{
  "status": "completed",
  "completed_count": 7,
  "applicable_count": 7,
  "failed_count": 0,
  "schema_version": 13,
  "config_hash": "354f3d1c2f1fbaf6",
  "data": {
    "scenes": [
      {
        "scene_index": 1,
        "heading": "정비소 안 – 저녁",
        "beat_title": "골목의 위급 상황 전달",
        "representative_moment": "손전등을 들고 정비소 문을 밀고 들어서며 내부를 응시하는 C02의 상반신",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Tight upper-body medium shot at chest height, tracking from inside the doorway",
            "t2i_prompt": "Photorealistic cinematic still. Framed tighter and slightly lower than the background reference, seen from inside at chest height offset beside the doorway, C02O02 in a yellow raincoat, a Korean woman in her twenties, occupies the middle-center midground, moving toward the garage interior, with the open garage doorway and its vertical interior frame standing in the middle-left background. She is captured in a tight upper-body three-quarter angle from the right, her body weight shifted forward onto her front foot to cross the threshold. One hand presses against the partially opened door plane at the left edge, and the opposite hand holds a P02 flashlight low. Her face angles past the camera, eyes fixed on an unseen figure deeper inside. The desaturated cool-toned overhead fluorescent light casts a restrained, slightly desaturated ambient light with moderate low-key contrast, catching the reflecting moisture on her wet shoulders. Deeper inside [L01: the narrow garage], an oil-stained workbench, an open toolbox, and a resting wrench sit in dim shadow, alongside the inverted frame of a worn bicycle seen obliquely beyond her.",
            "outfit_assignments": [
              {
                "character_id": "C02",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "연희(20대), 노란 우비를 입고 손전등을 든 채로 문을 밀고 들어선다.",
              "숨이 가쁘고 어깨가 젖어 있다.",
              "Shot description: 한 손에 손전등을 쥐고 좁은 정비소 문을 민 채 안으로 한 걸음을 내디딘 자세의 연희 상체."
            ],
            "visual_inferences": [
              "fixed_element yeonhee_raincoat_state: Her forward-leaning posture indicates urgency after stepping through the doorway.",
              "fixed_element garage_evening_lighting: The single light source casts distinct shadows that highlight the tight, confined space of the garage."
            ],
            "creative_decisions": [
              "Explicitly deviated from the wider standing-height background reference framing to a tighter chest-height medium shot to match the DP's direction for a tight upper-body three-quarter view.",
              "Integrated the forward zoom targets (open garage doorway, vertical interior frame, desaturated cool-toned fluorescent light) naturally into the environment description to ensure continuity for the next shot."
            ],
            "confidence": "high",
            "applied_frame_spatial_constraint_ids": [
              "fsc_001",
              "fsc_002"
            ],
            "reference_phrase_kinds": [
              "character",
              "background"
            ],
            "owned_object_usage": [
              {
                "owned_token": "bicycle",
                "usage_kind": "anchor",
                "source_phrase": "the inverted frame of a worn bicycle"
              },
              {
                "owned_token": "door",
                "usage_kind": "redraw",
                "source_phrase": "the open garage doorway and its vertical interior frame standing in the middle-left background"
              },
              {
                "owned_token": "flashlight",
                "usage_kind": "redraw",
                "source_phrase": "a P02 flashlight low"
              },
              {
                "owned_token": "fluorescent light",
                "usage_kind": "anchor",
                "source_phrase": "overhead fluorescent light"
              },
              {
                "owned_token": "toolbox",
                "usage_kind": "anchor",
                "source_phrase": "an open toolbox"
              },
              {
                "owned_token": "workbench",
                "usage_kind": "anchor",
                "source_phrase": "an oil-stained workbench"
              },
              {
                "owned_token": "wrench",
                "usage_kind": "anchor",
                "source_phrase": "a resting wrench"
              }
            ],
            "owned_validation": {
              "schema_version": 2,
              "t2i_prompt_hash": "b334d7c73ea0febe",
              "owned_hash": "197c74fcc1709f49",
              "camera_direction_hash": "36f42880d87054b8",
              "owned_usage_hash": "df9a1767f794e9ef",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "bicycle",
                  "violating_phrase": "the inverted frame of a worn bicycle",
                  "reason": "The prompt references the bicycle as a background element sitting in dim shadow, serving as an anchor reference with no redraw intent, matching the producer's declaration.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "door",
                  "violating_phrase": "the open garage doorway and its vertical interior frame standing in the middle-left background",
                  "reason": "The prompt references the doorway as a spatial anchor and describes the hand pressing against the door plane, which does not constitute redrawing the background door, correcting the producer's redraw declaration to anchor.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "flashlight",
                  "violating_phrase": "holds a P02 flashlight low",
                  "reason": "The flashlight is a held prop, and describing her hand holding it low falls under the allowed physical interaction exception, supported by the shot intent.",
                  "verdict": "allowed_prop_contact"
                },
                {
                  "owned_object": "fluorescent light",
                  "violating_phrase": "overhead fluorescent light",
                  "reason": "The prompt references the fluorescent light to describe the ambient lighting effect, serving as an anchor reference.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "toolbox",
                  "violating_phrase": "an open toolbox",
                  "reason": "The prompt references the toolbox as a background element sitting in shadow, serving as an anchor reference.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "workbench",
                  "violating_phrase": "an oil-stained workbench",
                  "reason": "The prompt references the workbench as a background element sitting in shadow, serving as an anchor reference.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "wrench",
                  "violating_phrase": "a resting wrench",
                  "reason": "The prompt references the wrench as a background element sitting in shadow, serving as an anchor reference.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "",
        "_shot_index": 2,
        "visible_entities": [
          "C02",
          "L01",
          "P02"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 1,
            "shot_index": 2
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "medium",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "From inside the garage at chest height and offset beside the doorway, track with Yeonhee in a tight upper-body three-quarter view as she pushes the door with one hand and transfers her weight through her first step. Keep her door hand at the left edge and the flashlight low in the opposite hand, while her face and gaze angle past the camera toward Minsu deeper inside the garage.",
            "lighting_mood": "Restrained, slightly desaturated ambient light appropriate to the evening garage, with moderate low-key contrast and clear texture on Yeonhee's wet shoulder.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction expresses a vertical camera height or viewpoint position 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": {
                "not_applicable": true,
                "not_applicable_reason": "이 샷의 visible_entities 에 인물이 1명이라 두 인물 사이의 공유 anchor 규칙은 성립하지 않는다",
                "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)"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, wide, insert} from `shot_staging.framing_scale` enum (render_strategy.framing_scale, production SOT). LLM consumes the enum only; keyword classification is forbidden.",
                "close_framing_rules": {
                  "not_applicable": true,
                  "not_applicable_reason": "framing_scale=medium — close 규칙은 이 샷에 해당하지 않는다"
                },
                "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",
                  "subject_reference_policy_cross_ref": "see id_policy.subject_reference_policy (subject_reference_policy SOT — policy enum 에 따라 ID/outlook 사용)"
                },
                "self_check_steps": [
                  "read framing_scale enum from render_strategy.framing_scale (do NOT infer 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."
