gray color palette for the night environment to amplify the sense of isolation and uncertainty."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 13,
                "shot_index": 4,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C02O04",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "4ced9b2e1e8d7c92"
      },
      {
        "scene_index": 14,
        "heading": "경찰서 사무실 / N",
        "beat_title": "불길한 예감과 위치추적 요구",
        "representative_moment": "혜수가 김형사의 얼굴 앞으로 P01(스마트폰)을 불쑥 들이밀며 화면을 보여주는 순간.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "A side-composed MCU at chest height, close framing on the immediate interaction",
            "t2i_prompt": "Photorealistic cinematic still. A side-composed MCU at chest height. C02O04 in a dark investigator jacket, an East Asian woman in her 40s, seen over the shoulder in the left third of the frame with back to camera, her tense shoulder and extended arm holding P01 out forward. P01, a smartphone, crosses the air between them just below the other figure's face, its lit screen angled partly toward the camera showing a saved contact number. Across the shared duty desk edge, C04O08 in a dark detective coat, an East Asian man in his 30s, occupies the mid-right in left profile, leaning back slightly, eyes fixed on P01. Focus on P01 and C04O08's tense profile. Flat greenish fluorescent overhead light falling on the scene, with the cold glow of P01 reflecting on C04O08's cheek and coat lapel. [L10: messy duty desk surface with scattered night-shift papers crossing the lower frame, the paper-covered front face of an office notice board visible directly behind C04O08 with official notices blurred and unreadable].",
            "outfit_assignments": [
              {
                "character_id": "C02",
                "outlook_id": "O04"
              },
              {
                "character_id": "C04",
                "outlook_id": "O08"
              }
            ],
            "source_facts": [
              "Scene text: '사무실 안에는 김형사 혼자 당직 근무 중이다. ... (핸드폰에 저장된 민숙번호 보여주며) 이 번호 위치 좀 따봐.'",
              "Shot 3 description: 'A side-composed MCU at chest height, dollying in from the duty desk side across 혜수’s shoulder as her arm thrusts a phone into 김형사’s personal space.'"
            ],
            "visual_inferences": [
              "fixed_element night_shift_office_environment: Flat greenish fluorescent light casting cool damp shadows",
              "C02 is partial OTS in the foreground left, placing C04 in full view as the primary subject reacting in mid-right profile",
              "The phone screen acts as a practical light source reflecting on C04's face, enhancing the night-shift tension"
            ],
            "creative_decisions": [
              "Applied skipped_close_framing background mode, excluding wider room geometry and solely utilizing L10 block for the messy desk edge and the blurred notice board behind C04",
              "Described the lit screen orientation and the blurred paper-covered front face of the board exactly per DP constraints"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "efb37b02f332d124",
              "owned_hash": "f747b52ba12b45e1",
              "camera_direction_hash": "558b44ea92fb4eb2",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "Not dependent on a prior visual scene output; background is constructed via close framing rules solely from frame edges.",
        "_shot_index": 3,
        "visible_entities": [
          "C02",
          "C04",
          "P01",
          "L10"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 14,
            "shot_index": 3
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A side-composed MCU at chest height, dollying in from the duty desk side across 혜수’s shoulder as her arm thrusts a phone into 김형사’s personal space. The phone screen cuts into the foreground just below 김형사’s face without overwhelming the frame, while 김형사 sits mid-right in profile and 혜수’s tense shoulder and extended arm occupy the left third.",
            "lighting_mood": "Flat greenish fluorescent office light is broken by the cold phone glow, giving the confrontation a damp, exhausted night-shift tension.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C02",
              "C04",
              "P01",
              "L10"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C02",
                "outlook_id": "O04"
              },
              {
                "character_id": "C04",
                "outlook_id": "O08"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L10B02",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "night_shift_office_environment",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The interior of an institutional office at night, illuminated by flat, sickly greenish fluorescent overhead lights. The harsh lighting casts cool, damp shadows across the room, establishing an exhausted and tense atmosphere. A faint, cold glow from a mobile device slightly pierces the ambient greenish light, while the background features muted, indistinct office furniture and duty desks.",
                "applies_to_shots": [
                  3,
                  5
                ],
                "source_facts": [
                  "Scene text: '경찰서 사무실 / N'",
                  "Shot 3 description: 'Flat greenish fluorescent office light is broken by the cold phone glow, giving the confrontation a damp, exhausted night-shift tension'",
                  "Shot 5 description: 'Sickly fluorescent light and the faint phone glow'"
                ],
                "visual_inferences": [
                  "Standard institutional office setting with overhead panel lights",
                  "The overall color palette is dominated by cold green and bluish tones due to the fluorescent fixtures"
                ],
                "creative_decisions": [
                  "Using the greenish tint to visually reinforce the fatigue and anxiety of the late-night shift",
                  "Keeping the background details relatively generic and muted to keep the focus on the foreground subjects"
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [
              {
                "scene_index": 14,
                "shot_index": 5,
                "description": "휴대폰을 꽉 쥔 채 허공을 노려보는 혜수의 불안한 표정.",
                "keep_elements": [
                  {
                    "label": "the same Korean woman in a dark jacket holding a smartphone",
                    "kind": "immobilized_character"
                  },
                  {
                    "label": "cold fluorescent police office lighting",
                    "kind": "environment"
                  },
                  {
                    "label": "blurred duty desk and office notice board in the background",
                    "kind": "environment"
                  },
                  {
                    "label": "the smartphone with faint cold screen glow",
                    "kind": "static_prop"
                  }
                ]
              }
            ],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C02O04",
                "policy": "required"
              },
              {
                "kind": "character_outlook",
                "id": "C04O08",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "bd9866e267dae55c"
      },
      {
        "scene_index": 14,
        "heading": "S#16. 경찰서 사무실 / N",
        "beat_title": "불길한 예감 속 긴장",
        "representative_moment": "혜수(C02)가 극심한 불안감에 휩싸여 P01(스마트폰)을 꽉 쥔 채 허공을 응시하고 있다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Eye-level close-up, sickly green fluorescent ambient mixed with cold screen glow, shallow depth of field",
            "t2i_prompt": "Photorealistic cinematic still. Close-up at eye-level, head-and-shoulder composition, the same figure. C02O04 in a dark investigator jacket, a Korean, early-40s woman with black hair, dark eyes, and a middle-aged female face, positioned at a three-quarter angle from the left. Her face sits slightly left of center as she freezes with anxiety, her eyes fixed on the empty space past the camera. Near the lower frame edge, her hands tightly clutch P01, a smartphone, the screen face angled downward and only partly visible to the camera, its glow leaking between her fingers. Flat, sickly greenish fluorescent overhead light washes her upper body, while the faint cold light from the phone reflects on her fingers and the lower side of her face, emphasizing her tension with cool damp shadows. Focus on her anxious expression. [: the same room, soft background depth, a duty desk blurred into horizontal lines, the printed date face of a wall calendar visible but completely defocused on the rear wall].",
            "outfit_assignments": [
              {
                "character_id": "C02",
                "outlook_id": "O04"
              }
            ],
            "source_facts": [
              "Scene text: '수리영에게 내색은 안했지만 왠지 불길한.'",
              "Scene text: '이상해, 예감이 안 좋아.'",
              "Shot 5 description: '휴대폰을 꽉 쥔 채 허공을 노려보는 혜수의 불안한 표정.'"
            ],
            "visual_inferences": [
              "fixed_element night_shift_office_environment: The interior of an institutional office at night, illuminated by flat, sickly greenish fluorescent overhead lights.",
              "The light from the phone screen creates a contrasting cold glow on her hands and lower face against the ambient green of the office."
            ],
            "creative_decisions": [
              "Used a head-and-shoulder close framing to intensely capture her anxiety while avoiding the 'floating head' artifact by keeping her upper body anchored in the frame.",
              "Applied a muted institutional blur to the background duty desk and calendar to maintain the zoom_in_detail continuity without drawing focus from her expression."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "95a23f92c3761b2b",
              "owned_hash": "f747b52ba12b45e1",
              "camera_direction_hash": "0ed96a841501adc9",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 14,
        "dependency_reason": "zoom_in_detail ref_usage for continuity with the prior shot, focusing tightly on C02 while preserving the same room and lighting conditions",
        "_shot_index": 5,
        "visible_entities": [
          "C02",
          "P01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 14,
            "shot_index": 5
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A CU at eye-level, pushing in to about two feet from 혜수’s face as she freezes with anxiety, staring past the camera into empty office space rather than at anyone. Her face sits slightly left of center, the clenched phone remains visible near the lower frame edge, and the duty room behind her falls soft into muted institutional blur.",
            "lighting_mood": "Sickly fluorescent light and the faint phone glow carve worry into her face, with cool shadows and no warm relief.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C02",
              "P01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C02",
                "outlook_id": "O04"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L10B02",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "night_shift_office_environment",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The interior of an institutional office at night, illuminated by flat, sickly greenish fluorescent overhead lights. The harsh lighting casts cool, damp shadows across the room, establishing an exhausted and tense atmosphere. A faint, cold glow from a mobile device slightly pierces the ambient greenish light, while the background features muted, indistinct office furniture and duty desks.",
                "applies_to_shots": [
                  3,
                  5
                ],
                "source_facts": [
                  "Scene text: '경찰서 사무실 / N'",
                  "Shot 3 description: 'Flat greenish fluorescent office light is broken by the cold phone glow, giving the confrontation a damp, exhausted night-shift tension'",
                  "Shot 5 description: 'Sickly fluorescent light and the faint phone glow'"
                ],
                "visual_inferences": [
                  "Standard institutional office setting with overhead panel lights",
                  "The overall color palette is dominated by cold green and bluish tones due to the fluorescent fixtures"
                ],
                "creative_decisions": [
                  "Using the greenish tint to visually reinforce the fatigue and anxiety of the late-night shift",
                  "Keeping the background details relatively generic and muted to keep the focus on the foreground subjects"
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 14,
                "shot_index": 3,
                "ref_usage": "zoom_in_detail",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C02O04",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "d2920836d7e779d5"
      },
      {
        "scene_index": 15,
        "heading": "S#17. 옥탑방 / 동틀무렵",
        "beat_title": "고요한 새벽의 옥탑방",
        "representative_moment": "어스름한 새벽빛 아래 불이 켜진 옥탑방 창문이 보이는 텅 빈 마당 전경.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Extreme wide shot, low angle waist height, pre-dawn blue haze with contrasting yellow practical glow",
            "t2i_prompt": "Photorealistic cinematic still. An extreme wide shot, low angle at waist height. The damp courtyard from the reference occupies the lower frame as empty negative space under a pre-dawn blue haze. The exterior stairs from the reference ascend on the right side. Focus on the lit rooftop window from the reference in the upper third, burning with a lonely yellow practical glow edged by humid fog.",
            "outfit_assignments": [],
            "source_facts": [
              "S#17 action: '새벽녘, 고양이 울음소리. 검은 고양이가 불이 켜져 있는 옥탑 방을 슬쩍 보더니 담을 넘는다'",
              "shot description: 어스름한 새벽빛 아래 불이 켜진 옥탑방 창문이 보이는 텅 빈 마당 전경."
            ],
            "visual_inferences": [
              "No human characters are present in the frame",
              "The pre-dawn blue haze and wet surfaces emphasize the isolation of the location"
            ],
            "creative_decisions": [
              "Extreme wide shot from a low angle to highlight the vast empty courtyard space in the foreground",
              "Contrasting the lonely yellow practical light from the window against the cool blue dawn ambient to create an omen-like mood"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "097ddece941615c4",
              "owned_hash": "9788465baca39ead",
              "camera_direction_hash": "659983f6edbfa365",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "courtyard",
                  "violating_phrase": "The damp courtyard from the reference",
                  "reason": "The courtyard is explicitly referenced as existing in the reference, which anchors it rather than requesting a redraw.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "window",
                  "violating_phrase": "the lit rooftop window from the reference",
                  "reason": "The window is explicitly described as coming from the reference, so this is an anchored existing object rather than a new one to create.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "stair",
                  "violating_phrase": "The exterior stairs from the reference",
                  "reason": "The stairs are tied to the reference and described as an existing background element, not as something to add or build.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 4,
        "dependency_reason": "exact_background preservation from prior shot established room",
        "_shot_index": 1,
        "visible_entities": [
          "L03"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 15,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "An EWS, low angle, held static from the far edge of the empty courtyard at waist height, facing the rooftop room across the yard. The courtyard occupies most of the lower frame as damp negative space, while the lit rooftop window sits small but sharp in the upper third, glowing before any cat is revealed.",
            "lighting_mood": "Pre-dawn blue haze mutes the courtyard while the rooftop window burns with a lonely yellow practical glow, edged by humid fog.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "L03"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L03B01",
            "reference_usage": "exact_background",
            "owned_objects": [
              "cat",
              "courtyard",
              "drain",
              "gate",
              "landing",
              "light fixture",
              "railing",
              "roof",
              "stair",
              "utility pipe",
              "wall",
              "window"
            ],
            "camera_reference": {
              "camera_position": "near number 4 (south entrance gate), inside number 1 (courtyard yard), facing diagonally upward/north toward number 5 (lit rooftop window), with number 2 (exterior stair run and base landing) along the right side and number 7 (black cat on wall) crossing the wall line",
              "camera_height": "eye-level standing ~1.6m",
              "lens_hint": "35mm wide angle",
              "framing_notes": "Emphasize the empty dawn courtyard mood by keeping the yard open in the foreground; include the stair base, the rooftop window, and the cat on the wall as the key readable anchors."
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "when focusing on an owned object, anchor it explicitly: 'the door from the reference'",
              "camera_reference specifies the framing under which the background was painted — match camera_position, camera_height, and framing",
              "if deviating from the reference framing (push-in, angle shift, ECU), state the deviation explicitly so the compositor can account for the scale difference"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [
              {
                "scene_index": 4,
                "shot_index": 1,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "background",
                "id": "L03B01",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "375fbfc0ce2db525"
      },
      {
        "scene_index": 16,
        "heading": "S#18. 안방 안 / D",
        "beat_title": "환각의 잔상과 안도",
        "representative_moment": "번쩍 눈을 뜬 C01이 베개에 머리를 댄 채 허공의 천장을 멍하니 응시하는 순간.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "ECU, slight low angle from bed level, cool dawn lighting",
            "t2i_prompt": "Photorealistic cinematic still. C01O02 in a beige hoodie, an East Asian young woman in her early 20s with black hair and dark eyes, tight framing on her face, neck and upper chest visible in the lower frame. She is propped against a creased, slept-on pillow at three-quarter angle from the right, her eyes fixed on the blank, pale ceiling above. A slight low angle from bed level beside her, aimed upward. Focus on her startled, glossy expression. Cold, dim, and muted grayish daylight characteristic of early dawn washes across her features. A thin curtain filters soft, cool-toned ambient light from the off-frame side, casting faint shadows on the pillow edge. The immediate background consists only of a sliver of the dimly lit ceiling, establishing an oppressive stillness without wide room details.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "Shot description: observer at bed level, ECU of 수리영's face, staring at ceiling.",
              "DP directives: awakened eyes, blank ceiling, pillow edge.",
              "S16 action: 고양이 울음소리에 번쩍 눈을 뜨고."
            ],
            "visual_inferences": [
              "fixed_element dim_dawn_lighting: cold, dim, and muted grayish daylight washing over the scene.",
              "Low angle emphasizes her fixed gaze and vulnerable state.",
              "Close framing omits the wider room context (skipped background ref) to focus purely on her immediate bed-level surroundings."
            ],
            "creative_decisions": [
              "Kept neck and upper chest in frame to prevent floating head artifacts, adhering to close-framing face phrasing rules.",
              "Used 'Focus on her startled, glossy expression' to direct emphasis without triggering body-part ID restrictions.",
              "Incorporated stable traits (black hair, dark eyes) naturally into the demographic descriptor."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "3a0c16cb72b3335d",
              "owned_hash": "6b550e5aab414324",
              "camera_direction_hash": "94de47cb1f3dd327",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 12,
        "dependency_reason": "previous_shot_refs에 명시된 S12_Shot8의 배경(방) 연속성 데이터를 기반으로 하나, 현재 샷의 close framing 설정으로 인해 배경 묘사는 최소화됨.",
        "_shot_index": 2,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 16,
            "shot_index": 2
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "An ECU, slight low angle, from bed level beside 수리영’s face, aimed upward enough to catch her fixed gaze toward the blank ceiling. Only her awakened eyes, brow, and upper face dominate the frame, with the pillow edge and a sliver of dim ceiling giving scale; the camera is static, intimate, and breath-held.",
            "lighting_mood": "Cold dawn seepage and faint room shadow leave her eyes glossy and startled, with no hallucination yet—only oppressive stillness.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B09",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "dim_dawn_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The indoor environment is bathed in a cold, dim, and muted grayish daylight characteristic of early dawn. Soft, cool-toned ambient light seeps into the spaces, casting faint shadows and establishing a consistently gloomy, low-contrast atmosphere across different rooms.",
                "applies_to_shots": [
                  2,
                  10,
                  14
                ],
                "source_facts": [
                  "Shot 2 lighting: 'Cold dawn seepage and faint room shadow'",
                  "Shot 10 lighting: 'cool humid daylight'",
                  "Shot 14 lighting: 'Muted gray daylight'"
                ],
                "visual_inferences": [
                  "The entire house shares the same overcast, early morning natural light coming from outside.",
                  "The lack of strong direct sunlight unifies the physically disconnected rooms (bedroom, bathroom, dining area) under one cohesive tone."
                ],
                "creative_decisions": [
                  "Applying a consistent cool-blue and grayish color grade to the environmental lighting to heighten the protagonist's psychological dread and the eerie occult undertones."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 12,
                "shot_index": 8,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "e535d09b027e9bda"
      },
      {
        "scene_index": 16,
        "heading": "S#18. 욕실 안 / D",
        "beat_title": "환각으로 인한 극심한 공포와 자기 불신",
        "representative_moment": "거울 표면에 반사된 욕실 타일 위로 붉은 원형 표식(P08)이 선명하게 나타난다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Tight eye-level close-up reflection shot, cool grayish dawn light mixed with dim practical lighting",
            "t2i_prompt": "Photorealistic cinematic still. Tight eye-level close-up framing directly against the [L04: clean but slightly humid bathroom mirror plane], capturing an uncanny, flat reflection. Focus on P08, a sharp and saturated red painted circle, centered slightly above the middle third of the reflected tile wall. The reflected background consists of pale, cold, wet-looking tiles forming a near-graphic grout grid. Dim practical lighting mixes with cool, grayish early dawn ambient light, making the crisp red mark appear aggressively vivid against the fog-softened white tiles. The reflection is completely devoid of any figures or human presence.",
            "outfit_assignments": [],
            "source_facts": [
              "S#18 action: 거울에 비친 욕실 벽에 갑자기 또 발생하는 붉은 원.",
              "Shot description: A static shot of a sharp red painted circle on the bathroom tiles, reflected on the mirror surface."
            ],
            "visual_inferences": [
              "fixed_element dim_dawn_lighting: cool, dim grayish daylight seeps into the space.",
              "Suriyoung is completely excluded from the reflection frame to emphasize the subjective hallucination."
            ],
            "creative_decisions": [
              "Tight close-up framing strictly on the mirror surface to flatten the perspective into a graphic, claustrophobic grid.",
              "High color contrast between the saturated red of P08 and the pale, cool tiles to heighten psychological dread."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "0e72968d11406085",
              "owned_hash": "cdfa7376ed4f4a07",
              "camera_direction_hash": "1c02a1e42137e393",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 12,
        "dependency_reason": "atmosphere_reference for cool gray ambient lighting tone",
        "_shot_index": 10,
        "visible_entities": [
          "P08",
          "L04"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 16,
            "shot_index": 10
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "From tight against the bathroom mirror plane, the camera holds a static CU, eye-level reflection shot; the reflected bathroom tiles flatten into a near-graphic grid, with the vivid red painted circle centered slightly above the middle third. This is 수리영’s subjective hallucinated perception, but she is excluded from frame, leaving only the mirror-surface image and the impossible mark.",
            "lighting_mood": "Dim bathroom practical light mixes with cool humid daylight, making the red circle feel aggressively vivid against fog-softened tile whites.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "P08",
              "L04"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B10",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "dim_dawn_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The indoor environment is bathed in a cold, dim, and muted grayish daylight characteristic of early dawn. Soft, cool-toned ambient light seeps into the spaces, casting faint shadows and establishing a consistently gloomy, low-contrast atmosphere across different rooms.",
                "applies_to_shots": [
                  2,
                  10,
                  14
                ],
                "source_facts": [
                  "Shot 2 lighting: 'Cold dawn seepage and faint room shadow'",
                  "Shot 10 lighting: 'cool humid daylight'",
                  "Shot 14 lighting: 'Muted gray daylight'"
                ],
                "visual_inferences": [
                  "The entire house shares the same overcast, early morning natural light coming from outside.",
                  "The lack of strong direct sunlight unifies the physically disconnected rooms (bedroom, bathroom, dining area) under one cohesive tone."
                ],
                "creative_decisions": [
                  "Applying a consistent cool-blue and grayish color grade to the environmental lighting to heighten the protagonist's psychological dread and the eerie occult undertones."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 12,
                "shot_index": 8,
                "ref_usage": "atmosphere_reference",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "cebb401104df2bf3"
      },
      {
        "scene_index": 16,
        "heading": "S#18. 거실 식탁 / D",
        "beat_title": "사진 속 엄마의 얼굴이 기괴하게 일그러지는 환각",
        "representative_moment": "식탁 위에 놓인 P03 사진 속 중년 여자의 웃는 얼굴이 기괴하게 흘러내리듯 왜곡된 정지 클로즈업 컷.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Close-up, slight Dutch angle at table height, cold dim dawn lighting with a cool-blue psychological dread color grade.",
            "t2i_prompt": "Photorealistic cinematic still. Close-up at table height with a slight Dutch angle, focusing on P03, a framed photograph standing upright on a [: dark wood dining table with a humid sheen]. Inside the physical frame, the reproduced image shows a middle-aged East Asian woman facing the camera; her smiling portrait occupies the center third but is grotesquely warped as if the ink is melting downward in a sickening hallucination. The frame's fingerprinted, dull glass reflects a faint streak of dim practical light across the distorted smile. A half-full water glass, wet with condensation, sits softly out of focus at the lower frame edge. The immediate tabletop is bathed in cold, muted grayish dawn daylight with a cool-blue ambient tone, leaving faint shadows and establishing a gloomy, eerie atmosphere.",
            "outfit_assignments": [],
            "source_facts": [
              "Shot 14 description: At dining-table height, the completed dolly-in locks into a CU, slight Dutch angle on the framed photograph surface; the mother’s smiling face sits in the center third, grotesquely warped as if the ink is melting downward.",
              "Scene text: 식탁에 놓인 엄마와 수리영의 사진을 보는데... 환하게 웃는 엄마 얼굴이 뭉개져 흘러내린다",
              "DP direction: water glass half-full, wet with droplets as a scale reference."
            ],
            "visual_inferences": [
              "fixed_element dim_dawn_lighting: The indoor environment is bathed in a cold, dim, and muted grayish daylight characteristic of early dawn.",
              "The physical P03 prop is geometrically grounded by the visible dark wood table surface and the blurred water glass at the frame edge.",
              "The subjective hallucination only affects the portrait face inside the reproduction surface, not the physical frame or table."
            ],
            "creative_decisions": [
              "Applied a cool-blue and grayish color grade to the environmental lighting to heighten psychological dread.",
              "Used a slight Dutch angle to emphasize the sickening, unstable nature of the hallucination.",
              "Skipped the background room entirely as per close framing constraints, focusing solely on the immediate tabletop surroundings."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "240e0ee3f18de31c",
              "owned_hash": "cdfa7376ed4f4a07",
              "camera_direction_hash": "bbd88148db5584f1",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 12,
        "dependency_reason": "previous_shot_refs(zoom_in_detail) points to Scene 12 Shot 8 to maintain the same dining table surface and lighting context.",
        "_shot_index": 14,
        "visible_entities": [
          "P03"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 16,
            "shot_index": 14
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "At dining-table height, the completed dolly-in locks into a CU, slight Dutch angle on the framed photograph surface; the mother’s smiling face sits in the center third, grotesquely warped as if the ink is melting downward. The frame edges and a sliver of tabletop remain visible to keep it physically grounded, while the distortion is perceived through 수리영’s hallucination.",
            "lighting_mood": "Muted gray daylight and weak warm practicals leave the table dim, with the photo’s warped face carrying the shot’s sickly occult emphasis.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "P03"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B10",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "dim_dawn_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The indoor environment is bathed in a cold, dim, and muted grayish daylight characteristic of early dawn. Soft, cool-toned ambient light seeps into the spaces, casting faint shadows and establishing a consistently gloomy, low-contrast atmosphere across different rooms.",
                "applies_to_shots": [
                  2,
                  10,
                  14
                ],
                "source_facts": [
                  "Shot 2 lighting: 'Cold dawn seepage and faint room shadow'",
                  "Shot 10 lighting: 'cool humid daylight'",
                  "Shot 14 lighting: 'Muted gray daylight'"
                ],
                "visual_inferences": [
                  "The entire house shares the same overcast, early morning natural light coming from outside.",
                  "The lack of strong direct sunlight unifies the physically disconnected rooms (bedroom, bathroom, dining area) under one cohesive tone."
                ],
                "creative_decisions": [
                  "Applying a consistent cool-blue and grayish color grade to the environmental lighting to heighten the protagonist's psychological dread and the eerie occult undertones."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 12,
                "shot_index": 8,
                "ref_usage": "zoom_in_detail",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "prop",
                "id": "P03",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [
            {
              "contract_id": "rc_001",
              "scope": {
                "scene_index": 16,
                "shot_indices": [
                  14
                ],
                "duration": "single_shot"
              },
              "targets": [
                {
                  "entity_id": "P03",
                  "role": "visual_target"
                }
              ],
              "requirements": [
                {
                  "dimension": "visual_identity",
                  "operation": "preserve",
                  "strength": "required",
                  "reference_policy": "use_entity_reference"
                }
              ]
            }
          ],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "3805982061531e6f"
      },
      {
        "scene_index": 17,
        "heading": "S#19. 마트 외부 / 이른 아침",
        "beat_title": "마트 입구에 도착한 수리영",
        "representative_moment": "안개가 옅게 낀 이른 아침, 마트 입구 유리문 앞에 자전거 핸들을 쥔 채 선 수리영(C01)의 전신.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Eye-level medium-wide establishing frame, 35mm wide angle lens, cool fog-muted dawn lighting.",
            "t2i_prompt": "Photorealistic cinematic still. Eye-level observer frame, the subject's shoes and lower legs occupying the lower frame with her upper body extending toward the upper edge. C01O02 in a beige hoodie, an East Asian young woman in her early 20s with black hair and dark eyes, stands in the center foreground. She is positioned at a three-quarter angle from the left, gripping the damp handlebar of her bicycle. The bicycle's frame, brake cable, and front tire are clearly visible in the lower frame beside her legs. Her eyes are fixed on the closed glass doors from the reference. The scene is bathed in cold, fog-muted dawn light, a visible layer of pale gray morning fog softening the quiet storefront from the reference behind her. Weak fluorescent spill from inside the mart washes the moisture-streaked windows from the reference. The rain-slick sidewalk from the reference fills the ground, reflecting the glass doors and bicycle tires in broken, dim patches. Focus on her paused, full-body stance and the bicycle, establishing the quiet morning arrival.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "Shot description: 안개가 옅게 낀 이른 아침, 마트 입구 유리문 앞에 자전거 핸들을 쥔 채 선 수리영의 전신.",
              "DP notes: Outside the mart entrance, eye-level establishing frame about 18 feet from the glass doors. 수리영’s full body and bicycle occupy the center foreground."
            ],
            "visual_inferences": [
              "fixed_element cold_dawn_lighting_state: Bathed in dim, cool-toned daylight with a visible layer of faint morning fog.",
              "수리영 holds the bicycle's handlebar in a stationary pose before entering, requiring the bicycle body to be visible according to prop rules.",
              "The reflection on the pavement is dictated directly by DP notes and wet weather conditions."
            ],
            "creative_decisions": [
              "Used a wide 35mm establishing framing to capture the character's full body, the bicycle, and the closed mart storefront in one coherent spatial layout.",
              "Emphasized the pale gray morning fog and reflective wet sidewalk to enhance the damp coastal thriller mood.",
              "Placed the camera at an eye-level standing height with geometric frame-edge alignment (feet at bottom, head at top) for realistic proportions."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "e74ffc71c8bb2bf9",
              "owned_hash": "32ecec4c97c04d6d",
              "camera_direction_hash": "3de681063f8513d2",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "bicycle",
                  "violating_phrase": "gripping the damp handlebar of her bicycle",
                  "reason": "The bicycle is mentioned as an existing object being held and framed in the scene, with no explicit redraw or add directive directed at it.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "door",
                  "violating_phrase": "closed glass doors from the reference",
                  "reason": "The doors are explicitly tied to 'from the reference,' which anchors them as pre-existing background elements rather than newly created ones.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "storefront",
                  "violating_phrase": "quiet storefront from the reference",
                  "reason": "The storefront is directly marked as coming from the reference, so this is an anchor reference rather than a redraw instruction.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "window",
                  "violating_phrase": "moisture-streaked windows from the reference",
                  "reason": "The windows are described with a reference anchor and no verb that explicitly creates or modifies them as new background objects.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "sidewalk",
                  "violating_phrase": "rain-slick sidewalk from the reference",
                  "reason": "The sidewalk is referenced as part of the existing scene from the reference, which makes it an anchor reference.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "N/A",
        "_shot_index": 1,
        "visible_entities": [
          "C01",
          "L07"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 17,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Outside the mart entrance, the camera holds a WS, eye-level establishing frame about 18 feet from the glass doors before the dolly-in begins; 수리영’s full body and bicycle occupy the center foreground, with the handlebar readable across her waist line. The quiet storefront sits symmetrically behind her through faint morning fog, emphasizing the pause before she enters.",
            "lighting_mood": "Cool fog-muted dawn light dominates, with weak fluorescent spill from inside the mart and no comforting warmth.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01",
              "L07"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L07B01",
            "reference_usage": "exact_background",
            "owned_objects": [
              "CCTV camera",
              "awning",
              "bicycle",
              "bicycle rack",
              "bollard",
              "curb",
              "door",
              "entrance mat",
              "sidewalk",
              "storefront",
              "street lane",
              "window"
            ],
            "camera_reference": {
              "camera_position": "near number 9 (parked bicycle) beside number 8 (bicycle rack), facing diagonally toward number 4 (glass double doors) with number 5 and number 6 (front windows) across the facade",
              "camera_height": "eye-level standing ~1.6m",
              "lens_hint": "35mm wide angle",
              "framing_notes": "include the bicycle arrival area in the foreground and the closed mart entrance beyond it"
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "when focusing on an owned object, anchor it explicitly: 'the door from the reference'",
              "camera_reference specifies the framing under which the background was painted — match camera_position, camera_height, and framing",
              "if deviating from the reference framing (push-in, angle shift, ECU), state the deviation explicitly so the compositor can account for the scale difference"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "cold_dawn_lighting_state",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A specific lighting state of the outdoor environment, featuring cold, fog-muted dawn light. The scene is bathed in dim, cool-toned daylight with a visible layer of faint morning fog that softens edges and mutes surrounding colors, creating a chilly and quiet atmosphere.",
                "applies_to_shots": [
                  1,
                  5
                ],
                "source_facts": [
                  "scene heading text: 이른 아침 (early morning)",
                  "Shot 1 lighting: Cool fog-muted dawn light dominates",
                  "Shot 5 lighting: Cold dawn light"
                ],
                "visual_inferences": [
                  "The light casts a bluish-gray tint over the exterior, lacking any warm sunlight."
                ],
                "creative_decisions": [
                  "Unifying the color grading and atmospheric fog texture to ensure the early morning mood remains consistent across wide and tight shots."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              },
              {
                "kind": "background",
                "id": "L07B01",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "ca6ce1f48e0975a1"
      },
      {
        "scene_index": 17,
        "heading": "S#19. 마트 / D",
        "beat_title": "CCTV를 통해 정체를 확인하기로 결심하다",
        "representative_moment": "C01이 굳은 표정으로 마트 밖 CCTV 렌즈를 똑바로 올려다보고 있다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "High angle surveillance-style tight CU, cold fog-muted dawn light with washed-out contrast and faint red indicator reflection.",
            "t2i_prompt": "Photorealistic cinematic still. High angle surveillance-style composition. C01O02 in a beige hoodie, a young Korean woman in her early 20s with black hair and dark eyes, facing the camera, chin lifted tense, eyes fixed on the camera lens. Head-and-shoulder composition, tight framing on her face, upper body and shoulders in frame. Focus on her tense expression staring into the lens. Cold fog-muted dawn light casts a dim bluish-gray tint over her features, creating a washed-out contrast. A faint red indicator light reflects on her cheek and forehead. In the frame-edge background, [: damp, stained, pale concrete exterior wall] is visible at a shallow angle from above, slightly blurred. The weathered side edge of a CCTV housing is partially visible near the top edge. Below her chest at the bottom frame edge, the rain-darkened wet entrance pavement faintly reflects pale storefront light.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "scene heading text: 이른 아침 (early morning)",
              "shot description: 수리영이 마트 밖에서 CCTV 위치를 확인하고 렌즈를 향해 턱을 치켜든 정면 구도"
            ],
            "visual_inferences": [
              "fixed_element cold_dawn_lighting_state: dim, cool-toned daylight with a visible layer of faint morning fog, casting a bluish-gray tint over the exterior",
              "character pose: facing the camera, chin lifted tense, eyes fixed on the camera lens",
              "surveillance-style high angle composition implies the physical position of the exterior CCTV"
            ],
            "creative_decisions": [
              "Used a head-and-shoulder composition to prevent the face from floating disconnected, keeping the upper body anchored in the frame per close_framing_face_phrasing constraints",
              "Skipped full background reference and composed the frame-edge surroundings directly with a damp concrete wall and wet pavement reflection to emphasize the tight, isolated focus on the subject",
              "Added a faint red indicator light reflecting on her cheek to visually enhance the feeling of staring into a recording device"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "a571dc68e608e2f0",
              "owned_hash": "9c311eb384197b50",
              "camera_direction_hash": "da6841d00ebbd309",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 17,
        "dependency_reason": "atmosphere_reference: shares the same cold dawn lighting state and exterior mood as the preceding establishing shots of Scene 17.",
        "_shot_index": 5,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 17,
            "shot_index": 5
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Mounted high beside the exterior CCTV axis, the camera holds a tight CU, high angle, surveillance-style frontal frame as 수리영 lifts her chin directly toward the lens. Her face occupies most of the center, while the wall and CCTV hardware are reduced to thin blurred edges, making the confrontation feel like she is staring into recorded evidence itself.",
            "lighting_mood": "Cold dawn light and washed-out CCTV-like contrast flatten her features, with a faint red indicator glow near the camera side adding menace.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L07B02",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "cold_dawn_lighting_state",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A specific lighting state of the outdoor environment, featuring cold, fog-muted dawn light. The scene is bathed in dim, cool-toned daylight with a visible layer of faint morning fog that softens edges and mutes surrounding colors, creating a chilly and quiet atmosphere.",
                "applies_to_shots": [
                  1,
                  5
                ],
                "source_facts": [
                  "scene heading text: 이른 아침 (early morning)",
                  "Shot 1 lighting: Cool fog-muted dawn light dominates",
                  "Shot 5 lighting: Cold dawn light"
                ],
                "visual_inferences": [
                  "The light casts a bluish-gray tint over the exterior, lacking any warm sunlight."
                ],
                "creative_decisions": [
                  "Unifying the color grading and atmospheric fog texture to ensure the early morning mood remains consistent across wide and tight shots."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 17,
                "shot_index": 1,
                "ref_usage": "atmosphere_reference",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "8a9103118c9fb461"
      },
      {
        "scene_index": 18,
        "heading": "S#20. 마트 총무과 / 실내",
        "beat_title": "어두운 CCTV 화면과 수리영의 불안",
        "representative_moment": "혜수 어깨 너머로 초조하게 모니터를 향해 목을 빼고 있는 수리영의 클로즈업.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Over-the-shoulder close-up, eye-level framing, shallow depth of field, cool blue-gray lighting contrast",
            "t2i_prompt": "Photorealistic cinematic still. Head-and-shoulder composition, eye-level framing. An out-of-focus shoulder and back of a Korean woman facing away from the camera forms a soft dark foreground edge on the left. Just behind her, C01O02 in a beige hoodie, a Korean woman in her early 20s with black hair and dark eyes, occupies the center-right at three-quarter angle from the right. She leans forward with her neck craned, sharing a single line of sight toward an unseen monitor off-frame. Focus on the psychological tension of the moment. Cold blue-gray monitor light washes across the right side of her face and reflects in her eyes, leaving the immediate foreground in heavy, dim shadows.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "Scene text: 수리영 불안이 깊어지고. CCTV화면을 돌려보는 보안팀 직원과 혜수, 뒤에서 초조한 모습으로 화면을 주시하는 수리영",
              "Shot description: Just behind and slightly beside 혜수, the camera tracks into an OTS CU, eye-level frame; 혜수’s shoulder becomes a soft dark foreground edge on the left while 수리영’s anxious face fills the center-right."
            ],
            "visual_inferences": [
              "fixed_element desk_cctv_monitor_dark_playback: The unseen CCTV monitor acts as the focal point for the characters' single line of sight.",
              "fixed_element cold_cctv_monitor_office_lighting: Blue-gray monitor light is the dominant light source illuminating the subjects in the dim setting.",
              "Hye-soo is rendered as an unregistered foreground common-noun entity to form the OTS framing without triggering a composite ID."
            ],
            "creative_decisions": [
              "Utilized close head-and-shoulder composition to emphasize C01's anxious expression while skipping unnecessary background elements.",
              "Applied cool blue-gray light washing across the face rather than cutting it, ensuring realistic lighting without generating physical cuts on the character."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "7959247666851c9d",
              "owned_hash": "93ac7ff8cd40c3b5",
              "camera_direction_hash": "8f130c24a43289f4",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "Not directly dependent on a prior scene's generated background due to close framing skipping background refs; relies solely on fixed lighting and prop states.",
        "_shot_index": 2,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 18,
            "shot_index": 2
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Just behind and slightly beside 혜수, the camera tracks into an OTS CU, eye-level frame; 혜수’s shoulder becomes a soft dark foreground edge on the left while 수리영’s anxious face fills the center-right. 수리영 leans past 혜수 toward the unseen monitor, her neck stretched forward and eyes fixed off-frame.",
            "lighting_mood": "Blue-gray monitor light cuts across humid, stale office darkness, leaving 혜수 as a soft silhouette and 수리영’s eyes tense and wet.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L11B02",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "applies_to_shots": [
                  2,
                  4
                ],
                "character_name": "",
                "confidence": "high",
                "creative_decisions": [
                  "Keep the blue-gray CCTV monitor glow as the dominant visual continuity cue across both selected shots.",
                  "Maintain a cold investigative atmosphere in the office rather than introducing warm ambient light."
                ],
                "description": "The supermarket administrative security office remains dim, with cold blue-gray light from a CCTV monitor washing across the people near the desk. Any overhead fluorescent light stays dull and dirty, mixing weakly with the monitor spill and leaving the room in muted, cool shadows.",
                "element_id": "cold_cctv_monitor_office_lighting",
                "element_type": "environment_state",
                "source_facts": [
                  "Scene text: CCTV footage is being replayed in the supermarket administrative office/security room.",
                  "Shot 2 lighting: 'Blue-gray monitor light cuts across humid, stale office darkness.'",
                  "Shot 4 lighting: 'Dirty fluorescent overheads mix with blue monitor spill.'"
                ],
                "visual_inferences": [
                  "The same cold monitor spill is treated as a continuous lighting condition across the selected shots.",
                  "The office darkness is kept consistent even when the camera angle changes."
                ]
              },
              {
                "applies_to_shots": [
                  2,
                  4
                ],
                "character_name": "",
                "confidence": "high",
                "creative_decisions": [
                  "Use the CCTV monitor as the central fixed prop motivating eyelines and blue light, even when it is partly off-frame.",
                  "Keep the dark surveillance segment visually unresolved so the scene’s investigative tension remains consistent."
                ],
                "description": "A CCTV monitor at the desk is replaying surveillance footage in which one crucial section appears abnormally dark and difficult to identify. The screen faces the people gathered around the desk and acts as the source of the cold blue-gray glow in the room.",
                "element_id": "desk_cctv_monitor_dark_playback",
                "element_type": "persistent_prop",
                "source_facts": [
                  "Scene text: the security staff and others are reviewing CCTV footage.",
                  "Scene text: '그 부분만 유독 어두워 식별이 불가능하다' — that portion is unusually dark and impossible to identify.",
                  "Shot 2 staging: eyes are fixed toward the CCTV monitor, with blue-gray monitor light present.",
                  "Shot 4 lighting: blue monitor spill is present during the exchange."
                ],
                "visual_inferences": [
                  "The CCTV monitor is placed on or immediately beside the desk in front of the characters.",
                  "The dark section remains on the same playback source across both shots."