            },
            "frame_spatial_contract": {
              "reason": "movement_direction",
              "constraints": [
                {
                  "constraint_id": "fsc_001",
                  "target_kind": "background",
                  "target_id": "",
                  "label": "open garage doorway",
                  "screen_zone": "middle_left",
                  "depth_plane": "background",
                  "gesture_action": "none",
                  "gesture_target_label": ""
                },
                {
                  "constraint_id": "fsc_002",
                  "target_kind": "character",
                  "target_id": "C02",
                  "label": "연희",
                  "screen_zone": "middle_center",
                  "depth_plane": "midground",
                  "gesture_action": "moves_toward",
                  "gesture_target_label": "garage interior"
                }
              ]
            },
            "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",
              "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",
              "L01",
              "P02"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C02",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "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"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — generic descriptor with demographic 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"
            },
            "subject_reference_policy": [],
            "constraints": [
              "ID format for visible subjects (C## bare, C##O## composite IDs, or common-noun descriptor) is determined per-subject by the id_policy per-subject policy array — see constraints[2] below",
              "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 generic descriptor with demographic instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the generic-descriptor rule applies only to a face reproduction, not to the prop entity.",
              "per-subject identity reference policy is provided in id_policy.subject_reference_policy[] — apply each subject's policy when emitting references: (a) id_and_outlook_required ⇒ subject must use C##O## composite ID; (b) base_id_required ⇒ subject must use bare C## only (no outlook); (c) generic_descriptor_allowed ⇒ subject must NOT use C## or C##O##, render with a common/demographic descriptor only. omitted subjects default to id_and_outlook_required",
              "when an ID is required for a subject (policy = id_and_outlook_required or base_id_required) and a C##O## (or C##) is introduced 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. does not apply to generic_descriptor_allowed subjects (no ID emitted)",
              "when an ID is required (policy = id_and_outlook_required or base_id_required), 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). does not apply to generic_descriptor_allowed subjects — those use a common/demographic descriptor with no C##/C##O##"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L01B01",
            "reference_usage": "exact_background",
            "owned_objects": [
              "bicycle",
              "door",
              "flashlight",
              "fluorescent light",
              "toolbox",
              "workbench",
              "wrench"
            ],
            "camera_reference": {
              "camera_position": "near number 3 (open floor zone), facing diagonally toward number 6 (entrance door), with number 7 (workbench) and number 8 (repair bicycle) visible along the side and rear",
              "camera_height": "eye-level standing, approximately 1.6m",
              "lens_hint": "35mm moderate wide angle",
              "framing_notes": "Show the narrow depth of the shop, the visitor entering through number 6, the mechanic stopping beside numbers 7 and 8, number 10 on the bench, and number 13 in the visitor's hand. Keep the central aisle legible beneath number 9."
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "if focusing on an owned object, declare its reference-kind via the reference_phrase_kinds sidecar (producer contract); do not enumerate forbidden phrasings here",
              "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": "yeonhee_raincoat_state",
                "element_type": "character_state",
                "character_name": "Yeonhee",
                "description": "A Korean woman in her twenties wearing a yellow raincoat. The shoulders of the raincoat are visibly wet, reflecting moisture. Her posture conveys motion, with her body weight shifted forward onto her front foot.",
                "applies_to_shots": [
                  2,
                  4
                ],
                "element_scope": "full",
                "source_facts": [
                  "연희(20대), 노란 우비를 입고",
                  "어깨가 젖어 있다.",
                  "weight crossing the threshold",
                  "weight held on front foot"
                ],
                "visual_inferences": [
                  "Her forward-leaning posture indicates urgency after stepping through the doorway."
                ],
                "creative_decisions": [
                  "Emphasizing the wet texture of the raincoat to vividly contrast with the dry, oily interior of the garage."
                ],
                "confidence": "high"
              },
              {
                "element_id": "garage_evening_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A narrow garage interior during the evening. The space is illuminated by a single fluorescent light hanging from the ceiling, providing a restrained, slightly desaturated ambient light. The lighting creates a moderate low-key contrast across the room.",
                "applies_to_shots": [
                  2,
                  4
                ],
                "element_scope": "full",
                "source_facts": [
                  "좁은 정비소. 형광등 하나가 천장에 매달려 있다.",
                  "Restrained, slightly desaturated ambient light appropriate to the evening garage",
                  "moderate low-key contrast"
                ],
                "visual_inferences": [
                  "The single light source casts distinct shadows that highlight the tight, confined space of the garage."