                ]
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "a985107b56b9e75a"
      },
      {
        "scene_index": 18,
        "heading": "S#20. 마트 총무과 / 실내",
        "beat_title": "식별 불가한 영상과 복사 요청",
        "representative_moment": "혜수가 턱을 괸 채 보안직원의 옆얼굴을 주시하며 CCTV 복사를 요청한다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "MCU from a slight low angle, mixing dim dirty fluorescent overheads with a cold blue-gray monitor spill.",
            "t2i_prompt": "Photorealistic cinematic still. MCU from a slight low angle. C02O05 in a charcoal gray suit jacket, a Korean woman in her 40s, occupies the left-center frame in right profile, upper body visible, her hand braced under her chin, eyes fixed on the figure across from her. In the right midground, an East Asian man in a security uniform is partially visible in left profile, leaning toward the desk, his eyes fixed on C02O05. The two figures are separated by the front edge of a security desk cutting across the lower frame. Dirty fluorescent overhead light mixes with cold blue-gray monitor spill that washes across C02O05's focused features. [L11: Close framing of the immediate office space; a messy security notice board with curling papers is visible behind the man. A CCTV monitor on the desk angles partly away from the camera, displaying a dark, muddy video section. A weak yellow desk lamp points downward at the desk surface.] Focus on C02O05's sharp right profile and the cold investigative tension in the space between them.",
            "outfit_assignments": [
              {
                "character_id": "C02",
                "outlook_id": "O05"
              }
            ],
            "source_facts": [
              "Scene text: CCTV footage is being replayed in the supermarket administrative office/security room.",
              "Scene text: '그 부분만 유독 어두워 식별이 불가능하다' — that portion is unusually dark and impossible to identify.",
              "Shot 4 lighting: 'Dirty fluorescent overheads mix with blue monitor spill.'"
            ],
            "visual_inferences": [
              "fixed_element cold_cctv_monitor_office_lighting: The same cold monitor spill is treated as a continuous lighting condition.",
              "fixed_element desk_cctv_monitor_dark_playback: The CCTV monitor is placed on or immediately beside the desk in front of the characters.",
              "The security guard is placed in the right midground opposite Hye-su, creating a direct eyeline exchange across the desk."
            ],
            "creative_decisions": [
              "Used a slight low angle to emphasize Hye-su's scrutinizing, calculating posture.",
              "Positioned the desk's front edge in the lower frame to act as a shared spatial anchor and physical divider.",
              "Applied the blue-gray monitor spill directly onto Hye-su's face to visually tie her to the unseen off-screen investigation target."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "56a040c57e9b4f7f",
              "owned_hash": "93ac7ff8cd40c3b5",
              "camera_direction_hash": "252dad1f347820eb",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "No previous scene has established this exact angle and close framing; scene continuity is handled internally via fixed elements from earlier shots within the same scene.",
        "_shot_index": 4,
        "visible_entities": [
          "C02",
          "L11"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 18,
            "shot_index": 4
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "The lateral move settles beside the desk into a steady MCU, slight low angle across 혜수 toward 보안직원’s side profile. 혜수 occupies the left-center with her hand braced under her chin, while 보안직원 remains in the right midground as her eyeline target, keeping the investigative pressure on their exchange.",
            "lighting_mood": "Dirty fluorescent overheads mix with blue monitor spill, giving 혜수’s focused profile a cold investigative hardness.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C02",
              "L11"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C02",
                "outlook_id": "O05"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L11B02",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "applies_to_shots": [
                  2,
                  4
                ],
                "character_name": "",
                "confidence": "high",
                "creative_decisions": [
                  "Keep the blue-gray CCTV monitor glow as the dominant visual continuity cue across both selected shots.",
                  "Maintain a cold investigative atmosphere in the office rather than introducing warm ambient light."
                ],
                "description": "The supermarket administrative security office remains dim, with cold blue-gray light from a CCTV monitor washing across the people near the desk. Any overhead fluorescent light stays dull and dirty, mixing weakly with the monitor spill and leaving the room in muted, cool shadows.",
                "element_id": "cold_cctv_monitor_office_lighting",
                "element_type": "environment_state",
                "source_facts": [
                  "Scene text: CCTV footage is being replayed in the supermarket administrative office/security room.",
                  "Shot 2 lighting: 'Blue-gray monitor light cuts across humid, stale office darkness.'",
                  "Shot 4 lighting: 'Dirty fluorescent overheads mix with blue monitor spill.'"
                ],
                "visual_inferences": [
                  "The same cold monitor spill is treated as a continuous lighting condition across the selected shots.",
                  "The office darkness is kept consistent even when the camera angle changes."
                ]
              },
              {
                "applies_to_shots": [
                  2,
                  4
                ],
                "character_name": "",
                "confidence": "high",
                "creative_decisions": [
                  "Use the CCTV monitor as the central fixed prop motivating eyelines and blue light, even when it is partly off-frame.",
                  "Keep the dark surveillance segment visually unresolved so the scene’s investigative tension remains consistent."
                ],
                "description": "A CCTV monitor at the desk is replaying surveillance footage in which one crucial section appears abnormally dark and difficult to identify. The screen faces the people gathered around the desk and acts as the source of the cold blue-gray glow in the room.",
                "element_id": "desk_cctv_monitor_dark_playback",
                "element_type": "persistent_prop",
                "source_facts": [
                  "Scene text: the security staff and others are reviewing CCTV footage.",
                  "Scene text: '그 부분만 유독 어두워 식별이 불가능하다' — that portion is unusually dark and impossible to identify.",
                  "Shot 2 staging: eyes are fixed toward the CCTV monitor, with blue-gray monitor light present.",
                  "Shot 4 lighting: blue monitor spill is present during the exchange."
                ],
                "visual_inferences": [
                  "The CCTV monitor is placed on or immediately beside the desk in front of the characters.",
                  "The dark section remains on the same playback source across both shots."
                ]
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C02O05",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "cfa0986c961123b6"
      },
      {
        "scene_index": 19,
        "heading": "S#21. 경찰서 안 / D",
        "beat_title": "충격적인 발견",
        "representative_moment": "수리영이 책상 위 살인사건 사진들을 보고 경악하는 얼굴 클로즈업.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Low angle CU with shallow DOF, harsh fluorescent lighting draining color.",
            "t2i_prompt": "Photorealistic cinematic still. [: Close framing at desk level, a blurred edge of a scratched gray laminate police desk cutting across the lower foreground, with an out-of-focus worn gray partition and a glowing monitor screen behind.] C01O02 in a beige hoodie, an East Asian woman in her early 20s, at three-quarter angle from the left, tight framing on her face with shoulders in frame. She is hunched close over the tabletop, dark eyes looking down toward the desk. Focus on her pale expression and dilated eyes frozen in shock. On the scratched desk surface in the lower foreground, several glossy crime scene photographs are chaotically scattered and overlapping, displaying gruesome dark red-brown torn flesh details, alongside dull silver paperclips, the glossy front faces lying upward beneath the shallow focus. The environment is illuminated by harsh, cold overhead fluorescent lighting casting a sterile, bluish-white glare that washes across her skin, draining her face pale while deepening the somber shadows around the desk.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "S19 action: '혜수 책상위에는 살인사건 확대 사진들이 어지럽게 널려있다. 넘겨보며 놀라는 수리영... 목, 어깨, 얼굴등 살점이 떨어져'",
              "Shot 4 description: 책상 위로 몸을 바싹 숙인 채 동공이 크게 확장된 수리영의 얼굴 클로즈업."
            ],
            "visual_inferences": [
              "fixed_element scattered_evidence_photos: glossy photographs with gruesome red-brown details",
              "fixed_element cold_fluorescent_lighting: harsh bluish-white glare and deep background gloom",
              "shallow focus isolates her expression while leaving the desk edge blurred"
            ],
            "creative_decisions": [
              "Using a low angle to place the desk surface in the lower foreground, incorporating the forward zoom targets (photos, scratched desk, paperclips)",
              "Applying skipped_close_framing background mode using [L01] block to describe immediate surroundings without referencing the whole room"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "f33571454342d770",
              "owned_hash": "704a968fbe0b5a95",
              "camera_direction_hash": "ac2555c3f6e685a3",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 14,
        "dependency_reason": "같은 방 안이지만 close framing으로 배경 참조 생략, 앞선 샷의 분위기 및 사진 소품 유지",
        "_shot_index": 4,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 19,
            "shot_index": 4
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "From the front corner of 혜수’s desk, the camera pushes into a CU, low angle with shallow DOF; the blurred desk edge cuts across the lower foreground while 수리영’s dilated eyes dominate the upper center. She is hunched close over the tabletop, frozen between recognition and panic as her gaze drops toward the evidence below.",
            "lighting_mood": "Harsh police-station fluorescents are softened by humid air and monitor spill, draining her face pale while the desk shadows feel heavy.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L10B01",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "scattered_evidence_photos",
                "element_type": "persistent_prop",
                "character_name": "",
                "description": "Several glossy crime scene photographs are chaotically scattered and overlapping on the surface of a standard office desk. The visible images display gruesome, torn flesh details with dark red-brown tones, contrasting sharply with the mundane desk surface.",
                "applies_to_shots": [
                  4,
                  5
                ],
                "source_facts": [
                  "scene text: '혜수 책상위에는 살인사건 확대 사진들이 어지럽게 널려있다. 목, 어깨, 얼굴등 살점이 떨어져'",
                  "Shot 4: 'her gaze drops toward the evidence below'",
                  "Shot 5: 'The images overlap chaotically across the desk, with torn neck and shoulder details isolated in the center'"
                ],
                "visual_inferences": [
                  "The photographs are directly resting on the desk under the character's immediate view, acting as a horrifying focal point among otherwise normal office items."
                ],
                "creative_decisions": [
                  "Highlighting the red-brown wound details within the photos to create a sickening, warm accent against the cold desk environment."
                ],
                "confidence": "high"
              },
              {
                "element_id": "cold_fluorescent_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The police station environment is illuminated by harsh, cold overhead fluorescent lights. This lighting casts a sterile, bluish-white glare across flat surfaces like the desk and glossy photos, while deepening the somber, heavy shadows in the background.",
                "applies_to_shots": [
                  4,
                  5,
                  7
                ],
                "source_facts": [
                  "Shot 4: 'Harsh police-station fluorescents are softened by humid air and monitor spill'",
                  "Shot 5: 'Cold fluorescent light makes the glossy photo surfaces glare in patches'",
                  "Shot 7: 'cutting through the police-office gloom'"
                ],
                "visual_inferences": [
                  "The overall ambient light in the room is distinctly cool-toned and uninviting, typical of institutional spaces at night or in tense situations."
                ],
                "creative_decisions": [
                  "Using the contrast between the cold, harsh overhead glares and the deep background gloom to amplify the psychological tension of the discovery."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 14,
                "shot_index": 5,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [
              {
                "scene_index": 19,
                "shot_index": 5,
                "description": "책상에 어지럽게 널린, 목과 어깨의 살점이 뜯긴 끔찍한 시신이 찍힌 살인사건 현장 사진들 클로즈업.",
                "keep_elements": [
                  {
                    "label": "scattered glossy crime scene photographs on the desk",
                    "kind": "static_prop"
                  },
                  {
                    "label": "scratched gray laminate police desk surface",
                    "kind": "environment"
                  },
                  {
                    "label": "dull silver paperclips beside the photos",
                    "kind": "static_prop"
                  }
                ]
              }
            ],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "2ff860396cbfde77"
      },
      {
        "scene_index": 19,
        "heading": "S#21. 경찰서 안 / D",
        "beat_title": "살인사건 사진을 보고 충격받는 수리영",
        "representative_moment": "책상 위에 어지럽게 널린 살인사건 현장 사진(P07) 클로즈업.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Steep overhead high angle close-up, cold fluorescent lighting with a nauseating warm red-brown accent",
            "t2i_prompt": "Photorealistic cinematic still. Overhead high angle from above, close-up tabletop insert. Focus on P07, several glossy crime scene photographs, chaotically scattered and overlapping across [: the same scratched gray laminate desk surface] that fills the entire frame. The glossy front faces of P07 are turned upward to the camera, displaying forensic images of a human figure, an Asian adult, with gruesome, torn neck and shoulder flesh details isolated in the center. Thin official case-number labels with small printed faces are attached to several photo borders, partly cropped. A loose cluster of dull silver paperclips rests beside the photo edges. The same harsh, cold overhead fluorescent lighting casts a sterile bluish-white glare in patches across the glossy photo surfaces, contrasting sharply with the deep red-brown wound details that create a sickening warm accent. All human presence is entirely cropped out.",
            "outfit_assignments": [],
            "source_facts": [
              "scene text: '혜수 책상위에는 살인사건 확대 사진들이 어지럽게 널려있다. 목, 어깨, 얼굴등 살점이 떨어져'",
              "Shot 5: 'The images overlap chaotically across the desk, with torn neck and shoulder details isolated in the center'"
            ],
            "visual_inferences": [
              "fixed_element scattered_evidence_photos: glossy crime scene photographs chaotically scattered and overlapping on the standard office desk",
              "fixed_element cold_fluorescent_lighting: police station environment illuminated by harsh, cold overhead fluorescent lights, sterile bluish-white glare"
            ],
            "creative_decisions": [
              "Steep overhead close-up point-of-view tabletop insert to strictly isolate the horrifying evidence, cropping out the observer entirely.",
              "Using the word 'same' to link the continuity of the desk surface and lighting directly from the previous shot as per the zoom_in_detail constraint."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "5cb1e91925ddca1d",
              "owned_hash": "704a968fbe0b5a95",
              "camera_direction_hash": "839082b79a8927af",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 19,
        "dependency_reason": "zoom_in_detail from the previous shot (Shot 4), focusing tightly on the scattered evidence photos under the same fluorescent lighting.",
        "_shot_index": 5,
        "visible_entities": [
          "P07"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 19,
            "shot_index": 5
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Following 수리영’s gaze, the camera holds a steep high angle CU, point-of-view tabletop insert over the scattered murder-scene photographs. The images overlap chaotically across the desk, with torn neck and shoulder details isolated in the center while all human presence is cropped out.",
            "lighting_mood": "Cold fluorescent light makes the glossy photo surfaces glare in patches, turning the red-brown wound details into the only warm, nauseating accent.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "P07"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L10B01",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "scattered_evidence_photos",
                "element_type": "persistent_prop",
                "character_name": "",
                "description": "Several glossy crime scene photographs are chaotically scattered and overlapping on the surface of a standard office desk. The visible images display gruesome, torn flesh details with dark red-brown tones, contrasting sharply with the mundane desk surface.",
                "applies_to_shots": [
                  4,
                  5
                ],
                "source_facts": [
                  "scene text: '혜수 책상위에는 살인사건 확대 사진들이 어지럽게 널려있다. 목, 어깨, 얼굴등 살점이 떨어져'",
                  "Shot 4: 'her gaze drops toward the evidence below'",
                  "Shot 5: 'The images overlap chaotically across the desk, with torn neck and shoulder details isolated in the center'"
                ],
                "visual_inferences": [
                  "The photographs are directly resting on the desk under the character's immediate view, acting as a horrifying focal point among otherwise normal office items."
                ],
                "creative_decisions": [
                  "Highlighting the red-brown wound details within the photos to create a sickening, warm accent against the cold desk environment."
                ],
                "confidence": "high"
              },
              {
                "element_id": "cold_fluorescent_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The police station environment is illuminated by harsh, cold overhead fluorescent lights. This lighting casts a sterile, bluish-white glare across flat surfaces like the desk and glossy photos, while deepening the somber, heavy shadows in the background.",
                "applies_to_shots": [
                  4,
                  5,
                  7
                ],
                "source_facts": [
                  "Shot 4: 'Harsh police-station fluorescents are softened by humid air and monitor spill'",
                  "Shot 5: 'Cold fluorescent light makes the glossy photo surfaces glare in patches'",
                  "Shot 7: 'cutting through the police-office gloom'"
                ],
                "visual_inferences": [
                  "The overall ambient light in the room is distinctly cool-toned and uninviting, typical of institutional spaces at night or in tense situations."
                ],
                "creative_decisions": [
                  "Using the contrast between the cold, harsh overhead glares and the deep background gloom to amplify the psychological tension of the discovery."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 19,
                "shot_index": 4,
                "ref_usage": "zoom_in_detail",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "81170ff2ae20d56d"
      },
      {
        "scene_index": 19,
        "heading": "S#21. 경찰서 안 / D",
        "beat_title": "해상 좌표 발견",
        "representative_moment": "모니터 화면에 뜬 해상 지도의 붉은 핀과 영문 좌표 숫자를 극단적 클로즈업으로 보여준다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Extreme close-up screen insert, cold monitor glow, sharp focus on digital text",
            "t2i_prompt": "Photorealistic cinematic still. Extreme close-up of a computer monitor screen displaying a sea map interface. The flat screen face is viewed directly, showing slightly pixelated digital textures. A bright, crisp red location pin sits near the upper-left third, while glowing white English-number coordinate text is sharply visible in the lower-right third. The cold blue-white light of the monitor dominates the framing, carrying a sterile, bluish-white glare across the screen surface. The vivid red pin stands out intensely against the cool map background. At the very bottom margin of the frame, a thin, completely out-of-focus strip of a stationary keyboard rests in deep shadow. Focus tightly on the glowing white coordinate text and the sharp red map pin on the display surface.",
            "outfit_assignments": [],
            "source_facts": [
              "S#21 action: '<화면내용> LAST LOCATION 해상 지도위에 좌표. 37.484400 N / 126. 072486 E'",
              "shot description: 모니터 화면 속 해상 지도의 붉은 핀과 영문 좌표 숫자가 적힌 글씨 클로즈업"
            ],
            "visual_inferences": [
              "fixed_element: cold_fluorescent_lighting contributing to the sterile, bluish-white glare on the screen",
              "keyboard edge at the bottom provides physical grounding for the extreme close-up of the digital screen"
            ],
            "creative_decisions": [
              "Positioned the red pin in the upper-left and coordinates in the lower-right to create a diagonal visual tension",
              "Applied a heavy blur to the keyboard edge to ensure the viewer's eye is locked entirely onto the digital interface"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "38461083253edf3d",
              "owned_hash": "704a968fbe0b5a95",
              "camera_direction_hash": "add66a580ceccd5e",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 19,
        "dependency_reason": "같은 공간의 샷으로서 차가운 형광등 조명 톤(cold_fluorescent_lighting)을 화면 반사에 유지",
        "_shot_index": 7,
        "visible_entities": [],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 19,
            "shot_index": 7
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Beside the monitor, the camera completes the move into an ECU, eye-level screen insert with the lens nearly parallel to the display; the red pin sits near the upper-left third and the English coordinate numbers sit sharply in the lower-right third. The bezel nearly disappears, but a thin edge of keyboard blur and desk clutter remain at the margins to keep the screen physically anchored.",
            "lighting_mood": "The monitor’s cold blue-white light becomes the dominant source, with the red pin cutting through the police-office gloom like a warning flare.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L10B01",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "cold_fluorescent_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The police station environment is illuminated by harsh, cold overhead fluorescent lights. This lighting casts a sterile, bluish-white glare across flat surfaces like the desk and glossy photos, while deepening the somber, heavy shadows in the background.",
                "applies_to_shots": [
                  4,
                  5,
                  7
                ],
                "source_facts": [
                  "Shot 4: 'Harsh police-station fluorescents are softened by humid air and monitor spill'",
                  "Shot 5: 'Cold fluorescent light makes the glossy photo surfaces glare in patches'",
                  "Shot 7: 'cutting through the police-office gloom'"
                ],
                "visual_inferences": [
                  "The overall ambient light in the room is distinctly cool-toned and uninviting, typical of institutional spaces at night or in tense situations."