                ],
                "creative_decisions": [
                  "Using a desaturated color palette to enhance the somber and gritty atmosphere of the scene."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [
              {
                "scene_index": 3,
                "shot_index": 2,
                "description": "문가에 선 연희가 열린 문 밖의 어두운 골목 쪽으로 시선을 고정한 채 서 있는 옆얼굴.",
                "keep_elements": [
                  {
                    "label": "the open garage door and its vertical interior frame",
                    "kind": "environment",
                    "subject_kind": "non_human_visual_element"
                  },
                  {
                    "label": "the desaturated cool-toned overhead fluorescent light",
                    "kind": "environment",
                    "subject_kind": "non_human_visual_element"
                  }
                ]
              }
            ],
            "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'"
                ],
                "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy applies per-subject; policy enum (id_and_outlook_required / base_id_required / generic_descriptor_allowed) determines C##/C##O## use",
                "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. third-person full body and close-up of the same character must not be combined in a single prompt.",
              "third_person_handling": "describe persons with full or partial body in frame; single-camera consistency forbids combining a partial close-up and the same character's full body in one prompt",
              "close_up_handling": "an isolated framed area of a single subject must not coexist with that subject's full-body description in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "any framing-intent wording designates the area shown by the single camera position; it does not switch the view",
              "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy — subjects with policy = 'base_id_required' or 'generic_descriptor_allowed' must not use outlook reference (C##O## form); additionally, subjects with policy = 'generic_descriptor_allowed' must not use bare C## either — render with a common/demographic descriptor only",
              "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 — third-person full body and close-up of the same character must not be combined in a single shot. any framing-intent wording designates the area shown by the single camera position, not a view switch (id_policy.subject_reference_policy applies per-subject)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C02O02",
                "policy": "required"
              },
              {
                "kind": "background",
                "id": "L01B01",
                "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": "4334905d1719d23b"
      },
      {
        "scene_index": 1,
        "heading": "#1. 정비소 안 – 저녁",
        "beat_title": "위급 상황 인지",
        "representative_moment": "자전거 수리를 멈추고 몸을 일으킨 민수가 문가에 나타난 연희를 응시하는 순간.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Over-the-shoulder medium shot from standing eye level, emphasizing the shared spatial tension across the narrow garage under moderate low-key lighting.",
            "t2i_prompt": "Photorealistic cinematic still. C02O02 in a yellow raincoat, a Korean young woman with natural East Asian facial features and dark hair, stands in the middle-left foreground, looking toward C01O01 across the garage, while C01O01 in a dark worker uniform, a Korean middle-aged man with natural East Asian facial features and dark hair, pauses in the middle-right midground, looking toward C02O02 at the entrance door. Seen at standing eye level from just behind and beside C02O02's shoulder, her back to the camera and near shoulder form a soft foreground edge. The two figures share the single aisle space of the narrow garage, separated by the concrete floor. C02O02's posture conveys motion, with her body weight shifted forward onto her front foot, the shoulders of her raincoat visibly wet and reflecting moisture. Across the aisle, C01O01 is seen at a three-quarter angle from the left, paused over the inverted bicycle chain, his chin raised and eyes fixed on C02O02. The bicycle is viewed from an oblique side beneath him, its frame visible. Behind him, the greasy workbench and open toolbox face partly toward the camera. The open entrance door sits immediately beside C02O02, the opening extending behind her. The L01B01 space is illuminated by a single fluorescent light hanging from the ceiling, providing a restrained, slightly desaturated evening-interior ambient light that maintains moderate low-key contrast across C01O01's lifted face and C02O02's near shoulder.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O01"
              },
              {
                "character_id": "C02",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "고개를 치켜든 채 문가에 선 연희의 얼굴을 똑바로 마주 보고 있는 민수.",
              "민수가 손을 멈추고 고개를 든다. 쥐고 있던 렌치를 작업대 위에 내려놓고 자리에서 일어선다.",
              "연희(20대), 노란 우비를 입고 손전등을 든 채로 문을 밀고 들어선다. 숨이 가쁘고 어깨가 젖어 있다."
            ],
            "visual_inferences": [
              "fixed_element yeonhee_raincoat_state: body weight shifted forward onto her front foot, raincoat visibly wet",
              "fixed_element garage_evening_lighting: moderate low-key contrast from single overhead light",
              "C02 over_shoulder framing forming a soft edge in the foreground"
            ],
            "creative_decisions": [
              "Positioning the camera at standing eye level to align the over-the-shoulder perspective with Minsu's lifted face, emphasizing their shared eyeline.",
              "Using the single fluorescent light to cast low-key contrast across Minsu's face and Yeonhee's wet raincoat shoulder to build a tense, interrupted moment."
            ],
            "confidence": "high",
            "applied_frame_spatial_constraint_ids": [
              "fsc_001",
              "fsc_002"
            ],
            "reference_phrase_kinds": [
              "character",
              "background"
            ],
            "owned_object_usage": [
              {
                "owned_token": "bicycle",
                "usage_kind": "anchor",
                "source_phrase": "paused over the inverted bicycle chain"
              },
              {
                "owned_token": "door",
                "usage_kind": "anchor",
                "source_phrase": "at the entrance door"
              },
              {
                "owned_token": "flashlight",
                "usage_kind": "absent",
                "source_phrase": ""
              },
              {
                "owned_token": "fluorescent light",
                "usage_kind": "anchor",
                "source_phrase": "illuminated by a single fluorescent light"
              },
              {
                "owned_token": "toolbox",
                "usage_kind": "anchor",
                "source_phrase": "open toolbox"
              },
              {
                "owned_token": "workbench",
                "usage_kind": "anchor",
                "source_phrase": "greasy workbench"
              },
              {
                "owned_token": "wrench",
                "usage_kind": "absent",
                "source_phrase": ""
              }
            ],
            "owned_validation": {
              "schema_version": 2,
              "t2i_prompt_hash": "dca978a82cb54146",
              "owned_hash": "197c74fcc1709f49",
              "camera_direction_hash": "5bb4bc1b7dfc7541",
              "owned_usage_hash": "c4070da4f40324b0",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "bicycle",
                  "violating_phrase": "paused over the inverted bicycle chain, his chin raised and eyes fixed on C02O02. The bicycle is viewed from an oblique side beneath him, its frame visible.",
                  "reason": "The prompt references the bicycle solely to establish the character's posture and camera angle relative to it, which matches the producer's anchor declaration.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "door",
                  "violating_phrase": "at the entrance door. ... The open entrance door sits immediately beside C02O02, the opening extending behind her.",
                  "reason": "The entrance door is used purely as a spatial reference for positioning the character and framing the camera, aligning with the anchor declaration.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "fluorescent light",
                  "violating_phrase": "illuminated by a single fluorescent light hanging from the ceiling",
                  "reason": "The fluorescent light is mentioned solely to establish the ambient light source of the scene, matching the anchor declaration.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "toolbox",
                  "violating_phrase": "open toolbox face partly toward the camera",
                  "reason": "The open toolbox is referenced as a static background element to define the scene's composition, matching the producer's anchor declaration.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "workbench",
                  "violating_phrase": "greasy workbench",
                  "reason": "The greasy workbench is mentioned as a background element to anchor the scene's environment, agreeing with the producer's anchor declaration.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 1,
        "dependency_reason": "S1_Shot2 exact_background: same room, different angle and moment",
        "_shot_index": 4,
        "visible_entities": [
          "C01",
          "C02",
          "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": "Hold static at Yeonhee's shoulder height just behind and beside her, using only her near shoulder as a soft foreground edge while leaving Minsu's lifted face unobstructed across the narrow garage. Minsu remains bent near the bicycle in the midground, chin raised and eyes locked on Yeonhee, so the changed emphasis is their shared eyeline rather than a change in camera distance.",
            "lighting_mood": "Restrained, slightly desaturated evening-interior illumination maintains moderate low-key contrast across Minsu's lifted face and Yeonhee's near shoulder.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction expresses a vertical camera height or viewpoint position 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, wide, insert} from `shot_staging.framing_scale` enum (render_strategy.framing_scale, production SOT). LLM consumes the enum only; keyword classification is forbidden.",
                "close_framing_rules": {
                  "not_applicable": true,
                  "not_applicable_reason": "framing_scale=medium — close 규칙은 이 샷에 해당하지 않는다"
                },
                "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",
                  "subject_reference_policy_cross_ref": "see id_policy.subject_reference_policy (subject_reference_policy SOT — policy enum 에 따라 ID/outlook 사용)"
                },
                "self_check_steps": [
                  "read framing_scale enum from render_strategy.framing_scale (do NOT infer 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."
            },
            "frame_spatial_contract": {
              "reason": "shared_space_relation",
              "constraints": [
                {
                  "constraint_id": "fsc_001",
                  "target_kind": "character",
                  "target_id": "C01",
                  "label": "민수",
                  "screen_zone": "middle_right",
                  "depth_plane": "midground",
                  "gesture_action": "looks_toward",
                  "gesture_target_label": "연희 at the doorway"
                },
                {
                  "constraint_id": "fsc_002",
                  "target_kind": "character",
                  "target_id": "C02",
                  "label": "연희",
                  "screen_zone": "middle_left",
                  "depth_plane": "foreground",
                  "gesture_action": "looks_toward",
                  "gesture_target_label": "민수 across the garage"
                }
              ]
            },
            "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",
              "L01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O01"
              },
              {
                "character_id": "C02",
                "outlook_id": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "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"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — generic descriptor with demographic 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"
            },
            "subject_reference_policy": [],
            "constraints": [
              "ID format for visible subjects (C## bare, C##O## composite IDs, or common-noun descriptor) is determined per-subject by the id_policy per-subject policy array — see constraints[2] below",
              "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 generic descriptor with demographic instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the generic-descriptor rule applies only to a face reproduction, not to the prop entity.",
              "per-subject identity reference policy is provided in id_policy.subject_reference_policy[] — apply each subject's policy when emitting references: (a) id_and_outlook_required ⇒ subject must use C##O## composite ID; (b) base_id_required ⇒ subject must use bare C## only (no outlook); (c) generic_descriptor_allowed ⇒ subject must NOT use C## or C##O##, render with a common/demographic descriptor only. omitted subjects default to id_and_outlook_required",
              "when an ID is required for a subject (policy = id_and_outlook_required or base_id_required) and a C##O## (or C##) is introduced 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. does not apply to generic_descriptor_allowed subjects (no ID emitted)",
              "when an ID is required (policy = id_and_outlook_required or base_id_required), 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). does not apply to generic_descriptor_allowed subjects — those use a common/demographic descriptor with no C##/C##O##"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L01B01",
            "reference_usage": "exact_background",
            "owned_objects": [
              "bicycle",
              "door",
              "flashlight",
              "fluorescent light",
              "toolbox",
              "workbench",
              "wrench"
            ],
            "camera_reference": {
              "camera_position": "near number 3 (open floor zone), facing diagonally toward number 6 (entrance door), with number 7 (workbench) and number 8 (repair bicycle) visible along the side and rear",
              "camera_height": "eye-level standing, approximately 1.6m",
              "lens_hint": "35mm moderate wide angle",
              "framing_notes": "Show the narrow depth of the shop, the visitor entering through number 6, the mechanic stopping beside numbers 7 and 8, number 10 on the bench, and number 13 in the visitor's hand. Keep the central aisle legible beneath number 9."
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "if focusing on an owned object, declare its reference-kind via the reference_phrase_kinds sidecar (producer contract); do not enumerate forbidden phrasings here",
              "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": "yeonhee_raincoat_state",
                "element_type": "character_state",
                "character_name": "Yeonhee",
                "description": "A Korean woman in her twenties wearing a yellow raincoat. The shoulders of the raincoat are visibly wet, reflecting moisture. Her posture conveys motion, with her body weight shifted forward onto her front foot.",
                "applies_to_shots": [
                  2,
                  4
                ],
                "element_scope": "full",
                "source_facts": [
                  "연희(20대), 노란 우비를 입고",
                  "어깨가 젖어 있다.",
                  "weight crossing the threshold",
                  "weight held on front foot"
                ],
                "visual_inferences": [
                  "Her forward-leaning posture indicates urgency after stepping through the doorway."
                ],
                "creative_decisions": [
                  "Emphasizing the wet texture of the raincoat to vividly contrast with the dry, oily interior of the garage."
                ],
                "confidence": "high"
              },
              {
                "element_id": "garage_evening_lighting",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A narrow garage interior during the evening. The space is illuminated by a single fluorescent light hanging from the ceiling, providing a restrained, slightly desaturated ambient light. The lighting creates a moderate low-key contrast across the room.",
                "applies_to_shots": [
                  2,
                  4
                ],
                "element_scope": "full",
                "source_facts": [
                  "좁은 정비소. 형광등 하나가 천장에 매달려 있다.",
                  "Restrained, slightly desaturated ambient light appropriate to the evening garage",
                  "moderate low-key contrast"
                ],
                "visual_inferences": [
                  "The single light source casts distinct shadows that highlight the tight, confined space of the garage."