                ],
                "creative_decisions": [
                  "Using the contrast between the cold, harsh overhead glares and the deep background gloom to amplify the psychological tension of the discovery."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 19,
                "shot_index": 4,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "150b87030fbd8116"
      },
      {
        "scene_index": 20,
        "heading": "S#22. 경찰서 밖 / D",
        "beat_title": "자전거를 타고 급히 출발하는 수리영",
        "representative_moment": "수리영(C01)이 경찰서 밖에서 자전거 안장에서 엉덩이를 뗀 채 급박하게 페달을 밟으며 출발한다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Low-angle tracking medium shot, muted overcast daylight with slick reflections",
            "t2i_prompt": "Photorealistic cinematic still. Deviating from the eye-level reference framing to a low-angle tracking medium shot. C01O02 in a beige hoodie, a Korean young woman, in left profile, mid-pedal on the bicycle from the reference. Her hips are lifted off the saddle, one foot driving down on the pedal, body leaning forward, eyes fixed on the distant path. Her dark hair flows back in the direction opposite to motion. The whole frame and wheels of the bicycle are visible, angled forward, the front wheel turned slightly toward the direction of travel. The wet pavement from the reference reflects the bicycle wheels and her lower legs in broken streaks beneath the moving frame. The damp concrete facade of the station building from the reference recedes obliquely behind her, with the pale gray curb line from the reference running diagonally across the lower foreground. Overcast coastal daylight washes the scene in a muted, humid atmosphere, while a faint red glow from the brake reflector accents the rear wheel. Focus on the dynamic forward tilt of the rider and bicycle as they accelerate.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "S#22 action: '주차된 자전거를 타고 이내 달리는 수리영'",
              "shot description: hips lift off the saddle, one foot drives down on the pedal, the whole bike angles forward across the wet police-station exterior"
            ],
            "visual_inferences": [
              "mid-pedal freeze with hair flowing back to indicate motion direction",
              "low-angle tracking emphasizes the physical force and urgency of the departure"
            ],
            "creative_decisions": [
              "Explicitly deviating from the reference's eye-level framing to a low angle to capture the full body and bicycle dynamics from below",
              "Highlighting the wet pavement reflections and faint red brake reflector to add depth and a sharp color accent against the muted overcast palette"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "cd31ba9c27621720",
              "owned_hash": "8ce1510a9509e291",
              "camera_direction_hash": "691f00811483660b",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "bicycle",
                  "violating_phrase": "the bicycle from the reference",
                  "reason": "The bicycle is mentioned as already present from the reference, which anchors it rather than explicitly redrawing it.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "station building",
                  "violating_phrase": "the station building from the reference",
                  "reason": "The station building is referenced as part of the existing scene from the reference, not requested as a new object to create or modify.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "curb",
                  "violating_phrase": "the pale gray curb line from the reference",
                  "reason": "The curb is only used as an existing reference element for composition, so this is an anchor reference.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "",
        "_shot_index": 2,
        "visible_entities": [
          "C01",
          "L12"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 20,
            "shot_index": 2
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Tracking parallel from a low three-quarter side position about two meters away, the camera frames a WS, low angle of 수리영’s full body and bicycle as she pushes off from the parking spot. Her hips lift off the saddle, one foot drives down on the pedal, and the whole bike angles forward across the wet police-station exterior, giving the departure urgent physical force.",
            "lighting_mood": "Overcast coastal daylight stays muted and humid, with slick ground reflections and a faint red brake reflector adding the only sharp color accent.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01",
              "L12"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L12B01",
            "reference_usage": "exact_background",
            "owned_objects": [
              "CCTV pole",
              "barrier gate",
              "bicycle",
              "bicycle rack",
              "curb",
              "door",
              "parking lot",
              "parking space",
              "patrol car",
              "planter",
              "road",
              "sidewalk",
              "signboard",
              "station building",
              "window"
            ],
            "camera_reference": {
              "camera_position": "near number 5 (main entrance door), facing diagonally across number 1 (parking lot) toward number 12 (bicycle departure route) and number 7 (driveway exit)",
              "camera_height": "eye-level standing ~1.6m",
              "lens_hint": "35mm wide angle",
              "framing_notes": "include number 11 (bicycle) and number 12 (departure route) prominently, with number 2 (station building frontage) behind and number 4 (public road) visible beyond the exit"
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "when focusing on an owned object, anchor it explicitly: 'the door from the reference'",
              "camera_reference specifies the framing under which the background was painted — match camera_position, camera_height, and framing",
              "if deviating from the reference framing (push-in, angle shift, ECU), state the deviation explicitly so the compositor can account for the scale difference"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              },
              {
                "kind": "background",
                "id": "L12B01",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "f14c5dce26e9b3d0"
      },
      {
        "scene_index": 21,
        "heading": "S#23. 어느 PC방 / 실내",
        "beat_title": "좌표 검색과 모니터 불빛",
        "representative_moment": "어두운 PC방 안, 푸른빛이 감도는 모니터 불빛에 비친 C01의 굳은 얼굴 정면 구도.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Static eye-level close-up (CU), cold blue monitor practical light, muted low-contrast coastal shadows",
            "t2i_prompt": "Photorealistic cinematic still. A static eye-level close-up holding a symmetrical frontal monitor-axis view. C01O02 in a beige hoodie, Korean female in her early 20s with black hair, a young face and dark eyes, hunched at a workstation facing the camera. The camera focuses on her tense expression in a head-and-shoulder composition, her neck and shoulders visible in the lower frame. Her dark eyes are fixed forward on the unseen monitor just past the implied screen line. Cold blue practical monitor light washes across her face and the front of her hoodie, flattening her features with no warm fill. Behind her, [: dark PC cafe computer booth partitions face the camera, receding into out-of-focus murky, low-contrast shadows in the soft background depth], reflecting muted coastal humidity.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "S#23 action: '모니터 화면에 앉아있는 수리영, 좌표를 찍고 위치를 확인하고...'",
              "shot description: A static close-up (CU) at eye-level holds a symmetrical frontal monitor-axis view... C01 hunched at workstation, facing camera, eyes on monitor."
            ],
            "visual_inferences": [
              "C01's face is illuminated entirely by the offscreen blue monitor glow",
              "The PC cafe background is dark and mostly empty, creating a tunnel-like recession of partitions behind her"
            ],
            "creative_decisions": [
              "Used a head-and-shoulder composition to fulfill the '70% face' DP note while explicitly avoiding the forbidden floating-head 'face filling the entire frame' effect",
              "Excluded the L01 ID as requested, assigning the background description directly to an L02 block",
              "Applied 'washes across' for the cold blue lighting to ensure it acts on surfaces without triggering physical cut/slice misinterpretations by the T2I model"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "0a72ff98fea906d9",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "b6cb85ef99f8d1f1",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "No prior visual dependency explicitly modeled for this specific framing",
        "_shot_index": 1,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 21,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A static close-up (CU) at eye-level holds a symmetrical frontal monitor-axis view from just beyond the implied screen line, with 수리영 centered and the dark PC café receding into soft rows behind her. The framing is direct observer vision, not a device POV; her face occupies about 70% of the frame while the monitor itself stays offscreen, suggested only by the blue light flattening her features.",
            "lighting_mood": "Cold blue monitor light is the dominant practical source, with muted coastal humidity in the shadows and almost no warm fill.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": null,
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "e73260821f51f4f1"
      },
      {
        "scene_index": 22,
        "heading": "S#24. 해안가 / 소나기",
        "beat_title": "폭풍우가 몰아치는 어두운 해안",
        "representative_moment": "먹구름이 짙게 깔린 어두운 해안가 하늘 위로 거대한 번개 줄기가 번쩍인다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Extreme wide establishing shot, high upward-tilted angle, cold slate-blue and charcoal palette overexposed by harsh lightning",
            "t2i_prompt": "Photorealistic cinematic still. An extreme wide establishing shot from a very high coastal position, with an upward-tilted angle capturing the storm ceiling. Focus on a massive, harsh white lightning bolt ripping across the upper-center frame, illuminating thick, heavy storm clouds. The static observer perspective compresses the dark sky and cold sea into one ominous plane, anchored by a thin silhouette of black shoreline along the bottom edge. Heavy diagonal rain streaks cut through the foreground air. The gloomy environment is drenched in cold slate-blue and charcoal tones, the clouds briefly overexposed by the brilliant flash before falling back into fog-softened darkness, entirely devoid of any human figure.",
            "outfit_assignments": [],
            "source_facts": [
              "Scene text: '천둥 번개와 함께 잔뜩 낀 먹구름이 세찬 비를 뿌린다'",
              "Shot description: extreme wide / establishing shot (EWS) from a very high coastal position, upward-tilted angle",
              "Shot description: holding the massive lightning bolt in the upper center while the black shoreline sits as a thin lower strip, no human figure visible"
            ],
            "visual_inferences": [
              "fixed_element stormy_coastal_weather: heavy diagonal rain streaks cutting through the dark atmosphere, cold slate-blue and charcoal tones",
              "atmosphere_reference from prior shot: maintaining the gloomy, fog-softened darkness and cold stormy mood"
            ],
            "creative_decisions": [
              "Framed the lightning bolt directly in the upper-center to create a stark focal point against the dark sky, compressing the sea and shoreline into a minimal lower strip to emphasize the overwhelming scale of the storm ceiling."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "aa08fc8f6097a299",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "230013d2afee4183",
              "validator": "scene_detail_owned_objects.v1",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 9,
        "dependency_reason": "분위기 레퍼런스(atmosphere_reference)로 이전 샷의 차갑고 어두운 폭풍우 색감과 톤(slate-blue, charcoal)을 이어받음",
        "_shot_index": 1,
        "visible_entities": [],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 22,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A near-frozen extreme wide / establishing shot (EWS) from a very high coastal position uses a high, upward-tilted angle across the storm ceiling, holding the massive lightning bolt in the upper center while the black shoreline sits as a thin lower strip. The static observer frame compresses sky and sea into one ominous plane, with rain streaks cutting diagonally through the foreground but no human figure visible.",
            "lighting_mood": "The palette is slate-blue and charcoal, briefly overexposed by the lightning flash before falling back into fog-softened darkness.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "not_applicable",
            "bg_id": null,
            "reference_usage": "none",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": false,
            "constraints": [
              "no chain_bg is bound to this shot — describe the location freely without referencing a background image",
              "do not use 'from the reference' / 'the existing room' / 'preserving the room' style phrasing — there is no reference image to refer to"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "stormy_coastal_weather",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A severe storm environment characterized by heavy diagonal rain streaks cutting through the dark atmosphere. The sky and surroundings are dominated by cold slate-blue, charcoal, and muted storm-gray tones, periodically illuminated by bright, harsh lightning flashes from thick, dark storm clouds.",
                "applies_to_shots": [
                  1,
                  4
                ],
                "source_facts": [
                  "Scene text: '천둥 번개와 함께 잔뜩 낀 먹구름이 세찬 비를 뿌린다'",
                  "Shot 1: 'rain streaks cutting diagonally', 'slate-blue and charcoal, briefly overexposed by the lightning flash'",
                  "Shot 4: 'Storm-muted gray light', 'faint distant lightning flicker'"
                ],
                "visual_inferences": [
                  "The intensity of the rain and the gloomy, oppressive darkness remain constant regardless of whether the camera is far away or up close."
                ],
                "creative_decisions": [
                  "Using a highly cohesive cold gray and slate-blue color palette to visually link the extreme wide landscape with the intimate close-up."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 9,
                "shot_index": 1,
                "ref_usage": "atmosphere_reference",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [],
            "readiness_policy": "not_applicable",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "f620272cd356a515"
      },
      {
        "scene_index": 22,
        "heading": "S#24. 해안가 / 소나기",
        "beat_title": "거센 비바람 속 필사적인 질주",
        "representative_moment": "C01 in left profile, her wet face reflecting cold light, eyes fixed straight ahead while frozen mid-pedal in the storm.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Tracking eye-level head-and-shoulder framing in profile, shallow DOF with motion blur, storm-muted cold gray lighting with faint lightning flicker.",
            "t2i_prompt": "Photorealistic cinematic still. A head-and-shoulder composition, C01O03 in a dark navy windbreaker, a young Korean woman in her early 20s, positioned in left profile at the center-right. Her neck and shoulder edge, along with a small slice of the wet bicycle handlebar, are visible in the lower frame. She is frozen mid-pedal leaning forward, eyes fixed on the distant space ahead. Wet hair flowing backward in the direction opposite to motion. Suspended raindrops clinging to her cheek and streaking past the lens plane. Storm-muted cold gray light washes her face, with faint lightning flicker catching on her wet skin. A dark, horizontally blurred wet road smears behind her in a soft background, conveying high-speed momentum without revealing the wider environment.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O03"
              }
            ],
            "source_facts": [
              "Scene text: '자전거를 타고 빗속을 달리는 수리영. 오로지 엄마를 찾아야한다는 생각 뿐.'",
              "Shot 4: '뺨을 타고 내리던 빗방울이 피부에 맺힌 채, 핏발 선 눈으로 정면을 고정하여 응시하는 수리영의 옆얼굴 클로즈업.'"
            ],
            "visual_inferences": [
              "fixed_element stormy_coastal_weather: heavy diagonal rain streaks and cold slate-blue storm-gray tones",
              "mid-action motion frozen with hair flowing opposite to movement",
              "bicycle handlebar included at the frame edge to ground the riding action"
            ],
            "creative_decisions": [
              "Used a head-and-shoulder composition to prevent floating head artifact while maintaining a tight focus on her facial expression.",
              "Applied horizontal motion blur to the background to represent the tracking shot speed without importing external background references.",
              "Placed lightning reflections gently on the wet skin rather than as direct beams, avoiding harsh visual cuts across the face."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "5c154a92cde6d630",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "d4e793989a73fafe",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 9,
        "dependency_reason": "zoom_in_detail: preserves the same continuous storm lighting and spatial mood established in the preceding sequence",
        "_shot_index": 4,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 22,
            "shot_index": 4
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A tracking extreme close-up (ECU) at eye-level runs shoulder-height beside 수리영, less than a meter from her rain-slick side profile, using profile framing and shallow DOF. Her bloodshot eye and one suspended raindrop sit in the right half of the frame, while the blurred wet road and bicycle motion smear horizontally behind her.",
            "lighting_mood": "Storm-muted gray light wraps her face with cold highlights on rain, punctuated by faint distant lightning flicker.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O03"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": null,
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "stormy_coastal_weather",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A severe storm environment characterized by heavy diagonal rain streaks cutting through the dark atmosphere. The sky and surroundings are dominated by cold slate-blue, charcoal, and muted storm-gray tones, periodically illuminated by bright, harsh lightning flashes from thick, dark storm clouds.",
                "applies_to_shots": [
                  1,
                  4
                ],
                "source_facts": [
                  "Scene text: '천둥 번개와 함께 잔뜩 낀 먹구름이 세찬 비를 뿌린다'",
                  "Shot 1: 'rain streaks cutting diagonally', 'slate-blue and charcoal, briefly overexposed by the lightning flash'",
                  "Shot 4: 'Storm-muted gray light', 'faint distant lightning flicker'"
                ],
                "visual_inferences": [
                  "The intensity of the rain and the gloomy, oppressive darkness remain constant regardless of whether the camera is far away or up close."
                ],
                "creative_decisions": [
                  "Using a highly cohesive cold gray and slate-blue color palette to visually link the extreme wide landscape with the intimate close-up."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 9,
                "shot_index": 1,
                "ref_usage": "zoom_in_detail",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O03",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "1c02aaa0b645bf8d"
      },
      {
        "scene_index": 23,
        "heading": "S#25. 옥탑방 안 / 해질 무렵",
        "beat_title": "난장판이 된 집안과 낯선 발자국",
        "representative_moment": "비에 젖은 거실 바닥에 찍힌 진흙 묻은 낯선 발자국을 매우 낮게 밀착해서 바라보는 스틸.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Very low close-up riding just above the floor, deep blue-black shadows with cool silver-gray dusk light.",
            "t2i_prompt": "Photorealistic cinematic still. A very low close-up riding just above the ground, lens nearly parallel. Focus on a trail of muddy unfamiliar footprints tracking across the wet surface, the smeared shoe shapes and dirty edges sharply detailed. The muddy tracks occupy the central band of the frame. [L04: Rain-wet floor covered in thin, reflective puddles under fading dusk. Deep blue-black shadows dominate the unlit space, while a weak, cool silver-gray evening light filters in, catching the moisture. In the soft, blurred rear background, an unlit entrance doorway forms a deep shadowed rectangular void, alongside out-of-focus shapes of overturned dining chairs with legs angled forward, and an open window revealing gray dusk and damp air, establishing a gloomy, sickly humid, and unsettling atmosphere.]",
            "outfit_assignments": [],
            "source_facts": [
              "Shot 1: close-up of muddy footprints on a rain-wet living room floor.",
              "Shot description notes open living-room window, fallen dining chairs, and reflective rainwater puddles.",
              "Fading dusk leaks through the open window, turning the wet floor silver-gray while the room stays underlit."
            ],
            "visual_inferences": [
              "fixed_element dusk_apartment_interior_lighting: The apartment interior is enveloped in dark, low-key dusk lighting with deep blue-black shadows.",
              "fixed_element shadowed_entrance_doorway: The main entrance doorway acts as a heavily shadowed rectangular void in the background."
            ],
            "creative_decisions": [
              "Replaced '15 cm' with 'just above the ground' to avoid real numerical dimensions while maintaining the extremely low camera angle.",
              "Applied a shallow depth of field to keep the muddy footprints sharp while blurring the doorway and scattered chairs into atmospheric background shapes to comply with the close framing and background binding rules."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "a250f3a3a9b5fe4f",
              "owned_hash": "4d307ad28636e606",
              "camera_direction_hash": "472430d9774ba92a",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 11,
        "dependency_reason": "동일한 옥탑방 내부 배경",
        "_shot_index": 1,
        "visible_entities": [
          "L04"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 23,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A very low close-up (CU) at a high-angle-to-surface perspective rides only about 15 cm above the wet living-room floor, lens nearly parallel as it dollies into muddy unfamiliar footprints. The muddy tracks occupy the central band of the frame, with the entrance side blurred behind and the disturbed room hinted by toppled shapes in the far background.",
            "lighting_mood": "Fading dusk leaks through the open window, turning the wet floor silver-gray while the room stays underlit and sickly humid.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "L04"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B03",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "dusk_apartment_interior_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The interior of the apartment is enveloped in dark, low-key dusk lighting. Deep blue-black shadows dominate the unlit space, while a weak, cool silver-gray evening light softly filters in from the outside, creating a gloomy, humid, and unsettling atmosphere across the rooms.",
                "applies_to_shots": [
                  1,
                  6,
                  11
                ],
                "source_facts": [
                  "Shot 1: Fading dusk leaks through the open window, turning the wet floor silver-gray while the room stays underlit.",
                  "Shot 6: Low dusk light mixes with a faint warm interior practical, gray-green shadows.",
                  "Shot 11: doorway falls into blue-black interior shadow while the exterior dusk gives her face a weak gray rim."
                ],
                "visual_inferences": [
                  "The lack of strong indoor lighting ensures the entire apartment interior consistently reads as an eerie, shadowed void heavily influenced by the cool dusk light from outside."
                ],
                "creative_decisions": [
                  "Unifying the color palette across different sections of the apartment with a blue-gray dusk tint and heavy black shadows to heighten the psychological tension and sense of abandonment."
                ],
                "confidence": "high"
              },
              {
                "element_id": "shadowed_entrance_doorway",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The main entrance doorway of the apartment stands as a stark, unlit architectural feature. The frame of the open door reveals a deep, heavily shadowed rectangular void, completely lacking any welcoming interior light.",
                "applies_to_shots": [
                  1,
                  11
                ],
                "source_facts": [
                  "Shot 1: the entrance side blurred behind.",
                  "Shot 11: faces back into the dark doorway... doorway remains a shadowed rectangle behind her."
                ],
                "visual_inferences": [
                  "Since the apartment is unlit, the entrance area receives no practical illumination, rendering it as a dark silhouette or a black rectangle in the background."
                ],
                "creative_decisions": [
                  "Using the unlit, shadowy doorway as a consistent framing device to emphasize the emptiness and unsettling nature of the ransacked apartment."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 11,
                "shot_index": 14,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "2eea09a9a8456390"
      },
      {
        "scene_index": 23,
        "heading": "S#25. 옥탑방 안 / 해질 무렵",
        "beat_title": "벽에 꽂힌 과거의 단서 발견",
        "representative_moment": "A close-up of the blood-stained old photograph, P02, pinned to the center of a red circular mark, P08, on the bedroom wall.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Eye-level chest-height close-up, shallow depth of field, cool dusk mixed with faint warm practical light",
            "t2i_prompt": "Photorealistic cinematic still. Eye-level chest-height close-up composition, shallow depth of field. Focus on the sharp metal pin, dark blood flecks, and cracked paper texture of P02. P02, a yellowed, wrinkled old photograph, occupies the center of the frame. The visible face of P02 shows a faded rural restaurant and two young Asian girls, partially obscured by dark blood smears across the aged paper. P02 is fixed by a metal pin directly to [L04: the same damp-stained bedroom wall surface]. An uneven, ominous red circular mark, P08, is painted directly onto the wall, fully encircling P02 like an occult target. Dark, low-key dusk lighting with deep blue-black shadow gradients surrounds the pinned photo. A weak, cool silver-gray ambient light mixes with a faint warm interior practical glow falling on the wall, making P08 appear unnaturally saturated against the otherwise dim gray-green background.",
            "outfit_assignments": [],
            "source_facts": [
              "S#25 text: 벽에 또 붉은 원이 그려져있다. 그 가운데 꽂혀있는 피묻은 사진 한 장!",
              "Shot 6 DP instruction: A slow dolly-in close-up (CU) at chest-height eye-level pushes toward the bedroom wall, using shallow DOF so the pin point, blood flecks, and cracked paper texture become the sharp central focus.",
              "DP instruction: showing a faded rural restaurant, two young girls, and dark blood smears across the aged paper."
            ],
            "visual_inferences": [
              "fixed_element dusk_apartment_interior_lighting applied: Deep blue-black shadows dominate the unlit space, while a weak, cool silver-gray evening light softly filters in.",
              "Reproduction surface rule applied: The two young girls in the photograph are described with common nouns and demographic descriptors ('two young Asian girls') instead of character IDs."
            ],
            "creative_decisions": [
              "Framing P02 in the center, encircled entirely by the red mark P08, emphasizing its presence as a grim focal target.",
              "Using a mix of cool dusk ambient and a faint warm practical glow on the wall surface to make the red hue of P08 unnaturally saturated, heightening the eerie tension."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "456e9dd21d658631",
              "owned_hash": "db5e9f1b6e8d2520",
              "camera_direction_hash": "a180b9b46640bcbf",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 11,
        "dependency_reason": "zoom_in_detail: zooms into the exact background detail (the damp-stained bedroom wall surface) established in Scene 11 Shot 14, preserving the exact lighting and environment.",
        "_shot_index": 6,
        "visible_entities": [
          "P02",
          "P08",
          "L04"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 23,
            "shot_index": 6
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A slow dolly-in close-up (CU) at chest-height eye-level pushes toward the bedroom wall, using shallow DOF so the pin point, blood flecks, and cracked paper texture become the sharp central focus. The old photograph stays within the middle 35–40% of the frame, surrounded by the red circle on the wall like an occult target rather than letting the prop unrealistically overtake the whole image.",
            "lighting_mood": "Low dusk light mixes with a faint warm interior practical, making the red circle feel unnaturally saturated against otherwise gray-green shadows.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "P02",
              "P08",
              "L04"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B02",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "dusk_apartment_interior_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The interior of the apartment is enveloped in dark, low-key dusk lighting. Deep blue-black shadows dominate the unlit space, while a weak, cool silver-gray evening light softly filters in from the outside, creating a gloomy, humid, and unsettling atmosphere across the rooms.",
                "applies_to_shots": [
                  1,
                  6,
                  11
                ],
                "source_facts": [
                  "Shot 1: Fading dusk leaks through the open window, turning the wet floor silver-gray while the room stays underlit.",
                  "Shot 6: Low dusk light mixes with a faint warm interior practical, gray-green shadows.",
                  "Shot 11: doorway falls into blue-black interior shadow while the exterior dusk gives her face a weak gray rim."
                ],
                "visual_inferences": [
                  "The lack of strong indoor lighting ensures the entire apartment interior consistently reads as an eerie, shadowed void heavily influenced by the cool dusk light from outside."
                ],
                "creative_decisions": [
                  "Unifying the color palette across different sections of the apartment with a blue-gray dusk tint and heavy black shadows to heighten the psychological tension and sense of abandonment."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 11,
                "shot_index": 14,
                "ref_usage": "zoom_in_detail",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "prop",
                "id": "P02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [
            {
              "contract_id": "rc_001",
              "scope": {
                "scene_index": 23,
                "shot_indices": [
                  6
                ],
                "duration": "single_shot"
              },
              "targets": [
                {
                  "entity_id": "P02",
                  "role": "visual_target"
                }
              ],
              "requirements": [
                {
                  "dimension": "visual_identity",
                  "operation": "preserve",
                  "strength": "required",
                  "reference_policy": "use_entity_reference"
                }
              ]
            }
          ],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "846a21619d101a97"
      },
      {
        "scene_index": 23,
        "heading": "S#25. 옥탑방 안 / 해질 무렵",
        "beat_title": "혼미한 상태로 방을 나서는 수리영",
        "representative_moment": "정신이 혼미해진 C01이 오래된 사진을 가슴에 품은 채 옥탑방 현관문을 나선다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Retreating medium shot at a slight low angle, cool silver-gray dusk lighting mixed with deep blue-black shadows",
            "t2i_prompt": "Photorealistic cinematic still. A retreating medium shot at a slight low angle, upper body and suspended knee in frame. C01O03 in a dark navy windbreaker, a young Korean woman, captured mid-stride, one foot suspended midair in mid-step, her hair faintly trailing in the direction opposite to motion. She is positioned center-left, at three-quarter angle from the left, her dark eyes staring into the void. She clutches an old creased photograph tightly against her chest, the image side turned toward her body so only a stained edge faces the camera. She wears P04, a worn backpack, its rear side visible over her shoulder. The open doorway from the reference frames her from behind, its interior enveloped in dark, low-key dusk lighting, forming a deep blue-black shadowed rectangular void. Exterior cool silver-gray evening dusk catches the side of her face with a weak gray rim light. Focus on her vacant trance and the tension of her suspended mid-stride pose against the dark background.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O03"
              }
            ],
            "source_facts": [
              "S#25 텍스트: 마치 이끌리듯 백팩을 메고 밖으로 나간다. 사진을 손에 꼭 쥔 채.",
              "Shot 11: 백팩을 멘 채 어두워진 옥탑방 현관문 밖으로 걸어 나가는 도중, 한쪽 발이 공중에 떠 있는 찰나의 자세로 멈춘 수리영의 몽롱한 표정."
            ],
            "visual_inferences": [
              "fixed_element dusk_apartment_interior_lighting: The interior behind her reads as an eerie, shadowed void.",
              "fixed_element shadowed_entrance_doorway: The doorframe forms a dark rectangular void."
            ],
            "creative_decisions": [
              "Employed a slight low angle retreating medium shot to capture her suspension mid-stride while keeping the shadowed doorway looming behind her.",
              "Kept the photograph's image facing inward to strictly maintain the DP's orientation constraints and avoid revealing unprompted details."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "a4ba1473856850ca",
              "owned_hash": "4d307ad28636e606",
              "camera_direction_hash": "90d2421ae6e9807f",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "door",
                  "violating_phrase": "open doorway from the reference",
                  "reason": "The prompt explicitly references the existing doorway as a framing anchor from the reference, not as a newly drawn object.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 23,
        "dependency_reason": "exact_background (maintains the same location as S23_Shot1 but facing back into the dark doorway)",
        "_shot_index": 11,
        "visible_entities": [
          "C01",
          "P04",
          "L04"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 23,
            "shot_index": 11
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "From just outside the rooftop-room entrance, a retreating medium shot (MS) at a slight low angle faces back into the dark doorway as 수리영 steps out with one foot suspended midair. The observer frame holds her in the center-left, her old photograph clutched close at chest level, while the doorway remains a shadowed rectangle behind her rather than competing for attention.",
            "lighting_mood": "The doorway falls into blue-black interior shadow while the exterior dusk gives her face a weak gray rim, emphasizing her vacant trance.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01",
              "P04",
              "L04"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O03"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L04B03",
            "reference_usage": "exact_background",
            "owned_objects": [
              "TV",
              "chair",
              "coffee table",
              "cooktop",
              "counter",
              "curtain",
              "dining table",
              "door",
              "footprint",
              "hallway",
              "herb pot",
              "kitchenette",
              "living room",
              "refrigerator",
              "shoe cabinet",
              "sink",
              "smartphone",
              "sofa",
              "washing machine",
              "window"
            ],
            "camera_reference": {
              "camera_position": "near number 8 (front door), facing through number 3 (living area) toward number 19 (openable living sliding window), number 40 (muddy footprint trail), and number 41 (fallen chair pair)",
              "camera_height": "eye-level standing ~1.6m",
              "lens_hint": "28mm wide angle",
              "framing_notes": "Show the ransacked path from the entrance through the room to the window."
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "when focusing on an owned object, anchor it explicitly: 'the door from the reference'",
              "camera_reference specifies the framing under which the background was painted — match camera_position, camera_height, and framing",
              "if deviating from the reference framing (push-in, angle shift, ECU), state the deviation explicitly so the compositor can account for the scale difference"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "dusk_apartment_interior_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The interior of the apartment is enveloped in dark, low-key dusk lighting. Deep blue-black shadows dominate the unlit space, while a weak, cool silver-gray evening light softly filters in from the outside, creating a gloomy, humid, and unsettling atmosphere across the rooms.",
                "applies_to_shots": [
                  1,
                  6,
                  11
                ],
                "source_facts": [
                  "Shot 1: Fading dusk leaks through the open window, turning the wet floor silver-gray while the room stays underlit.",
                  "Shot 6: Low dusk light mixes with a faint warm interior practical, gray-green shadows.",
                  "Shot 11: doorway falls into blue-black interior shadow while the exterior dusk gives her face a weak gray rim."
                ],
                "visual_inferences": [
                  "The lack of strong indoor lighting ensures the entire apartment interior consistently reads as an eerie, shadowed void heavily influenced by the cool dusk light from outside."
                ],
                "creative_decisions": [
                  "Unifying the color palette across different sections of the apartment with a blue-gray dusk tint and heavy black shadows to heighten the psychological tension and sense of abandonment."
                ],
                "confidence": "high"
              },
              {
                "element_id": "shadowed_entrance_doorway",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The main entrance doorway of the apartment stands as a stark, unlit architectural feature. The frame of the open door reveals a deep, heavily shadowed rectangular void, completely lacking any welcoming interior light.",
                "applies_to_shots": [
                  1,
                  11
                ],
                "source_facts": [
                  "Shot 1: the entrance side blurred behind.",
                  "Shot 11: faces back into the dark doorway... doorway remains a shadowed rectangle behind her."
                ],
                "visual_inferences": [
                  "Since the apartment is unlit, the entrance area receives no practical illumination, rendering it as a dark silhouette or a black rectangle in the background."
                ],
                "creative_decisions": [
                  "Using the unlit, shadowy doorway as a consistent framing device to emphasize the emptiness and unsettling nature of the ransacked apartment."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 23,
                "shot_index": 1,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O03",
                "policy": "required"
              },
              {
                "kind": "background",
                "id": "L04B03",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "5ae0871ecb64183a"
      },
      {
        "scene_index": 24,
        "heading": "S#26. 포구 / 해질 무렵",
        "beat_title": "수리영의 단절된 소통과 고립감",
        "representative_moment": "안개 낀 포구의 갯바위 위에서, 수리영(C01)의 젖은 손이 혜수이모의 전화가 울리는 스마트폰(P01) 화면을 물끄러미 내려다보고 있다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "A tight close-up from seated shoulder height tilting down to hands, shallow DOF, cool dusk tone",
            "t2i_prompt": "Photorealistic cinematic still. A tight close-up from an eye-level-to-hands angle, focus on the vibrating lit front screen of P01 facing the camera at a slight diagonal in the lower third of the frame. The device is tightly gripped by the tense, rain-damp fingers of a young Korean woman in her early 20s wearing the sleeve of a dark navy windbreaker. The phone's small bluish glow breaks the cold gray fog light, catching dim practical reflections on the wet skin. Below the hands, dark, slick, uneven harbor rocks provide a grounding surface at the frame edge. Beyond the immediate foreground, a dense, pervasive gray-blue harbor fog acts as a semi-opaque veil in the shallow depth of field, heavily obscuring distant elements and softening the moody, low-contrast dusk atmosphere.",
            "outfit_assignments": [],
            "source_facts": [
              "S#26 action: '진동벨이 울린다. 얼핏 보니 혜수이모다. 벌써 여러통 째 부재중 전화.'",
              "Shot description: smartphone screen close-up, held by Suri-young, vibrating with incoming call"
            ],
            "visual_inferences": [
              "fixed_element thick_harbor_fog: dense gray-blue haze obscuring the background",
              "body_part_focus_rule applied: C01 is represented by 'tense, rain-damp fingers of a young Korean woman in her early 20s' without a composite ID",
              "DP framing: focus on hands holding P01, catching bluish glow from the screen"
            ],
            "creative_decisions": [
              "Placed P01 in the lower third of the frame to anchor the visual weight of the device against the empty foggy background",
              "Used shallow depth of field to isolate the trembling, wet hand and sharp screen details from the soft, uniform fog and slick rocks"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "287e280f9daac258",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "d45e101f45e98b8a",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "N/A",
        "_shot_index": 3,
        "visible_entities": [
          "C01",
          "P01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 24,
            "shot_index": 3
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A tight close-up (CU) from seated shoulder height beside 수리영 tilts down to her hands, using an eye-level-to-hands angle and shallow DOF to make the vibrating smartphone screen readable without letting it exceed the lower third of the frame. This is direct observer vision: her wet fingers and sleeve frame the display showing ‘혜수이모,’ while the slick rocks and foggy harbor remain soft scale references around it.",
            "lighting_mood": "Cold gray fog light is broken by the phone’s small bluish glow, with wet skin and stone catching dim practical reflections.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01",
              "P01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O03"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": null,
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "thick_harbor_fog",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A dense, pervasive fog blankets the coastal area during dusk. It creates a thick gray-blue haze that heavily obscures distant elements and softens the ambient lighting into a moody, low-contrast atmosphere.",
                "applies_to_shots": [
                  3,
                  4
                ],
                "source_facts": [
                  "S#26 action: '안개가 자욱하다. 텅 빈 포구.'",
                  "Shot 3 description: 'foggy harbor remain soft scale references'",
                  "Shot 4 description: 'compressed wall of fog', 'Fog-diffused dusk light'"
                ],
                "visual_inferences": [
                  "The fog imparts a cold, uniform gray-blue atmospheric tint across the harbor area due to the dusk timing."
                ],
                "creative_decisions": [
                  "Use the heavy fog as a consistent backdrop to visually isolate the subjects and establish a gloomy, mysterious mood across varying framings."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "266f8540c3efcdde"
      },
      {
        "scene_index": 24,
        "heading": "S#26. 포구 / 해질 무렵",
        "beat_title": "유일한 가능성의 등장",
        "representative_moment": "An old, weathered fishing boat emerges bow-first from the thick dusk fog toward the empty pier.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "eye-level medium perspective, telephoto compression, fog-diffused low-contrast gray-blue dusk lighting",
            "t2i_prompt": "Photorealistic cinematic still. [L14: A wet and empty pier edge leading into a dark, foggy harbor at dusk]. An old, small fishing boat emerges bow-first from a compressed wall of thick, dense fog, the bow occupying the middle third of the frame and facing directly forward. The boat exhibits a visibly weathered exterior with chipped rust-red and red-brown paint accents, its hull completely wet and slick with moisture. The fog is semi-transparent only at the center where the bow breaks through, heavily obscuring the stern so the vessel seems to push out of a blank gray-white void. Below, dark horizontal bands of harbor water ripples form, the surface reflecting the broken gray fog and the dark underside of the boat in wavering fragments. A slice of the wet pier edge runs across the lower foreground corner. Fog-diffused dusk light flattens the contrast into a moody gray-blue ambient tone, with the dim rust-red paint adding a restrained warmth. Focus on the weathered bow of the boat piercing the fog wall.",
            "outfit_assignments": [],
            "source_facts": [
              "S#26 action: '안개를 뚫고 배 한척이 다가오는 게 보인다.'",
              "Shot 4 description: 'old fishing boat emerges bow-first in the center', 'compressed wall of fog', 'Fog-diffused dusk light flattens contrast into gray-blue'"
            ],
            "visual_inferences": [
              "fixed_element thick_harbor_fog: dense gray-blue atmospheric tint",
              "fixed_element weathered_fishing_boat: visibly aged exterior and wet from recent rain",
              "water surface reflects the bow and broken fog as described in the shot constraints"
            ],
            "creative_decisions": [
              "Telephoto lens compression to make the fog appear as a dense, impenetrable wall",
              "Placing a slice of the pier edge in the foreground corner to establish the observer's static position on land"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "beea3c0186691453",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "db7a884e0b54ae8c",
              "validator": "scene_detail_owned_objects.v1",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 24,
        "dependency_reason": "Carrying over the thick harbor fog and dusk atmosphere established in the prior shot (atmosphere_reference).",
        "_shot_index": 4,
        "visible_entities": [
          "L14"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 24,
            "shot_index": 4
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A static medium wide shot (MWS) with a telephoto eye-level angle waits at the pier edge, facing straight into a compressed wall of fog as the old fishing boat emerges bow-first in the center. The bow occupies the middle third of the frame, water ripples form dark horizontal bands below, and the fog erases the stern so the boat seems to push out of a blank void.",
            "lighting_mood": "Fog-diffused dusk light flattens contrast into gray-blue, with the boat’s rust-red accent cutting faintly through the gloom.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "L14"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "not_applicable",
            "bg_id": null,
            "reference_usage": "none",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": false,
            "constraints": [
              "no chain_bg is bound to this shot — describe the location freely without referencing a background image",
              "do not use 'from the reference' / 'the existing room' / 'preserving the room' style phrasing — there is no reference image to refer to"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "thick_harbor_fog",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A dense, pervasive fog blankets the coastal area during dusk. It creates a thick gray-blue haze that heavily obscures distant elements and softens the ambient lighting into a moody, low-contrast atmosphere.",
                "applies_to_shots": [
                  3,
                  4
                ],
                "source_facts": [
                  "S#26 action: '안개가 자욱하다. 텅 빈 포구.'",
                  "Shot 3 description: 'foggy harbor remain soft scale references'",
                  "Shot 4 description: 'compressed wall of fog', 'Fog-diffused dusk light'"
                ],
                "visual_inferences": [
                  "The fog imparts a cold, uniform gray-blue atmospheric tint across the harbor area due to the dusk timing."
                ],
                "creative_decisions": [
                  "Use the heavy fog as a consistent backdrop to visually isolate the subjects and establish a gloomy, mysterious mood across varying framings."
                ],
                "confidence": "high"
              },
              {
                "element_id": "weathered_fishing_boat",
                "element_type": "environment_state",
                "character_name": "",
                "description": "An old, small fishing boat exhibits a visibly weathered exterior with chipped rust-red and red-brown paint accents. Its hull and deck surfaces are slick with moisture, reflecting the dim ambient light.",
                "applies_to_shots": [
                  4,
                  9
                ],
                "source_facts": [
                  "Shot 4 description: 'old fishing boat emerges bow-first', 'rust-red accent cutting faintly through the gloom'",
                  "Shot 9 description: 'wet boat deck', 'dim red-brown boat paint accents adding a restrained ominous warmth'"
                ],
                "visual_inferences": [
                  "The boat looks visibly aged, and its exposed surfaces are completely wet due to the recent heavy rain shower mentioned in the scene text."
                ],
                "creative_decisions": [
                  "Highlight the muted red-brown paint to provide a subtle, consistent color contrast against the otherwise monochromatic gray-blue foggy environment."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 24,
                "shot_index": 3,
                "ref_usage": "atmosphere_reference",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [],
            "readiness_policy": "not_applicable",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "c13f1deb816f427b"
      },
      {
        "scene_index": 24,
        "heading": "S#26. 포구 / 해질 무렵",
        "beat_title": "수리영의 승선과 결심",
        "representative_moment": "수리영(C01)이 낡은 어선 갑판 위로 한 발을 내딛으며 승선하는 정지된 순간.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Low-angle medium close-up (MCU) from deck level, muted gray ambient lighting with subtle red-brown accents",
            "t2i_prompt": "Photorealistic cinematic still. A low-angle medium close-up from the deck level. C01O03 in a dark navy windbreaker, a young Korean female in her early 20s with black hair and dark eyes, at three-quarter angle from the right, mid-stride boarding the vessel. Her shoe and shin occupy the lower center frame, her foot planted firmly on the wet surface. Her upper body leans diagonally into the frame from the top edge, eyes looking down at her footing. Focus on her suspended action as she transfers her weight. [: Frame edges are defined by wet, scuffed wooden deck planks running away toward the dark, slick vertical side of the quay visible behind her lower leg. The inner side of the boat gunwale, exhibiting weathered rust-red and red-brown chipped paint, cuts along the right edge. Dense damp fog completely softens the background beyond the quay boundary.] Low gray ambient light reflects off the slick deck moisture, while the dim red-brown paint accents provide a restrained ominous warmth.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O03"
              }
            ],
            "source_facts": [
              "S#26 action: 배 한척이 다가오는 게 보인다. 배로 다가가는 수리영.",
              "S#26 environment: 비는 그치고 안개가 자욱하다.",
              "Shot 9 description: first foot lands on the wet boat deck, shore-to-boat transfer."
            ],
            "visual_inferences": [
              "fixed_element: weathered_fishing_boat (visibly weathered exterior with chipped rust-red/red-brown paint, slick with moisture).",
              "Due to the low camera position on the deck, the upper body is framed leaning down from above while the shoe acts as the lower visual anchor."
            ],
            "creative_decisions": [
              "Used L02 block instead of L01 to strictly describe the frame-edge immediate surroundings for this close framing.",
              "Emphasized the contrast between the cool gray fog and the ominous red-brown chipped paint on the gunwale to reinforce a grounded thriller tone."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "e91b3504cc5f547f",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "a0e6b7a70715f5ea",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 24,
        "dependency_reason": "S24_Shot4에서 설정된 낡은 어선 갑판의 질감과 조명(red-brown accents)을 연속성 요소로 유지",
        "_shot_index": 9,
        "visible_entities": [
          "C01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 24,
            "shot_index": 9
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A handheld medium close-up (MCU) at a low angle from deck level looks back toward the quay as 수리영’s first foot lands on the wet boat deck. Her shoe and shin occupy the lower center with realistic scale, while her upper body leans diagonally into the frame from above, turning the shore-to-boat transfer into a suspended action beat.",
            "lighting_mood": "Low gray ambient light reflects off the wet deck, with dim red-brown boat paint accents adding a restrained ominous warmth.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O03"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": null,
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "weathered_fishing_boat",
                "element_type": "environment_state",
                "character_name": "",
                "description": "An old, small fishing boat exhibits a visibly weathered exterior with chipped rust-red and red-brown paint accents. Its hull and deck surfaces are slick with moisture, reflecting the dim ambient light.",
                "applies_to_shots": [
                  4,
                  9
                ],
                "source_facts": [
                  "Shot 4 description: 'old fishing boat emerges bow-first', 'rust-red accent cutting faintly through the gloom'",
                  "Shot 9 description: 'wet boat deck', 'dim red-brown boat paint accents adding a restrained ominous warmth'"
                ],
                "visual_inferences": [
                  "The boat looks visibly aged, and its exposed surfaces are completely wet due to the recent heavy rain shower mentioned in the scene text."