                ],
                "creative_decisions": [
                  "Using a desaturated color palette to enhance the somber and gritty atmosphere of the scene."
                ],
                "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'"
                ],
                "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy applies per-subject; policy enum (id_and_outlook_required / base_id_required / generic_descriptor_allowed) determines C##/C##O## use",
                "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. third-person full body and close-up of the same character must not be combined in a single prompt.",
              "third_person_handling": "describe persons with full or partial body in frame; single-camera consistency forbids combining a partial close-up and the same character's full body in one prompt",
              "close_up_handling": "an isolated framed area of a single subject must not coexist with that subject's full-body description in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "any framing-intent wording designates the area shown by the single camera position; it does not switch the view",
              "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy — subjects with policy = 'base_id_required' or 'generic_descriptor_allowed' must not use outlook reference (C##O## form); additionally, subjects with policy = 'generic_descriptor_allowed' must not use bare C## either — render with a common/demographic descriptor only",
              "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 — third-person full body and close-up of the same character must not be combined in a single shot. any framing-intent wording designates the area shown by the single camera position, not a view switch (id_policy.subject_reference_policy applies per-subject)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O01",
                "policy": "required"
              },
              {
                "kind": "character_outlook",
                "id": "C02O02",
                "policy": "required"
              },
              {
                "kind": "background",
                "id": "L01B01",
                "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": "5931f19592934b6e"
      },
      {
        "scene_index": 2,
        "heading": "#2. 갈래길 – 저녁",
        "beat_title": "사거리 한가운데 방치된 노인",
        "representative_moment": "켜진 가로등 불빛이 흔들리는 젖은 사거리 한가운데에 C03이 지팡이를 떨어뜨린 채 쓰러져 있다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Elevated oblique wide angle, restrained slightly desaturated low-key palette",
            "t2i_prompt": "Photorealistic cinematic still. Seen from an elevated oblique wide angle above one corner, the wet paved pavement of the four-way intersection fills the frame, its four branches radiating toward the edges. At the exact visual center, C03O03 in a grey coat, an elderly East Asian man with a wrinkled face and grey-white hair, lies motionless and unconscious. He is positioned at a three-quarter angle from the left, curled on one side with his eyes closed. A dropped wooden cane on the wet pavement rests at an oblique angle beside his body. An upright flickering streetlamp is visible obliquely in the upper frame, casting unstable, subdued evening illumination. The reflection of the streetlamp light on the damp pavement forms an elongated luminous patch around and beyond the small, isolated figure, establishing a restrained, slightly desaturated low-key palette.",
            "outfit_assignments": [
              {
                "character_id": "C03",
                "outlook_id": "O03"
              }
            ],
            "source_facts": [
              "길이 네 갈래로 갈라지는 사거리. 가로등 하나가 깜빡인다. 젖은 바닥에 불빛이 길게 번져 있다.",
              "사거리 한가운데에 노인(70대)이 회색 외투 차림으로 눈을 감은 채 옆으로 누워 있다. 곁에 지팡이가 떨어져 있다.",
              "Shot description: 켜진 가로등 불빛이 흔들리는 젖은 사거리 한가운데에 노인이 눈을 감고 누워 있는 전경."
            ],
            "visual_inferences": [
              "fixed_element environment_state: wet_intersection_environment",
              "immobilized subject constraint: C03 lying curled on one side with the dropped cane firmly beside him"
            ],
            "creative_decisions": [
              "Elevated oblique view to emphasize the isolation of the figure within the wide, wet intersection",
              "Integration of the forward zoom targets (wet pavement, cane, reflection) directly into the wide composition",
              "Conversion of 'crane descent' dynamic motion into a static 'elevated oblique wide angle' to capture the single frozen shutter moment"
            ],
            "confidence": "high",
            "applied_frame_spatial_constraint_ids": [],
            "reference_phrase_kinds": [
              "character"
            ],
            "owned_object_usage": [],
            "owned_validation": {
              "schema_version": 2,
              "t2i_prompt_hash": "be676b72f8537159",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "21861eecf555f2b5",
              "owned_usage_hash": "4f53cda18c2baa0c",
              "validator": "scene_detail_owned_objects.v1",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "No prior visual scene dependency.",
        "_shot_index": 1,
        "visible_entities": [
          "C03",
          "L02"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 2,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "wide",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Hold an elevated oblique wide view from above one corner of the four-way intersection before beginning the crane descent, placing the elderly man as a small isolated figure at the exact visual center. The four roads spread toward the frame edges while his side-lying body and fallen cane remain clearly separated against the wet pavement beneath the flickering streetlamp.",
            "lighting_mood": "The flickering streetlamp casts unstable evening light that stretches across the wet pavement in a restrained, slightly desaturated low-key palette.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction expresses a vertical camera height or viewpoint position 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": {
                "not_applicable": true,
                "not_applicable_reason": "이 샷의 visible_entities 에 인물이 1명이라 두 인물 사이의 공유 anchor 규칙은 성립하지 않는다",
                "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)"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, wide, insert} from `shot_staging.framing_scale` enum (render_strategy.framing_scale, production SOT). LLM consumes the enum only; keyword classification is forbidden.",
                "close_framing_rules": {
                  "not_applicable": true,
                  "not_applicable_reason": "framing_scale=wide — close 규칙은 이 샷에 해당하지 않는다"
                },
                "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",
                  "subject_reference_policy_cross_ref": "see id_policy.subject_reference_policy (subject_reference_policy SOT — policy enum 에 따라 ID/outlook 사용)"
                },
                "self_check_steps": [
                  "read framing_scale enum from render_strategy.framing_scale (do NOT infer 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."
            },
            "frame_spatial_contract": null,
            "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",
              "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",
              "L02"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C03",
                "outlook_id": "O03"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "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"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — generic descriptor with demographic 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"
            },
            "subject_reference_policy": [],
            "constraints": [
              "ID format for visible subjects (C## bare, C##O## composite IDs, or common-noun descriptor) is determined per-subject by the id_policy per-subject policy array — see constraints[2] below",
              "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 generic descriptor with demographic instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the generic-descriptor rule applies only to a face reproduction, not to the prop entity.",
              "per-subject identity reference policy is provided in id_policy.subject_reference_policy[] — apply each subject's policy when emitting references: (a) id_and_outlook_required ⇒ subject must use C##O## composite ID; (b) base_id_required ⇒ subject must use bare C## only (no outlook); (c) generic_descriptor_allowed ⇒ subject must NOT use C## or C##O##, render with a common/demographic descriptor only. omitted subjects default to id_and_outlook_required",
              "when an ID is required for a subject (policy = id_and_outlook_required or base_id_required) and a C##O## (or C##) is introduced 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. does not apply to generic_descriptor_allowed subjects (no ID emitted)",
              "when an ID is required (policy = id_and_outlook_required or base_id_required), 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). does not apply to generic_descriptor_allowed subjects — those use a common/demographic descriptor with no C##/C##O##"
            ]
          },
          "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"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "wet_intersection_environment",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The paved surface of the four-way intersection is noticeably wet, forming reflective areas across the ground. An unstable, flickering streetlamp casts stretched, subdued evening illumination over the damp pavement.",
                "applies_to_shots": [
                  1,
                  3,
                  6
                ],
                "element_scope": "full",
                "source_facts": [
                  "길이 네 갈래로 갈라지는 사거리",
                  "가로등 하나가 깜빡인다",
                  "젖은 바닥에 불빛이 길게 번져 있다",
                  "flickering streetlamp casts unstable evening light"
                ],
                "visual_inferences": [
                  "The wet asphalt acts as a dark mirror, reflecting the faint ambient light and the streetlamp's beam."
                ],
                "creative_decisions": [
                  "A desaturated, low-key color palette is maintained to enhance the cold, desolate, and solitary evening atmosphere."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [],
            "forward_zoom_targets": [
              {
                "scene_index": 2,
                "shot_index": 3,
                "description": "연희가 앞으로 뻗은 손전등의 원형 불빛이 바닥에 누운 노인의 눈 감은 얼굴 위를 비춘 찰나.",
                "keep_elements": [
                  {
                    "label": "wet paved pavement of the four-way intersection",
                    "kind": "environment",
                    "subject_kind": "non_human_visual_element"
                  },
                  {
                    "label": "reflection of the streetlamp light on the damp pavement",
                    "kind": "environment",
                    "subject_kind": "non_human_visual_element"
                  },
                  {
                    "label": "dropped wooden cane on the wet pavement",
                    "kind": "static_prop",
                    "subject_kind": "non_human_visual_element"
                  }
                ]
              }
            ],
            "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'"
                ],
                "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy applies per-subject; policy enum (id_and_outlook_required / base_id_required / generic_descriptor_allowed) determines C##/C##O## use",
                "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. third-person full body and close-up of the same character must not be combined in a single prompt.",
              "third_person_handling": "describe persons with full or partial body in frame; single-camera consistency forbids combining a partial close-up and the same character's full body in one prompt",
              "close_up_handling": "an isolated framed area of a single subject must not coexist with that subject's full-body description in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "any framing-intent wording designates the area shown by the single camera position; it does not switch the view",
              "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy — subjects with policy = 'base_id_required' or 'generic_descriptor_allowed' must not use outlook reference (C##O## form); additionally, subjects with policy = 'generic_descriptor_allowed' must not use bare C## either — render with a common/demographic descriptor only",
              "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 — third-person full body and close-up of the same character must not be combined in a single shot. any framing-intent wording designates the area shown by the single camera position, not a view switch (id_policy.subject_reference_policy applies per-subject)",
              "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": "4cbbc249dd463062"
      },
      {
        "scene_index": 2,
        "heading": "갈래길 – 저녁",
        "beat_title": "의식 없는 노인을 비추는 손전등 불빛",
        "representative_moment": "C02가 쓰러져 있는 C03의 얼굴을 P02 손전등으로 비추고 있다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Close framing from knee height, low-key lighting with localized flashlight beam",
            "t2i_prompt": "Photorealistic cinematic still. C03O03 in a grey coat, an elderly Korean man in his 70s with a wrinkled face, lies in the lower-center foreground. C02O02 in a yellow raincoat, a young Korean woman in her 20s, leans forward in the upper-right background, her eyes fixed on his illuminated face. Seen from knee height beside the lying figure, C03O03 is in right profile, lying curled on his side on the [L02: wet paved pavement of the intersection, reflecting subdued evening streetlamp light] they both share. He remains completely still with his eyes firmly closed, displaying an unconscious state. C02O02, at a three-quarter angle from the left, has her arm extended, holding a P02 portable flashlight. The P02 flashlight casts a localized circle of bright illumination directly across C03O03's face, establishing a restrained low-key contrast against the darker surroundings. The empty wet ground area immediately beside C03O03's head remains clearly visible within the frame.",
            "outfit_assignments": [
              {
                "character_id": "C02",
                "outlook_id": "O02"
              },
              {
                "character_id": "C03",
                "outlook_id": "O03"
              }
            ],
            "source_facts": [
              "사거리 한가운데에 노인(70대)이 회색 외투 차림으로 눈을 감은 채 옆으로 누워 있다.",
              "연희가 손전등을 켜서 노인의 얼굴 쪽을 조심스럽게 비춘다."
            ],
            "visual_inferences": [
              "fixed_element old_man_close_state: lies on his side with his eyes firmly closed, displaying an unconscious state.",
              "fixed_element wet_intersection_environment: The paved surface of the four-way intersection is noticeably wet.",
              "fixed_element localized_flashlight_beam: A localized circle of bright flashlight illumination highlights the area near the ground."
            ],
            "creative_decisions": [
              "Placed C03 in lower-center foreground and C02 in upper-right background to fulfill the spatial contract fsc_001 and fsc_002.",
              "Described the ground immediately beside the face as clearly visible to prepare for the forward zoom target of Shot 6.",
              "Used L02 block directly for the shared ground surface as the background reference is skipped due to close framing."