                ],
                "creative_decisions": [
                  "Highlight the muted red-brown paint to provide a subtle, consistent color contrast against the otherwise monochromatic gray-blue foggy environment."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 24,
                "shot_index": 4,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O03",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "dae4459bde845465"
      },
      {
        "scene_index": 25,
        "heading": "S#27. 옥탑방 안 / 해질 무렵",
        "beat_title": "수리영의 좌표 확인과 혜수의 불안",
        "representative_moment": "어스름한 옥탑방에서 혜수(C02)가 불안한 표정으로 휴대폰을 귀에 댄 채 뒤쪽의 인물을 향해 시선을 돌린다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Eye-level head-and-shoulder composition with shallow depth of field, blue-gray dusk lighting.",
            "t2i_prompt": "Photorealistic cinematic still. Head-and-shoulder composition, occupying the center-right. C02O05 in a charcoal gray suit jacket, Korean woman in her early 40s, at three-quarter angle from the right, eyes fixed on a figure deep in the room. A dark P01 smartphone is pressed to her ear, the dark screen side angled away from the camera, only its edge visible along the left foreground. Her face reflects damp-eyed anxiety. In the blurred background, an adult Korean man in his 40s stands at three-quarter angle from the left, leaning out of focus. Blue-gray dusk light from the off-camera window mixes with weak interior practical spill, falling across the side of her face and the phone edge. [: dim shadowed frame-edge wall, hollow ambient]. Focus on C02O05's anxious demeanor.",
            "outfit_assignments": [
              {
                "character_id": "C02",
                "outlook_id": "O05"
              }
            ],
            "source_facts": [
              "S#27 action: 혜수가 수리영의 녹음메시지를 확인하며 김형사에게 불안하게 질문함",
              "shot description: phone pressed to ear, anxious face in three-quarter view, eyes toward the other figure"
            ],
            "visual_inferences": [
              "C02 is holding P01 pressed to her ear, keeping the screen angled away",
              "The unregistered detective is an out-of-focus background figure sharing the same room anchor",
              "Blue-gray dusk mixes with weak practical spill, establishing the dim lighting mood"
            ],
            "creative_decisions": [
              "Head-and-shoulder composition chosen to avoid floating face artifact while maintaining close framing",
              "Shallow depth of field effectively isolates C02 while reducing the background figure to a soft presence"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "a087a47482534257",
              "owned_hash": "47c60fddbcd37070",
              "camera_direction_hash": "40c254c2ecc75eb4",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 23,
        "dependency_reason": "exact_background (background reference is intentionally skipped at composition time due to close framing rules)",
        "_shot_index": 1,
        "visible_entities": [
          "C02",
          "P01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 25,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "A slow dolly-in close-up (CU) at eye-level sits just off 혜수’s phone side, catching her face in three-quarter view as her eyes turn toward 김형사. The phone edge remains barely visible along the left foreground, while 김형사 is a soft presence deeper in the dim rooftop room, making her anxious look define the screen direction.",
            "lighting_mood": "Blue-gray dusk from the window mixes with weak interior practical spill, leaving 혜수’s face damp-eyed and anxious against a hollow, shadowed room.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C02",
              "P01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C02",
                "outlook_id": "O05"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L04B11",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [
              {
                "scene_index": 23,
                "shot_index": 11,
                "ref_usage": "exact_background",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C02O05",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "958a6e99971d01ab"
      },
      {
        "scene_index": 26,
        "heading": "바다 + 갑판 / 해질 무렵",
        "beat_title": "인우의 당황",
        "representative_moment": "C05가 내밀어진 낡은 사진 모서리를 보고 눈을 크게 뜨며 당황한 순간",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Tight eye-level composition from an observer's perspective, shallow depth of field, dim amber sunset mixed with cold sea-gray fill",
            "t2i_prompt": "Photorealistic cinematic still. C05O10 in a worn waterproof fisher jacket, Korean, mid-20s male with young male face, black hair, and dark eyes, positioned at a three-quarter angle from the left, head-and-shoulder composition with his face occupying the center-right of the frame. His widened eyes and tightly sealed mouth are set against the dim light, his gaze fixed on the old photograph. In the left foreground, the blurred edge of a creased old black-and-white photograph occupies less than one-third of the image, angled toward the camera with its worn corner and silvered emulsion edge visibly held up. Dim amber sunset mixes with a cold sea-gray fill, washing over his cheek and leaving his eyes wet and shadowed, while the photograph's edge catches a dull highlight. [: Dim, humid sunset marine atmosphere, fog-softened cold sea-gray haze, and wet deck gloom softly frame the background behind his shoulder].",
            "outfit_assignments": [
              {
                "character_id": "C05",
                "outlook_id": "O10"
              }
            ],
            "source_facts": [
              "Shot 6: 낡은 흑백 사진을 마주한 채 눈을 크게 뜨고 입술을 굳게 다문 인우의 당황한 얼굴 클로즈업.",
              "Scene heading: 바다 + 갑판 / 해질 무렵"
            ],
            "visual_inferences": [
              "fixed_element sunset_marine_atmosphere: humid sunset merging with cold sea-gray haze",
              "fixed_element rain_slick_deck: wet deck gloom",
              "The old photograph acts as a near-foreground occluder, crowding his space"
            ],
            "creative_decisions": [
              "Placed the blurred photograph edge prominently in the left foreground to establish the forward zoom target for Shot 8.",
              "Used a head-and-shoulder composition rather than a strict face-only framing to prevent floating-head synthesis issues.",
              "Applied 'washing over' terminology for the lighting to avoid slice/cut triggers on the face."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "a5ae06c4ce62b11a",
              "owned_hash": "42cfab1203ec82d1",
              "camera_direction_hash": "a2d1544025c478c0",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "",
        "_shot_index": 6,
        "visible_entities": [
          "C05"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 26,
            "shot_index": 6
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Tight at Inwoo's face from just beside the old photograph, this extreme close-up (ECU) at eye-level compresses his widened eyes and sealed mouth into the center-right of frame while the blurred photo edge slices up the left foreground at less than one-third of the image. The camera is an observer's direct vision, holding the dolly-in pressure as if the photograph itself is crowding him, with only wet deck gloom and sunset haze behind his cheek.",
            "lighting_mood": "Dim amber sunset mixes with cold sea-gray fill, leaving Inwoo's eyes wet and shadowed while the photograph edge catches a dull, ghostly highlight.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C05"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C05",
                "outlook_id": "O10"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L15B03",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "sunset_marine_atmosphere",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The surrounding environment is characterized by a dim, humid sunset merging with a cold sea-gray haze. Weak amber and orange-gray light softly illuminates the damp air, creating a moody and isolated marine atmosphere. The background remains enveloped in a muted grayish tone reflecting the fading light over the ocean.",
                "applies_to_shots": [
                  6,
                  8,
                  11
                ],
                "source_facts": [
                  "Scene heading: 바다 + 갑판 / 해질 무렵 (Sea + deck / around sunset)",
                  "Shot 6: Dim amber sunset mixes with cold sea-gray fill, wet deck gloom and sunset haze",
                  "Shot 8: Low, humid sunset light, shallow-focus gray",
                  "Shot 11: dim sea glow, weak orange-gray sunset backlights rain"
                ],
                "visual_inferences": [
                  "The combination of the setting sun and sea spray creates a persistently hazy, damp air quality across the entire scene."
                ],
                "creative_decisions": [
                  "Using the contrasting warm amber and cool sea-gray tones to heighten the emotional tension and sense of isolation on the small boat."
                ],
                "confidence": "high"
              },
              {
                "element_id": "rain_slick_deck",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The floor of the small boat's deck is noticeably slick with rainwater and sea spray, presenting a highly reflective, glossy surface under the faint light. The dark, weathered texture of the deck emphasizes the wet, cold, and uncomfortable conditions of the exterior environment.",
                "applies_to_shots": [
                  6,
                  11
                ],
                "source_facts": [
                  "Shot 6: wet deck gloom",
                  "Shot 11: rain-slick deck, rain on their faces"
                ],
                "visual_inferences": [
                  "The deck surface is likely made of aged, dark material that heavily reflects the ambient sunset light when coated in water."
                ],
                "creative_decisions": [
                  "Emphasizing the glossy, wet texture of the deck floor to visually anchor the characters in a harsh physical space before they move inside."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [
              {
                "scene_index": 26,
                "shot_index": 8,
                "description": "사진 모서리에 작게 적힌 '금월'이라는 글씨 클로즈업.",
                "keep_elements": [
                  {
                    "label": "the blurred edge of the creased old black-and-white photograph",
                    "kind": "static_prop"
                  },
                  {
                    "label": "wet deck gloom with sunset haze in the background",
                    "kind": "environment"
                  }
                ]
              }
            ],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C05O10",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "50ae1d03469b051c"
      },
      {
        "scene_index": 26,
        "heading": "바다 + 갑판 / 해질 무렵",
        "beat_title": "사진 속 단서 확인",
        "representative_moment": "An extreme close-up of the creased corner of P02 revealing the handwritten word '금월' under dim sunset light.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Extreme close-up (ECU) macro insert, slight Dutch angle, shallow depth of field with amber and sea-gray tones",
            "t2i_prompt": "Photorealistic cinematic still. An extreme macro close-up focus on the bottom corner of P02, the same old black-and-white photograph. Held almost parallel to the camera at a slight Dutch angle, the creased, water-spotted, and yellowed paper fibers are sharply visible. The tiny handwritten text '금월' occupies the central band of the frame beside faded photographic grain. A dirty, worn white margin frames the front edge. The rest of the framing falls away into a shallow-focus cold sea-gray haze. Low, humid amber sunset light skims across the paper texture, catching the salt stains and making the ink appear deep red-brown against the muted grayish environment.",
            "outfit_assignments": [],
            "source_facts": [
              "Shot 8 description: 사진 모서리에 작게 적힌 '금월'이라는 글씨 클로즈업.",
              "Scene heading: 바다 + 갑판 / 해질 무렵"
            ],
            "visual_inferences": [
              "fixed_element sunset_marine_atmosphere: weak amber and orange-gray light softly illuminates the damp air",
              "macro lens compression blurs the marine background into abstract sea-gray tones"
            ],
            "creative_decisions": [
              "Extreme macro close-up parallel to the surface to emphasize the physical texture of the paper, paper fibers, and salt stains",
              "Use of skimming warm amber sunset light against cool sea-gray ambient to isolate the clue and heighten tension"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "cd0ea14e34a9059f",
              "owned_hash": "42cfab1203ec82d1",
              "camera_direction_hash": "c8a39e3e3625535a",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 26,
        "dependency_reason": "zoom_in_detail from previous shot in the same scene, preserving lighting, marine atmosphere, and the exact physical texture of the photograph",
        "_shot_index": 8,
        "visible_entities": [
          "P02"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 26,
            "shot_index": 8
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "An extreme close-up (ECU) macro insert holds almost parallel to the photograph surface at a slight Dutch angle, isolating the tiny handwritten '금월' on the lower corner while the paper fibers and salt stains remain visible. The clue-bearing corner occupies the central band of frame, with the rest of the faded image falling away into shallow-focus gray.",
            "lighting_mood": "Low, humid sunset light skims across the paper texture, making the ink look almost red-brown against the monochrome photograph.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "P02"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L15B03",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "sunset_marine_atmosphere",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The surrounding environment is characterized by a dim, humid sunset merging with a cold sea-gray haze. Weak amber and orange-gray light softly illuminates the damp air, creating a moody and isolated marine atmosphere. The background remains enveloped in a muted grayish tone reflecting the fading light over the ocean.",
                "applies_to_shots": [
                  6,
                  8,
                  11
                ],
                "source_facts": [
                  "Scene heading: 바다 + 갑판 / 해질 무렵 (Sea + deck / around sunset)",
                  "Shot 6: Dim amber sunset mixes with cold sea-gray fill, wet deck gloom and sunset haze",
                  "Shot 8: Low, humid sunset light, shallow-focus gray",
                  "Shot 11: dim sea glow, weak orange-gray sunset backlights rain"
                ],
                "visual_inferences": [
                  "The combination of the setting sun and sea spray creates a persistently hazy, damp air quality across the entire scene."
                ],
                "creative_decisions": [
                  "Using the contrasting warm amber and cool sea-gray tones to heighten the emotional tension and sense of isolation on the small boat."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 26,
                "shot_index": 6,
                "ref_usage": "zoom_in_detail",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "prop",
                "id": "P02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [
            {
              "contract_id": "rc_001",
              "scope": {
                "scene_index": 26,
                "shot_indices": [
                  8
                ],
                "duration": "single_shot"
              },
              "targets": [
                {
                  "entity_id": "P02",
                  "role": "visual_target"
                }
              ],
              "requirements": [
                {
                  "dimension": "visual_identity",
                  "operation": "preserve",
                  "strength": "required",
                  "reference_policy": "use_entity_reference"
                }
              ]
            }
          ],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "166d36534b8d704c"
      },
      {
        "scene_index": 26,
        "heading": "S#28. 바다 + 갑판 / 해질 무렵",
        "beat_title": "조타실을 향한 돌진과 차단",
        "representative_moment": "조타실 문틀을 잡고 인우()의 진입을 몸으로 막아선 수리영(C01).",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Chest-height framing from inside the wheelhouse, utilizing the dark doorjambs as a tight border, with cool interior shadows and faint red accents contrasting the weak orange-gray sunset backlight.",
            "t2i_prompt": "Photorealistic cinematic still. Framed at chest height from just inside the wheelhouse from the reference, looking outward. The dark doorway from the reference forms a hard vertical border on the left and right edges. C01O03 in a dark navy windbreaker, an East Asian woman in her early 20s with dark hair and dark eyes, stands facing the camera in the center-left, her hand clamped firmly on the doorway from the reference. She physically blocks an East Asian man in his 40s in a damp work jacket, who is positioned at three-quarter angle from the right. The man's shoulder is driven forward against her blocking arm, his eyes fixed on C01O03, as they share the narrow threshold. C01O03's eyes are fixed on the man. Focus on the constrained physical tension between the two bodies at the doorway. Behind them, the deck from the reference is highly reflective and slick with rainwater. A dim, humid sunset merges with a cold sea-gray haze over the water from the reference. Cool interior shadow wraps their foreground bodies, while a weak orange-gray ambient sunset glow backlights the damp air and falling rain around them. A faint ominous red glimmer catches the inner edge of the doorway from the reference.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O03"
              }
            ],
            "source_facts": [
              "S#28 action: '인우, 발작적으로 일어나 조타실로 달려가 뱃머리를 돌리려 한다. 이를 눈치 챈 수리영, 조타실로 따라들어가 막으면서'",
              "Shot 11 description: 조타실 문틀을 한 손으로 꽉 쥔 채 인우의 어깨를 막아선 수리영의 정면 구도.",
              "DP Instruction: C01 facing_camera, hand clamped doorframe, eyes→인우 / In-woo three_quarter_right, shoulder driven forward, eyes→수리영"
            ],
            "visual_inferences": [
              "fixed_element rain_slick_deck: deck floor is highly reflective, slick with rainwater",
              "fixed_element sunset_marine_atmosphere: dim humid sunset, cold sea-gray haze, weak amber/orange-gray light",
              "In-woo is an unregistered character, visually inferred as an East Asian man in his 40s based on world rules and context."
            ],
            "creative_decisions": [
              "Using the doorway from the reference as a natural framing device to emphasize the constrained physical space and tension of the confrontation.",
              "Applying cool interior shadows and a faint ominous red glimmer against the ambient sunset backlight to heighten the thriller mood and physical block."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "ac4962aa9c9735e4",
              "owned_hash": "42cfab1203ec82d1",
              "camera_direction_hash": "96866822470c2087",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "wheelhouse",
                  "violating_phrase": "from just inside the wheelhouse from the reference",
                  "reason": "The phrase references the existing wheelhouse from the reference rather than instructing a new wheelhouse to be created.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "doorway",
                  "violating_phrase": "The dark doorway from the reference",
                  "reason": "The doorway is explicitly referenced as already present from the reference, so it is an anchor reference, not a redraw instruction.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "deck",
                  "violating_phrase": "the deck from the reference",
                  "reason": "Deck is mentioned as part of the existing reference environment, with no instruction to add or remake it.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "water",
                  "violating_phrase": "over the water from the reference",
                  "reason": "Water is used as an existing background element from the reference and is only being referenced spatially.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 26,
        "dependency_reason": "Relies on the established sunset marine atmosphere and wet deck environment from earlier shots in the same scene.",
        "_shot_index": 11,
        "visible_entities": [
          "C01",
          "L15"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 26,
            "shot_index": 11
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "From just inside the pilothouse doorway at chest height, the stopped track resolves into a medium wide shot (MWS), eye-level, framed by the dark doorjambs like a hard frontal collision. 수리영 occupies the center-left with one hand clamped to the frame, blocking 인우's shoulder on the right as the rain-slick deck and dim sea glow recede behind them.",
            "lighting_mood": "Cool interior shadow traps the foreground while a weak orange-gray sunset backlights rain on their faces, with a faint ominous red glimmer bleeding from inside the cabin.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01",
              "L15"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O03"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L15B02",
            "reference_usage": "exact_background",
            "owned_objects": [
              "cleat",
              "crate",
              "deck",
              "doorway",
              "fish crate",
              "gate",
              "hatch",
              "photograph",
              "railing",
              "rope",
              "utility light",
              "water",
              "wheelhouse",
              "window"
            ],
            "camera_reference": {
              "camera_position": "just aft of number 3 on the open deck, facing into the narrow wheelhouse doorway choke point between number 3 and number 2",
              "camera_height": "eye-level standing ~1.6m",
              "lens_hint": "35mm wide angle",
              "framing_notes": "emphasize the constrained doorway width and nearby railings numbers 6 and 7 for physical blocking during confrontation"
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "when focusing on an owned object, anchor it explicitly: 'the door from the reference'",
              "camera_reference specifies the framing under which the background was painted — match camera_position, camera_height, and framing",
              "if deviating from the reference framing (push-in, angle shift, ECU), state the deviation explicitly so the compositor can account for the scale difference"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "sunset_marine_atmosphere",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The surrounding environment is characterized by a dim, humid sunset merging with a cold sea-gray haze. Weak amber and orange-gray light softly illuminates the damp air, creating a moody and isolated marine atmosphere. The background remains enveloped in a muted grayish tone reflecting the fading light over the ocean.",
                "applies_to_shots": [
                  6,
                  8,
                  11
                ],
                "source_facts": [
                  "Scene heading: 바다 + 갑판 / 해질 무렵 (Sea + deck / around sunset)",
                  "Shot 6: Dim amber sunset mixes with cold sea-gray fill, wet deck gloom and sunset haze",
                  "Shot 8: Low, humid sunset light, shallow-focus gray",
                  "Shot 11: dim sea glow, weak orange-gray sunset backlights rain"
                ],
                "visual_inferences": [
                  "The combination of the setting sun and sea spray creates a persistently hazy, damp air quality across the entire scene."
                ],
                "creative_decisions": [
                  "Using the contrasting warm amber and cool sea-gray tones to heighten the emotional tension and sense of isolation on the small boat."
                ],
                "confidence": "high"
              },
              {
                "element_id": "rain_slick_deck",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The floor of the small boat's deck is noticeably slick with rainwater and sea spray, presenting a highly reflective, glossy surface under the faint light. The dark, weathered texture of the deck emphasizes the wet, cold, and uncomfortable conditions of the exterior environment.",
                "applies_to_shots": [
                  6,
                  11
                ],
                "source_facts": [
                  "Shot 6: wet deck gloom",
                  "Shot 11: rain-slick deck, rain on their faces"
                ],
                "visual_inferences": [
                  "The deck surface is likely made of aged, dark material that heavily reflects the ambient sunset light when coated in water."
                ],
                "creative_decisions": [
                  "Emphasizing the glossy, wet texture of the deck floor to visually anchor the characters in a harsh physical space before they move inside."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 26,
                "shot_index": 6,
                "ref_usage": "atmosphere_reference",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O03",
                "policy": "required"
              },
              {
                "kind": "background",
                "id": "L15B02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "34f56ab706d800b9"
      },
      {
        "scene_index": 27,
        "heading": "S#29. 조타실 / 해질 무렵",
        "beat_title": "배를 돌리려는 인우와 그를 강하게 막아서는 수리영",
        "representative_moment": "조타실 방향키를 돌리려는 인우의 손목을 수리영이 강하게 낚아채어 막는 손 클로즈업.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Very close to the helm at wrist height, low angle close-up, shallow depth of field, stark amber and cold blue-gray lighting contrast",
            "t2i_prompt": "Photorealistic cinematic still. Low angle at wrist height beside the helm. Focus on the tense gripping hand of a young Korean woman in a dark navy windbreaker, her fingers forcefully locked around the twisting wrist of a mid-20s Korean man in a worn waterproof fisher suit. The hands collide in the lower center of the frame over a dark, damp, salt-stained metal helm housing. The side rim of P17, a worn, wet-finger smudged steering wheel, cuts diagonally through the right side as a cramped mechanical barrier. A sliver of a dark, slightly worn horizontal control console surface anchors the bottom edge beneath the hands. A sickly, muddy amber practical light grazes their knuckles, contrasting sharply against the cold blue-gray dusk gloom spilling from [: a fogged, rain-specked window surface in the soft out-of-focus background].",
            "outfit_assignments": [],
            "source_facts": [
              "S#29 action: 인우는 갑자기 배를 돌리려 하고, 수리영은 막아선다.",
              "shot description: 방향키를 돌리려는 인우의 손목을 강하게 붙잡은 수리영의 손 클로즈업."
            ],
            "visual_inferences": [
              "fixed_element wheelhouse_dusk_lighting: muddy amber practical from the console and cold blue-gray window light",
              "fixed_element dark_boat_console: partially illuminated shadowy base for the interaction",
              "helm housing texture: salt-stained, dark, damp under the physical struggle"
            ],
            "creative_decisions": [
              "Triggered 'Focus on' phrasing to apply id_policy.body_part_focus_rule, substituting C## IDs with common nouns and demographic descriptors to prevent floating head artifacts in the hand close-up.",
              "Used the steering wheel rim to diagonally crop the right frame, creating a claustrophobic mechanical barrier that emphasizes the tight, locked struggle.",
              "Adhered to close framing rules by blurring the background window surface via [L01: ...] block and skipping any mention of the larger room layout."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "9a484b38d91b0726",
              "owned_hash": "7a1f375a15beb5d2",
              "camera_direction_hash": "d9887f5067e9415d",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "Not directly dependent on a preceding visual scene for background layout due to the extreme close-up framing, but maintaining lighting continuity via fixed_elements.",
        "_shot_index": 2,
        "visible_entities": [
          "C01",
          "C05",
          "P17"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 27,
            "shot_index": 2
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Very close to the helm at wrist height, this close-up (CU), low angle tightens beside the steering wheel so only 인우's tense wrist, 수리영's gripping hand, and a sliver of the control area remain visible. The dolly-in pressure makes the hands collide in the lower center of frame, while the wheel rim cuts diagonally through the right side as a cramped mechanical barrier.",
            "lighting_mood": "A muddy amber practical from the console grazes their knuckles, while cold blue-gray window light leaves the cramped cabin humid and bruised.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01",
              "C05",
              "P17"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O03"
              },
              {
                "character_id": "C05",
                "outlook_id": "O10"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L16B03",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "wheelhouse_dusk_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The cramped interior of the boat cabin is defined by a striking dual-tone lighting environment. A sickly, muddy amber practical light from the control area casts a weak, yellowish glow over the nearby surfaces and subjects. This warm artificial light contrasts sharply against the cold, damp blue-gray gloom of dusk spilling in from the windows, creating a humid and oppressive atmosphere.",
                "applies_to_shots": [
                  2,
                  5,
                  8
                ],
                "source_facts": [
                  "Shot 2: A muddy amber practical from the console... cold blue-gray window light",
                  "Shot 5: lit by a weak yellow cabin practical and cold window spill",
                  "Shot 8: Sickly cabin yellow and cold dusk blue"
                ],
                "visual_inferences": [
                  "The consistent contrast between the interior yellow practical light and the exterior blue dusk gloom is the primary environmental characteristic visually linking these disparate shots."
                ],
                "creative_decisions": [
                  "Using the stark warm/cold color contrast to unify the spatial continuity and emotional tension across the different framing scales and camera angles."
                ],
                "confidence": "high"
              },
              {
                "element_id": "dark_boat_console",
                "element_type": "persistent_prop",
                "character_name": "",
                "description": "A dark, slightly worn horizontal control console surface located near the helm inside the boat cabin. It is partially illuminated by the nearby amber practical light, providing a flat, shadowy base for interactions and objects.",
                "applies_to_shots": [
                  2,
                  5
                ],
                "source_facts": [
                  "Shot 2: a sliver of the control area remain visible",
                  "Shot 5: The photograph lies across the dark console"
                ],
                "visual_inferences": [
                  "The 'control area' near the steering wheel in Shot 2 and the 'dark console' supporting the photograph in Shot 5 refer to the same physical structure anchoring the close-up actions."
                ],
                "creative_decisions": [
                  "Maintaining a consistent dark, worn texture for the console surface to ground the tight hand-level shots in the same spatial context."
                ],
                "confidence": "medium"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "skipped_by_policy",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "55e166f83012f0fe"
      },
      {
        "scene_index": 27,
        "heading": "S#29. 조타실 / 해질 무렵",
        "beat_title": "사진 속 아이를 지목하는 수리영",
        "representative_moment": "낡은 사진 속 두 소녀 중 한 명을 손가락으로 강하게 짚은 수리영의 손 클로즈업.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Overhead high angle close-up, striking dual-tone amber and blue-gray lighting.",
            "t2i_prompt": "Photorealistic cinematic still. Overhead high angle close-up from above. Focus on a young Korean woman's index finger extending from the cuff of a dark navy windbreaker, pressing forcefully onto the surface of P02, an old, creased, damp black-and-white photograph. P02 lies across [: a dark, scratched, salt-stained horizontal console surface], its rectangular edge tilted slightly across the frame. The printed content of P02 is fully visible, showing the faces of two young girls; the fingertip presses exactly onto the printed face of one girl in the lower-right foreground. At the upper edge of the frame, a cropped, dusty round gauge face peeks in, its numbers barely visible. A weak, sickly muddy amber practical light washes the console surface, contrasting sharply with a cold blue-gray dusk spill that turns the monochrome image green-gray. Condensed, humid cabin air creates a soft, hazy texture around the hand.",
            "outfit_assignments": [],
            "source_facts": [
              "S#29 action: 수리영 points to a child in the photo (이 사진속의 얘가 우리 엄마예요).",
              "Shot 5 description: Close above the old photograph at hand level... 수리영's finger presses one of the two girls in the image near the lower-right foreground."
            ],
            "visual_inferences": [
              "fixed_element wheelhouse_dusk_lighting: sickly amber practical contrasting with cold blue-gray dusk spill.",
              "fixed_element dark_boat_console: the photograph rests on a dark, slightly worn horizontal console.",
              "Per id_policy.body_part_focus_rule and reproduction_surface_rule, full character ID C01 is omitted for the hand close-up, replaced with demographic descriptor."
            ],
            "creative_decisions": [
              "Strict overhead framing to anchor the hand and photograph against the worn console top, excluding background distractions.",
              "Using the dual-tone (amber/blue-gray) lighting contrast to amplify the eerie, humid tension in the tight composition."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "70df32c6b261689d",
              "owned_hash": "cc274d34a2a21947",
              "camera_direction_hash": "2279388094c678ad",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "No explicit scene-level dependency from prior scenes; continuity is anchored within the current scene's previous shots.",
        "_shot_index": 5,
        "visible_entities": [
          "C01",
          "P02"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 27,
            "shot_index": 5
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Close above the old photograph at hand level, this close-up (CU), high angle holds completely still as 수리영's finger presses one of the two girls in the image near the lower-right foreground. The photograph lies across the dark console, its rectangular edge tilted slightly across the frame, while her fingertip creates the only forceful movement in an otherwise frozen evidentiary composition.",
            "lighting_mood": "The photo is lit by a weak yellow cabin practical and cold window spill, with the old monochrome image turning green-gray under the damp coastal gloom.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C01",
              "P02"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O03"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L16B01",
            "reference_usage": "skipped_close_framing",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": true,
            "constraints": [
              "this shot uses close framing — the background reference is intentionally omitted at composition time",
              "do not mention the existing room, the chain_bg reference image, or any item from the owned_objects list",
              "do not use any of: 'the existing X', 'from the reference', 'use the X from the reference', 'preserving the same room perspective', 'maintaining the reference's framing', 'do not generate a new X' — these phrases are invalid when the reference is absent",
              "do not use any of: 'matching the reference camera', 'deviation from reference', 'same … angle as the reference', 'match the reference framing' — Rule A close-framing invalidation: close framing has no reference camera to match or deviate from",
              "compose the close subject from its immediate surroundings only — use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "wheelhouse_dusk_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The cramped interior of the boat cabin is defined by a striking dual-tone lighting environment. A sickly, muddy amber practical light from the control area casts a weak, yellowish glow over the nearby surfaces and subjects. This warm artificial light contrasts sharply against the cold, damp blue-gray gloom of dusk spilling in from the windows, creating a humid and oppressive atmosphere.",
                "applies_to_shots": [
                  2,
                  5,
                  8
                ],
                "source_facts": [
                  "Shot 2: A muddy amber practical from the console... cold blue-gray window light",
                  "Shot 5: lit by a weak yellow cabin practical and cold window spill",
                  "Shot 8: Sickly cabin yellow and cold dusk blue"
                ],
                "visual_inferences": [
                  "The consistent contrast between the interior yellow practical light and the exterior blue dusk gloom is the primary environmental characteristic visually linking these disparate shots."
                ],
                "creative_decisions": [
                  "Using the stark warm/cold color contrast to unify the spatial continuity and emotional tension across the different framing scales and camera angles."
                ],
                "confidence": "high"
              },
              {
                "element_id": "dark_boat_console",
                "element_type": "persistent_prop",
                "character_name": "",
                "description": "A dark, slightly worn horizontal control console surface located near the helm inside the boat cabin. It is partially illuminated by the nearby amber practical light, providing a flat, shadowy base for interactions and objects.",
                "applies_to_shots": [
                  2,
                  5
                ],
                "source_facts": [
                  "Shot 2: a sliver of the control area remain visible",
                  "Shot 5: The photograph lies across the dark console"
                ],
                "visual_inferences": [
                  "The 'control area' near the steering wheel in Shot 2 and the 'dark console' supporting the photograph in Shot 5 refer to the same physical structure anchoring the close-up actions."
                ],
                "creative_decisions": [
                  "Maintaining a consistent dark, worn texture for the console surface to ground the tight hand-level shots in the same spatial context."
                ],
                "confidence": "medium"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "prop",
                "id": "P02",
                "policy": "required"
              }
            ],
            "forbidden_refs": [
              {
                "kind": "background",
                "reason": "close framing skips chain_bg reference"
              }
            ],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [
            {
              "contract_id": "rc_001",
              "scope": {
                "scene_index": 27,
                "shot_indices": [
                  5
                ],
                "duration": "single_shot"
              },
              "targets": [
                {
                  "entity_id": "P02",
                  "role": "visual_target"
                }
              ],
              "requirements": [
                {
                  "dimension": "visual_identity",
                  "operation": "preserve",
                  "strength": "required",
                  "reference_policy": "use_entity_reference"
                }
              ]
            }
          ],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": true,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "aa134303e22d8e5d"
      },
      {
        "scene_index": 27,
        "heading": "S#29. 조타실 / 해질 무렵",
        "beat_title": "조타실 벽면의 기괴한 붉은 원형 표식",
        "representative_moment": "C05 stands in silhouette in the left foreground, looking toward the rear wall densely covered with P08 red circular markings.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Eye-level over-the-shoulder reveal with sickly amber and cold blue dusk dual-tone contrast",
            "t2i_prompt": "Photorealistic cinematic still. Eye-level over-the-shoulder composition inside the shared space of the cramped wheelhouse. C05O10 in a worn waterproof jumper, East Asian young male in his mid-20s with black hair, occupies the left foreground. He stands with his back to the camera, a sliver of his side profile visible to reveal his jawline, shoulders locked rigid. He appears as a heavily shadowed backlit figure, body volume preserved against the dim ambient fill, with sickly muddy amber practical light reflecting on the edge of his shoulder. Focus on the P08 hand-drawn red circular markings densely covering the marked wall panel from the reference in the middle distance. The room maintains a striking dual-tone atmosphere: the warm yellowish glow from the small cabin lamp and console contrasts sharply against the cold, damp blue-gray gloom of dusk washing through the fogged side window glass from the reference. The P08 red circles remain the only saturated color pulsing against the dark, humid shadows.",
            "outfit_assignments": [
              {
                "character_id": "C05",
                "outlook_id": "O10"
              }
            ],
            "source_facts": [
              "S#29 action: '인우 얼굴 너머에 보이는 조타실안 수많은 원. 빙글빙글 돌고 여기에서 또다시 발견된 붉은 원!'",
              "Shot 8 description: 인우의 등 뒤, 어두운 조타실 벽면을 빼곡하게 채운 기괴한 붉은 원형 표식들의 정지 화면."
            ],
            "visual_inferences": [
              "fixed_element wheelhouse_dusk_lighting: sickly amber practical light contrasting against cold blue-gray dusk gloom.",
              "Inwoo's posture is locked and rigid, rendered as a backlit silhouette while preserving body volume to emphasize the shock.",
              "The wall panel from the reference is explicitly anchored to serve as the background for the red markings."
            ],
            "creative_decisions": [
              "Used an over-the-shoulder composition to physically separate the observer from the background wall while anchoring Inwoo in the foreground.",
              "Included a sliver of his side profile to comply with the face identifiability rules for entity ID binding, while fully preserving the DP's 'back to camera' instruction.",
              "Saturated the red color of the P08 markings specifically to create a visual contrast against the muted, humid coastal gloom of the background."
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "fdc7d26f73dbc181",
              "owned_hash": "cc274d34a2a21947",
              "camera_direction_hash": "72352fd2c28fe81a",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "wall panel",
                  "violating_phrase": "marked wall panel from the reference",
                  "reason": "The phrase explicitly references the existing wall panel from the reference, so it is an anchor rather than a request to redraw it.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "window",
                  "violating_phrase": "fogged side window glass from the reference",
                  "reason": "The window is mentioned as part of the referenced existing environment and used as a background anchor, not as a new object to generate.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "No direct scene dependency provided in the context; spatial continuity elements (wheelhouse_dusk_lighting) are applied via fixed_elements rather than cross-scene dependency.",
        "_shot_index": 8,
        "visible_entities": [
          "C05",
          "P08",
          "L16"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 27,
            "shot_index": 8
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "After sliding behind 인우, the camera stops at shoulder height in a medium shot (MS), eye-level, using an over-the-shoulder reveal: his dark back anchors the left foreground while the rear wall fills the middle distance with dense red circular markings. The frame freezes into direct observer vision, keeping 인우 mostly silhouette and letting the occult pattern stare back from the dim wheelhouse wall.",
            "lighting_mood": "Sickly cabin yellow and cold dusk blue leave the room nearly black, while the red circles pulse as the only saturated color against the humid shadows.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "C05",
              "P08",
              "L16"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C05",
                "outlook_id": "O10"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L16B01",
            "reference_usage": "exact_background",
            "owned_objects": [
              "GPS",
              "bench locker",
              "chair",
              "console",
              "door",
              "fuse panel",
              "helm wheel",
              "marking",
              "photo",
              "radio",
              "switch panel",
              "throttle",
              "wall panel",
              "window",
              "windshield"
            ],
            "camera_reference": {
              "camera_position": "near number 13 (storage bench) in the aft standing zone, facing diagonally toward number 11 (marked wall panel) and number 12 (old photo clue), with number 7 (helm wheel and throttle) peripheral in the foreground",
              "camera_height": "eye-level standing ~1.55m",
              "lens_hint": "35mm moderate wide angle",
              "framing_notes": "include number 11 and number 12 prominently; keep the wheelhouse visibly cramped"
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "when focusing on an owned object, anchor it explicitly: 'the door from the reference'",
              "camera_reference specifies the framing under which the background was painted — match camera_position, camera_height, and framing",
              "if deviating from the reference framing (push-in, angle shift, ECU), state the deviation explicitly so the compositor can account for the scale difference"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "wheelhouse_dusk_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The cramped interior of the boat cabin is defined by a striking dual-tone lighting environment. A sickly, muddy amber practical light from the control area casts a weak, yellowish glow over the nearby surfaces and subjects. This warm artificial light contrasts sharply against the cold, damp blue-gray gloom of dusk spilling in from the windows, creating a humid and oppressive atmosphere.",
                "applies_to_shots": [
                  2,
                  5,
                  8
                ],
                "source_facts": [
                  "Shot 2: A muddy amber practical from the console... cold blue-gray window light",
                  "Shot 5: lit by a weak yellow cabin practical and cold window spill",
                  "Shot 8: Sickly cabin yellow and cold dusk blue"
                ],
                "visual_inferences": [
                  "The consistent contrast between the interior yellow practical light and the exterior blue dusk gloom is the primary environmental characteristic visually linking these disparate shots."
                ],
                "creative_decisions": [
                  "Using the stark warm/cold color contrast to unify the spatial continuity and emotional tension across the different framing scales and camera angles."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C05O10",
                "policy": "required"
              },
              {
                "kind": "background",
                "id": "L16B01",
                "policy": "required"
              }
            ],
            "forbidden_refs": [],
            "readiness_policy": "block_if_missing",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "694420a2bc207ff3"
      },
      {
        "scene_index": 28,
        "heading": "S#30. 바다 / 해질 무렵",
        "beat_title": "안개 속으로 사라지는 배",
        "representative_moment": "짙은 안개벽 속으로 뱃머리가 반쯤 먹혀 들어가는 낡은 어선의 뒷모습을 바다 위 낮게 뜬 앵글로 포착한다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Low angle observer view directly behind the stern, muted sunset gold flattened by heavy gray fog.",
            "t2i_prompt": "Photorealistic cinematic still. Focus on the battered rear rail and dark cabin outline of an old, weathered fishing boat dissolving into a thick fog wall. Low angle observer view from directly behind at stern height. The boat's damp silhouette occupies the lower-middle frame, its bow pointing away and disappearing halfway into a heavy band of semi-opaque, blank gray fog that swallows the upper half of the image. [L17: Sunset sea enveloped by dense fog]. A churned, gray-white foaming wake trails back toward the camera in the foreground. Muted ambient lighting with cold gray fog flattening the twilight; faint bronze highlights catch on the wake, while the restless, rain-dimpled, black-green sea surface below reflects a broken smear of gold sunset and the boat's stern shadow beneath the hull.",
            "outfit_assignments": [],
            "source_facts": [
              "S#30 action: '서서히 지는 해가 바다위에 황금빛 그림자를 드리운다. 세상과 경계를 지우는 것 같은 짙은 안개 띠. 서서히 안개 안으로 들어가는 인우의 배.'",
              "shot description: 짙은 안개벽 속으로 뱃머리가 반쯤 집어삼켜진 낡은 어선의 뒷모습."