            ],
            "confidence": "high",
            "applied_frame_spatial_constraint_ids": [
              "fsc_001",
              "fsc_002"
            ],
            "reference_phrase_kinds": [
              "character"
            ],
            "owned_object_usage": [],
            "owned_validation": {
              "schema_version": 2,
              "t2i_prompt_hash": "0d6fcd58500f3641",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "9967a51b0d1ea89e",
              "owned_usage_hash": "4f53cda18c2baa0c",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "",
        "_shot_index": 3,
        "visible_entities": [
          "C02",
          "C03",
          "L02",
          "P02"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 2,
            "shot_index": 3
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "Dolly in from knee height beside the elderly man's upper body, looking diagonally across his closed face toward Yeonhee beyond him. Hold the circular flashlight beam precisely over his closed eyes and face; his head occupies the lower foreground while Yeonhee leans forward in the upper background with her extended hand and flashlight kept at a realistic relative scale.",
            "lighting_mood": "The flashlight creates a localized circle across the elderly man's face while the flickering streetlamp and its reflection provide subdued evening context.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction expresses a vertical camera height or viewpoint position 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, wide, insert} from `shot_staging.framing_scale` enum (render_strategy.framing_scale, production SOT). LLM consumes the enum only; keyword classification is forbidden.",
                "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"
                  ],
                  "subject_reference_policy_cross_ref": "see id_policy.subject_reference_policy (subject_reference_policy SOT — policy enum 에 따라 ID/outlook 사용)"
                },
                "wide_medium_rules": {
                  "not_applicable": true,
                  "not_applicable_reason": "framing_scale=close — wide/medium 규칙은 이 샷에 해당하지 않는다"
                },
                "self_check_steps": [
                  "read framing_scale enum from render_strategy.framing_scale (do NOT infer 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."
            },
            "frame_spatial_contract": {
              "reason": "shared_space_relation",
              "constraints": [
                {
                  "constraint_id": "fsc_001",
                  "target_kind": "character",
                  "target_id": "C02",
                  "label": "연희",
                  "screen_zone": "upper_right",
                  "depth_plane": "background",
                  "gesture_action": "looks_toward",
                  "gesture_target_label": "노인의 illuminated face"
                },
                {
                  "constraint_id": "fsc_002",
                  "target_kind": "character",
                  "target_id": "C03",
                  "label": "노인",
                  "screen_zone": "lower_center",
                  "depth_plane": "foreground",
                  "gesture_action": "none",
                  "gesture_target_label": ""
                }
              ]
            },
            "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",
              "C03",
              "L02",
              "P02"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C02",
                "outlook_id": "O02"
              },
              {
                "character_id": "C03",
                "outlook_id": "O03"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "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"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — generic descriptor with demographic 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"
            },
            "subject_reference_policy": [],
            "constraints": [
              "ID format for visible subjects (C## bare, C##O## composite IDs, or common-noun descriptor) is determined per-subject by the id_policy per-subject policy array — see constraints[2] below",
              "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 generic descriptor with demographic instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the generic-descriptor rule applies only to a face reproduction, not to the prop entity.",
              "per-subject identity reference policy is provided in id_policy.subject_reference_policy[] — apply each subject's policy when emitting references: (a) id_and_outlook_required ⇒ subject must use C##O## composite ID; (b) base_id_required ⇒ subject must use bare C## only (no outlook); (c) generic_descriptor_allowed ⇒ subject must NOT use C## or C##O##, render with a common/demographic descriptor only. omitted subjects default to id_and_outlook_required",
              "when an ID is required for a subject (policy = id_and_outlook_required or base_id_required) and a C##O## (or C##) is introduced 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. does not apply to generic_descriptor_allowed subjects (no ID emitted)",
              "when an ID is required (policy = id_and_outlook_required or base_id_required), 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). does not apply to generic_descriptor_allowed subjects — those use a common/demographic descriptor with no C##/C##O##"
            ]
          },
          "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",
              "describe the close subject and its immediate surroundings without relying on the absent background reference",
              "use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "old_man_close_state",
                "element_type": "character_state",
                "character_name": "Old man",
                "description": "The elderly man, in his 70s, lies on his side with his eyes firmly closed, displaying an unconscious state. He wears a grey coat visible around his upper body. His face rests near the wet pavement, remaining completely still.",
                "applies_to_shots": [
                  3,
                  6
                ],
                "element_scope": "close",
                "source_facts": [
                  "노인(70대)",
                  "눈을 감은 채 옆으로 누워 있다",
                  "회색 외투 차림으로",
                  "gaze=closed_eyes, state=unconscious"
                ],
                "visual_inferences": [
                  "The side of his face remains resting heavily near or against the pavement.",
                  "His facial muscles are completely relaxed due to his unconscious condition."
                ],
                "creative_decisions": [
                  "The grey coat texture contrasts slightly with the wet ground, anchoring his upper body within the tight framing."
                ],
                "confidence": "high"
              },
              {
                "element_id": "wet_intersection_environment",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The paved surface of the four-way intersection is noticeably wet, forming reflective areas across the ground. An unstable, flickering streetlamp casts stretched, subdued evening illumination over the damp pavement.",
                "applies_to_shots": [
                  1,
                  3,
                  6
                ],
                "element_scope": "full",
                "source_facts": [
                  "길이 네 갈래로 갈라지는 사거리",
                  "가로등 하나가 깜빡인다",
                  "젖은 바닥에 불빛이 길게 번져 있다",
                  "flickering streetlamp casts unstable evening light"
                ],
                "visual_inferences": [
                  "The wet asphalt acts as a dark mirror, reflecting the faint ambient light and the streetlamp's beam."
                ],
                "creative_decisions": [
                  "A desaturated, low-key color palette is maintained to enhance the cold, desolate, and solitary evening atmosphere."