            ],
            "visual_inferences": [
              "observer point of view positioned directly behind the stern just above the water level",
              "stern silhouette providing a scale reference against the massive fog wall",
              "foaming wake indicating the mid-action forward movement of the boat"
            ],
            "creative_decisions": [
              "heavy band of gray fog occupying the upper frame to emphasize isolation and crossing a boundary",
              "rain-dimpled black-green water texture to ground the thriller/noir tone"
            ],
            "confidence": "high",
            "owned_validation": {
              "schema_version": 1,
              "t2i_prompt_hash": "3e1b594485407560",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "c26f81ddb5927bac",
              "validator": "scene_detail_owned_objects.v1",
              "violations": []
            }
          }
        ],
        "scene_type": "transition",
        "dependent_scene_index": -1,
        "dependency_reason": "none",
        "_shot_index": 3,
        "visible_entities": [
          "L17"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 28,
            "shot_index": 3
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Directly behind the old fishing boat at stern height, this wide shot (WS), low angle tracks along the centerline just above the wake as the stern silhouette occupies the lower middle of frame and the bow disappears halfway into the fog wall. The composition stays tight enough to read the boat's battered rear rail and cabin outline, but leaves a heavy band of blank gray fog swallowing the upper half of the image.",
            "lighting_mood": "The last gold of sunset is flattened by cold gray fog, turning the boat into a damp silhouette with only faint bronze highlights on the wake.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction specifies a vertical height/position (low / hip-height / overhead / ground level / quay level / floor level / from above) and the scene description includes frame-edge surfaces",
                "consistency_check_summary": "elements described inside the frame must be at vertical/depth positions consistent with the camera's height — low cameras place feet/lower-body in the lower frame, overhead cameras place head/shoulders in upper frame, hip-height cameras place torso in frame with head/feet implied beyond edges",
                "core_principles": [
                  "low cameras at ground/quay/floor level place feet and lower legs in the lower frame; chest-height frame-edge surfaces behind the subject are geometrically forbidden",
                  "overhead high angles from above place the top of head and shoulders in the upper-center frame; floor surfaces visible below feet at the lower frame edge are forbidden",
                  "hip-height side cameras place torso and arms in frame; head and feet are implied beyond frame edges, not both fully visible together",
                  "full body shots require either eye-level or high-angle framing where head and feet both fit; low cameras + full silhouette in one frame is forbidden"
                ],
                "forbidden_combinations": [
                  "low camera at ground/quay/floor level + frame-edge surface described at chest/waist height behind the subject",
                  "overhead high angle + feet visible at the lower frame edge with floor visible below them",
                  "hip-height side camera + full silhouette including head and feet visible in same frame",
                  "low angle looking up + ceiling and floor both fully visible in same frame"
                ],
                "recommended_phrasings": [
                  "low observer at ground level, shoes and lower legs occupying the lower frame, upper body extending toward the upper edge",
                  "overhead high angle from above, the top of head and shoulders fill the upper-center frame, the floor visible around them",
                  "at hip height beside the subject, torso and arms in frame, head and lower legs implied beyond frame edges",
                  "low angle looking up at the figure, ceiling lamps in the upper frame, the wall behind extending downward"
                ],
                "self_check_steps": [
                  "identify the camera position phrasing (height + angle)",
                  "list elements described inside the frame (foreground prop, body parts, background surfaces)",
                  "verify each element's vertical/depth placement is geometrically consistent with that camera position",
                  "if mismatch found, adjust frame-edge description to match camera position"
                ],
                "rationale_summary": "T2I 모델은 카메라 위치와 frame 안 요소 vertical 좌표 모순 시 절충 합성 — 인물이 가구에 박히거나 신체 일부가 잘리거나 prop 이 떠 있는 부자연스러운 결과"
              },
              "fg_bg_shared_anchor_rule": {
                "applies_when": "two characters share an interaction (dialogue / handing object / facing / co-watching) AND are described as foreground/background separated AND a shared surface or spatial anchor exists between them (bench / table / door / room)",
                "shared_anchor_required": "describe the shared surface or spatial anchor explicitly so both figures occupy the same physical space",
                "forbidden_phrasings": [
                  "fg/bg separation without naming the shared surface/space",
                  "isolated foreground hand reaching with another figure 'in the rear background' (no shared anchor named)",
                  "one figure 'in the foreground' and another 'in the back' without naming the room/surface they share"
                ],
                "recommended_keywords": [
                  "shared bench / shared table / shared surface",
                  "the same X they both occupy (X = bench, sofa, table, doorway, room)",
                  "across the X from each other",
                  "<prop> crosses the air between them (interaction prop visually links the two)",
                  "the <surface> sharing a single line of sight from the camera"
                ],
                "self_check_steps": [
                  "do the two characters share a single interaction (dialogue, handing, facing, co-watching)?",
                  "is one foreground and the other background-separated?",
                  "is the shared surface/space anchor explicitly named?",
                  "if not, add one of the recommended keywords"
                ],
                "rationale_summary": "fg/bg 분리만 묘사하면 모델이 두 인물을 다른 공간에 배치하거나 인터랙션이 끊긴 어색한 합성 — 공유 anchor 명시가 같은 물리적 공간 강제"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, insert} from `_derive_framing_scale()` (render_strategy.framing_scale, production enum). wide-class shots are mapped to medium and use `wide_medium_rules`. The LLM may *additionally* infer wide from prose keywords (`wide shot` / `전경` / `establishing` / `aerial`) for stricter checking — these are LLM-side reasoning hints, not the production framing_scale enum (RO-4 binding — card-only classification).",
                "framing_scale_keywords": {
                  "_comment": "LLM-facing reasoning hints — not the production framing_scale enum. card framing_scale is close/medium/insert from _derive_framing_scale(); wide is LLM hint only (RO-4 / RO-5 binding). Canary close-framing shot scope filter uses production _CLOSE_FRAMING_RE (9-entry alignment) — separate from this 11-entry LLM list.",
                  "close": [
                    "close-up",
                    "CU",
                    "MCU",
                    "ECU",
                    "XCU",
                    "extreme close-up",
                    "클로즈업",
                    "손가락이",
                    "손이",
                    "눈이",
                    "얼굴이"
                  ],
                  "wide": [
                    "wide shot",
                    "establishing",
                    "aerial",
                    "전경",
                    "전신"
                  ],
                  "medium_default": "medium / mid / two-shot — anything not matching close/wide keywords"
                },
                "close_framing_rules": {
                  "primary_subject_only_fully_visible": "the named primary subject (hand / face / object) is the only fully visible entity in the frame",
                  "other_entities_appearance_options": [
                    "partial visibility — shoulder slice / arm slice / sleeve / fingertips entering the frame edge",
                    "soft background — out-of-focus silhouette via shallow depth of field",
                    "absent — not rendered at all even if listed in visible_entities"
                  ],
                  "forbidden_combinations": [
                    "same character's full body and body-part close-up in the same frame (view mixing)",
                    "two characters both rendered as sharp close-up faces in the same frame"
                  ],
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "wide_medium_rules": {
                  "primary_and_secondary_both_visible": "primary subject + secondary entities all renderable",
                  "body_part_close_up_forbidden": "body-part close-up of any character is forbidden in wide/medium framing — close-up belongs in a separate shot",
                  "fg_bg_separation_requires_shared_anchor": "if two characters are rendered fg/bg separated, also apply fg_bg_shared_anchor_rule",
                  "body_part_focus_cross_ref": "see id_policy.body_part_focus_rule (G4.3 sub-field — body-part Focus 시 C##O## 금지)"
                },
                "self_check_steps": [
                  "identify framing_scale from camera_direction",
                  "if close: verify only primary subject is fully visible; other entities are partial / soft bg / absent",
                  "verify same character's full body + body-part close-up are not combined",
                  "if close: verify two faces are not both rendered sharp"
                ],
                "rationale_summary": "한 shot 안 framing scale 가 일관되어야 모델이 단일 카메라 셔터로 해석. close/wide/medium 혼합은 multi-camera 합성으로 인식되어 인물 분열 / 부위 분리 / 시점 혼합 합성"
              },
              "rationale_summary": "Rule F/G/J 의 spatial consistency 룰 통합 — camera 위치 vs frame edge 모순, fg/bg 분리 시 공유 anchor 누락, framing scale 혼합. 3 sub-rule 모두 single t2i_prompt = single camera shutter 원칙의 spatial 측면."
            },
            "constraints": [
              "one prompt captures one still moment",
              "do not combine full-body framing and body-part close-up in one frame",
              "camera position and frame-edge elements must be geometrically consistent — see render_strategy.spatial_consistency.camera_frame_rule",
              "two characters in fg/bg separation must share a named surface/space anchor — see render_strategy.spatial_consistency.fg_bg_shared_anchor_rule",
              "framing_scale binds entity rendering — close = primary subject only fully visible, wide/medium = no body-part close-up — see render_strategy.spatial_consistency.primary_framing_rule"
            ]
          },
          "id_policy": {
            "allowed_base_entity_ids": [
              "L17"
            ],
            "allowed_outlook_pairs": [],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "body-part close-up",
              "a reproduced face inside any declared reproduction surface",
              "unregistered extra"
            ],
            "id_use_required_when": [
              "a physical photo/poster/document/picture-frame/map/key prop appearing in the scene — use its P## ID and preserve canonical content from entity_canon"
            ],
            "demographic_fallback_required": true,
            "face_identifiability_rule": {
              "use_entity_id_when": [
                "face identifiable: front, profile, three-quarter, or eyes closed",
                "OTS framing with any visible face"
              ],
              "common_noun_required_when": [
                "back to camera with no visible face",
                "silhouette or blurred outline only",
                "OTS with only back of head/shoulder, face entirely hidden"
              ],
              "id_use_summary": "C##/C##O## allowed only when face is identifiable AND framing is not dominated by an isolated body part",
              "rationale_summary": "C## 사용 시 얼굴 ref 가 자동 결합 — 얼굴이 안 보이는 인물에 C## 부착하면 ref 와 본문 묘사가 충돌"
            },
            "body_part_focus_rule": {
              "trigger_phrases": [
                "focus on",
                "close on",
                "tight on",
                "detail on"
              ],
              "applies_to": "any body part (face or non-face) when the framing is dominated by that part",
              "id_use": "forbidden — use a common noun + demographic descriptor instead",
              "common_noun_examples_generic": [
                "a trembling hand",
                "an exposed forearm",
                "the furrowed brow of a figure in shadow",
                "the clenched jaw of a figure in profile"
              ]
            },
            "close_framing_face_phrasing": {
              "forbidden_phrases": [
                "face filling the entire frame",
                "face filling nearly the entire frame",
                "her face fills the frame",
                "shoulders only entering the frame",
                "head snapped sideways",
                "his face fills the frame"
              ],
              "recommended_phrasings": [
                "tight framing on her face, upper body in frame",
                "close on his expression, shoulder visible at the edge",
                "head-and-shoulder composition, face occupying the center-right",
                "face in profile at the left, neck and upper chest visible"
              ],
              "rationale_summary": "T2I 모델은 얼굴이 프레임 전체를 가득 채우는 묘사를 받으면 몸과 분리된 얼굴만 공중에 떠 있는 합성을 낸다 — 어깨/상반신 일부를 프레임 안에 유지"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — common noun + demographic descriptor only",
              "rationale_summary": "C##O## 은 얼굴 reference 이미지를 원본 해상도로 inject 하므로 사진/화면/반사 표면 안 얼굴이 표면 밖 실물 크기로 합성됨. 이 rule 은 표면 안 얼굴 (face) 에만 적용 — 물리적 사진/액자/문서 prop 자체는 P## ID 보존 (Patch A)."
            },
            "demographic_descriptor_policy": {
              "required_on_first_appearance": true,
              "applies_to_id_forms": [
                "C##",
                "C##O##"
              ],
              "token_count_range": [
                1,
                2
              ],
              "components": {
                "ethnicity": [
                  "Asian",
                  "East Asian",
                  "South Asian",
                  "Southeast Asian",
                  "Black",
                  "Middle Eastern",
                  "Hispanic",
                  "Latina",
                  "Latino",
                  "Caucasian"
                ],
                "gender": [
                  "man",
                  "woman",
                  "figure"
                ],
                "age_band": [
                  "in his/her 20s",
                  "in his/her 30s",
                  "in his/her 40s",
                  "young",
                  "middle-aged",
                  "elderly"
                ],
                "role_hint_from_outfit": [
                  "in worker uniform",
                  "in fisher workwear",
                  "in detective coat",
                  "in business suit"
                ]
              },
              "format_template_a": "C##O## in <옷 1-3 단어>, <demographic descriptor>",
              "format_template_b": "C##O##, <demographic descriptor>, <자세 표현>",
              "source_priority": [
                "visual_world_rules.region",
                "visual_world_rules.era",
                "shot context outfit hint"
              ],
              "scenario_dependency_ban": "no work-specific proper nouns (character names, place names) — common nouns only",
              "regional_consistency": "한 작품 내 region 표기는 일관 — 선택된 세분 ethnicity (East Asian / Southeast Asian / South Asian / Caucasian / Black / Hispanic / Latina|o / Middle Eastern) 를 작품 단위 고정"
            },
            "constraints": [
              "visible real people with outfit binding use C##O## composite IDs",
              "do not use C##O## for a reproduced face inside any element flagged by id_policy.reproduction_surface_rule.applies (declared reproduction surface) — use a common noun + demographic descriptor instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the common-noun rule applies only to a face reproduction, not to the prop entity.",
              "do not use C##O## for isolated body-part focus (face or non-face) — when the framing is dominated by a single body part triggered by 'focus on / close on / tight on / detail on' phrasing, use a common noun (hand / eye / brow / jaw / etc.)",
              "for close-up of a face, do not use 'face filling the entire frame' / 'face filling nearly the entire frame' / 'her face fills the frame' / 'shoulders only entering the frame' / 'head snapped sideways' — keep upper body or shoulder edge visible in the frame ('tight framing on her face, upper body in frame')",
              "when introducing a C##O## (or C##) for the first time in this t2i_prompt, append a 1-2 token demographic descriptor (ethnicity + age band, or ethnicity + role hint) consistent with visual_world_rules.region — never use scenario-specific proper nouns",
              "use the format `C##O## in <옷 1-3 단어>, <demographic descriptor>` or `C##O##, <demographic descriptor>, <자세 표현>` — main characters with face refs may include this descriptor too (image inject takes precedence)"
            ]
          },
          "background_binding": {
            "mode": "not_applicable",
            "bg_id": null,
            "reference_usage": "none",
            "owned_objects": [],
            "camera_reference": null,
            "close_framing_skips_background_ref": false,
            "constraints": [
              "no chain_bg is bound to this shot — describe the location freely without referencing a background image",
              "do not use 'from the reference' / 'the existing room' / 'preserving the room' style phrasing — there is no reference image to refer to"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [],
            "previous_shot_refs": [],
            "forward_zoom_targets": [],
            "cross_shot_id_substitution_rule": {
              "applies_when": "fixed_elements[i].character_name maps to a C## allowed in id_policy.allowed_base_entity_ids OR a C##O## in id_policy.allowed_outlook_pairs",
              "substitution": "replace the common-noun person reference inside fixed_elements[i].description (e.g. 'An Asian man' / 'a woman' / 'a figure') with the matched C## or C##O##",
              "double_description_forbidden": "do not describe the same character with both a C##/C##O## and a separate common-noun person reference in the same t2i_prompt — produces phantom extra figures",
              "preserve_pose_unchanged": "fixed_elements[i].description 의 자세/위치/상태 keyword 는 그대로 유지 — 인물 참조 토큰만 교체",
              "no_repeat_after_pose": "이미 t2i_prompt 안에서 같은 인물의 자세를 묘사했다면, 같은 fixed_element description 을 별도 문장으로 반복하지 마라",
              "rationale_summary": "fixed_element description 이 보통명사로 작성되었고 같은 인물의 C## 묘사가 t2i_prompt 에 따로 있으면 모델이 두 인물로 인식해 인원 수가 부풀려진다"
            },
            "ref_usage_constraints": {
              "zoom_in_detail": {
                "scope_summary": "same moment, same space, same primary subject; camera moved/zoomed onto a sub-region only",
                "forbidden_additions": [
                  "new persons not in the prior shot",
                  "new props not in the prior shot",
                  "new background elements not in the prior shot",
                  "new poses, gestures, or expressions not in the prior shot",
                  "different moments of action"
                ],
                "required_phrasings": [
                  "describe only the focused sub-region in concrete terms",
                  "explicitly state the same environment (lighting, background) is preserved",
                  "use continuity tokens such as 'the same figure', 'the same room', 'the same surface'"
                ],
                "body_part_cross_ref": "id_policy.body_part_focus_rule applies — when the focused sub-region is a body part, use a common noun + demographic descriptor (no C##/C##O##)",
                "rationale_summary": "zoom_in_detail = 앞 샷의 특정 영역을 확대한 '같은 사진'. 새 정보 추가 금지 — 현재 프레임이 보여주는 부분만 정확히 묘사"
              },
              "exact_background": {
                "scope_summary": "same room, possibly different moment or angle; background carries through, persons/actions are described fresh for this shot",
                "permitted_additions": [
                  "new persons, poses, expressions, actions for this shot's moment",
                  "different camera angle within the same room"
                ],
                "background_consistency_rule": "background elements established in the prior shot must remain consistent (no contradictory walls, doors, furniture)",
                "ignore_keep_handled_elsewhere": "ignore/keep directives are auto-applied via reference image processing — do not echo them inside t2i_prompt",
                "rationale_summary": "같은 공간, 새 순간 — 인물·동작은 이 샷의 instantaneous moment 로 묘사하되 배경은 prior shot 과 일관"
              },
              "atmosphere_reference": {
                "scope_summary": "different room/floor/angle entirely — only mood/tone is shared with the prior shot",
                "forbidden_imports": [
                  "furniture layout from the prior shot",
                  "wall positions from the prior shot",
                  "specific architectural features from the prior shot"
                ],
                "required_handling": "describe the new space's background directly in this shot; carry through only color tone or lighting mood (e.g. 'cool gray ambient' / 'dim warm pool')",
                "rationale_summary": "atmosphere_reference = 분위기 reference 만, 레이아웃 reference 아님. 가구·벽 배치를 끌어오면 모델이 두 공간을 한 공간으로 합성"
              }
            },
            "view_consistency": {
              "framing_scope_options": [
                "third_person",
                "close_up"
              ],
              "single_camera_rule": "one t2i_prompt = one camera position. do not combine a third-person full-body description and a close-up of the same character's body part.",
              "third_person_handling": "describe persons with full or partial body in frame; their hands/arms remain attached to their bodies (no isolated hand close-up alongside the full body)",
              "close_up_handling": "isolated body part (hand, wrist, fingers, brow) — do not include the same character's full body in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "the phrasing 'Focus on ...' specifies the area of focus while the camera position remains the same; it is NOT a view switch",
              "id_policy_cross_ref_for_body_part_focus": "id_policy.body_part_focus_rule applies — body-part Focus 시 C##O## 금지 (얼굴 ref 합성 차단)",
              "prop_orientation_rule": "props the character holds must face one direction only — toward camera OR toward character, not both. forbidden contradiction example: 'photo angled partly toward the camera' + 'eyes fixed on the photograph' (양방향 모순)",
              "mixing_forbidden": true,
              "rationale_summary": "single t2i_prompt = single camera shutter at 1/1000s. 시점 혼합은 모델이 두 카메라 합성으로 해석해 인물이 두 번 등장 또는 부위 분리"
            },
            "constraints": [
              "fixed_elements are continuity inputs derived from prior shots — treat them as binding context, not as source facts to be invented",
              "when fixed_elements[i].character_name maps to a C## in id_policy.allowed_base_entity_ids or a C##O## in id_policy.allowed_outlook_pairs, replace the common-noun person reference inside fixed_elements[i].description with that C##/C##O## and do not also write a separate common-noun reference for the same character in the same t2i_prompt",
              "previous_shot_refs with ref_usage='zoom_in_detail' must preserve the same moment, the same space, and the same lighting as the prior shot — no new persons, props, background elements, poses, or moments",
              "previous_shot_refs with ref_usage='exact_background' may freely describe this shot's persons, poses, and moment, but must keep background elements consistent with the prior shot's established room",
              "previous_shot_refs with ref_usage='atmosphere_reference' must describe a new space directly — do not import furniture layout or wall positions from the prior shot, only mood/tone",
              "one t2i_prompt = one camera position — do not combine a third-person full-body description and a close-up of the same character's body part; 'Focus on' specifies the area of focus, not a view switch (id_policy.body_part_focus_rule applies for any body-part Focus)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [],
            "forbidden_refs": [],
            "readiness_policy": "not_applicable",
            "constraints": [
              "do not imply or describe a reference image that is not listed in required_refs (no phantom references)",
              "if a required character_outlook ref is missing, scene image generation must block before the image stage"
            ]
          },
          "render_contracts": [],
          "_card_metadata": {
            "lift_status": {
              "rule_a_lifted": true,
              "rule_c_lifted": true,
              "rule_e_lifted": false,
              "rule_h_lifted": true,
              "id_policy_composite_lifted": true,
              "id_policy_close_framing_face_lifted": true,
              "id_policy_reproduction_surface_lifted": true,
              "continuity_cross_shot_id_substitution_lifted": true,
              "continuity_ref_usage_constraints_lifted": true,
              "continuity_view_consistency_lifted": true,
              "continuity_constraints_extended": true,
              "spatial_camera_frame_lifted": true,
              "spatial_fg_bg_shared_anchor_lifted": true,
              "spatial_primary_framing_lifted": true,
              "spatial_constraints_extended": true
            },
            "rule_source": {
              "camera_rule": "A",
              "owned_rule": "C",
              "close_skip_rule": "E",
              "demographic_rule": "H",
              "composite_id_rule": "C## 사용 규칙 — 참조 이미지 연동 (v17:32-64)",
              "close_framing_face_rule": "극단 클로즈업 표현 (v17:66-85)",
              "reproduction_surface_rule": "사진·포스터·화면·거울 속 인물 규칙 (v17:133-146)",
              "cross_shot_id_substitution_rule": "v20/scene_detail/system.md#continuity-section",
              "ref_usage_rule": "v20/scene_detail/system.md#continuity-section",
              "view_consistency_rule": "v20/scene_detail/system.md#continuity-section",
              "camera_frame_rule": "v20/scene_detail/system.md#spatial-consistency",
              "fg_bg_shared_anchor_rule": "v20/scene_detail/system.md#spatial-consistency",
              "primary_framing_rule": "v20/scene_detail/system.md#spatial-consistency"
            }
          }
        },
        "render_prompt_card_hash": "2b597d76cc077558"
      }
    ],
    "consumed_bg_catalog_hash": "d416ac4b47723c22",
    "consumed_shot_binding_hash": "7e96fc96a43e8f42"
  },
  "step_id": "scene_detail",
  "run_id": "645ca9e9-847d-430b-b901-ba5bf45c2130",
  "resolved_model": "gemini-pro",
  "updated_at": "2026-05-13T20:23:18.996714+00:00",
  "project_config_snapshot": {}
}