                ],
                "confidence": "high"
              },
              {
                "element_id": "localized_flashlight_beam",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A localized circle of bright flashlight illumination highlights the area near the ground. This distinct beam creates a restrained, low-key contrast against the surrounding darker, wet environment.",
                "applies_to_shots": [
                  3,
                  6
                ],
                "element_scope": "close",
                "source_facts": [
                  "연희가 손전등을 켜서 노인의 얼굴 쪽을 조심스럽게 비춘다",
                  "flashlight creates a localized circle across the elderly man's face",
                  "localized flashlight illumination retain restrained low-key contrast"
                ],
                "visual_inferences": [
                  "The distinct edge of the circular beam sharply separates the lit faces from the much darker background."
                ],
                "creative_decisions": [
                  "The stark practical light from the flashlight provides a dramatic focus, emphasizing the urgency of checking the unconscious man."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 2,
                "shot_index": 1,
                "ref_usage": "zoom_in_detail",
                "kind": "location"
              }
            ],
            "forward_zoom_targets": [
              {
                "scene_index": 2,
                "shot_index": 6,
                "description": "바닥에 바짝 엎드리듯 고개를 숙인 채 노인의 얼굴 가까이 입을 댄 민수의 측면.",
                "keep_elements": [
                  {
                    "label": "wet paved pavement of the intersection",
                    "kind": "environment",
                    "subject_kind": "non_human_visual_element"
                  },
                  {
                    "label": "localized circular beam of flashlight illumination on the damp pavement",
                    "kind": "environment",
                    "subject_kind": "non_human_visual_element"
                  }
                ]
              }
            ],
            "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'"
                ],
                "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy applies per-subject; policy enum (id_and_outlook_required / base_id_required / generic_descriptor_allowed) determines C##/C##O## use",
                "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. third-person full body and close-up of the same character must not be combined in a single prompt.",
              "third_person_handling": "describe persons with full or partial body in frame; single-camera consistency forbids combining a partial close-up and the same character's full body in one prompt",
              "close_up_handling": "an isolated framed area of a single subject must not coexist with that subject's full-body description in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "any framing-intent wording designates the area shown by the single camera position; it does not switch the view",
              "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy — subjects with policy = 'base_id_required' or 'generic_descriptor_allowed' must not use outlook reference (C##O## form); additionally, subjects with policy = 'generic_descriptor_allowed' must not use bare C## either — render with a common/demographic descriptor only",
              "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 — third-person full body and close-up of the same character must not be combined in a single shot. any framing-intent wording designates the area shown by the single camera position, not a view switch (id_policy.subject_reference_policy applies per-subject)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C02O02",
                "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": "de909d06b87de232"
      },
      {
        "scene_index": 2,
        "heading": "#2. 갈래길 – 저녁",
        "beat_title": "노인의 의식 유무를 확인하는 민수",
        "representative_moment": "바닥에 바짝 엎드리듯 고개를 숙인 채 노인의 얼굴 가까이 입을 댄 민수의 측면.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "static ground-level perpendicular side view, localized flashlight illumination, low-key contrast",
            "t2i_prompt": "Photorealistic cinematic still. Seen from a static ground-level side view beside the head, the frame focuses closely on two upper bodies sharing the central horizontal band without overlapping. On the left, C03O03 in a grey coat, an East Asian elderly man, appears in right profile, lying curled on his side with his eyes firmly closed, displaying an unconscious state. His face rests heavily on the same wet pavement, remaining completely still. On the right, C01O01 in a dark work uniform, an East Asian middle-aged man, appears in left profile, kneeling with his torso folded low near the ground, his mouth positioned near the elderly man's face, eyes fixed on the unconscious figure. A localized circle of bright flashlight illumination highlights the narrow gap between their faces, creating a restrained, low-key contrast. The immediate ground-level anchor is the [L02: same wet paved surface], acting as a dark mirror reflecting stretched, broken streetlamp light beneath and behind the two close profiles.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O01"
              },
              {
                "character_id": "C03",
                "outlook_id": "O03"
              }
            ],
            "source_facts": [
              "노인(70대) 눈을 감은 채 옆으로 누워 있다",
              "회색 외투 차림으로",
              "민수가 노인 옆에 무릎을 꿇고",
              "shot description: 바닥에 바짝 엎드리듯 고개를 숙인 채 노인의 얼굴 가까이 입을 댄 민수의 측면"
            ],
            "visual_inferences": [
              "fixed_element old_man_close_state: His face rests near the wet pavement, remaining completely still.",
              "fixed_element wet_intersection_environment: The paved surface of the four-way intersection is noticeably wet, forming reflective areas across the ground.",
              "fixed_element localized_flashlight_beam: A localized circle of bright flashlight illumination highlights the area near the ground."
            ],
            "creative_decisions": [
              "A static ground-level side view isolates the two profiles to emphasize the urgency and intimacy of the close inspection.",
              "The wet pavement reflects the broken streetlamp light, acting as a dark mirror to anchor the lower frame."
            ],
            "confidence": "high",
            "applied_frame_spatial_constraint_ids": [],
            "reference_phrase_kinds": [
              "character"
            ],
            "owned_object_usage": [],
            "owned_validation": {
              "schema_version": 2,
              "t2i_prompt_hash": "3c5da8e6fe104e8b",
              "owned_hash": "e3b0c44298fc1c14",
              "camera_direction_hash": "9dd3fefae8b93a9e",
              "owned_usage_hash": "4f53cda18c2baa0c",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": -1,
        "dependency_reason": "zoom_in_detail refers to prior shots within the same scene, not a cross-scene dependency.",
        "_shot_index": 6,
        "visible_entities": [
          "C01",
          "C03",
          "L02"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 2,
            "shot_index": 6
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "At ground level beside the elderly man's head, hold a static perpendicular side view across the narrow gap between his face and Minsu's lowered profile. Minsu kneels with his torso folded close to the pavement and his mouth near the elderly man's face as he speaks, while both faces share the central band of the frame without overlapping.",
            "lighting_mood": "Flickering streetlamp light and the localized flashlight illumination retain restrained low-key contrast on the two close profiles and the wet ground.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction expresses a vertical camera height or viewpoint position 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, wide, insert} from `shot_staging.framing_scale` enum (render_strategy.framing_scale, production SOT). LLM consumes the enum only; keyword classification is forbidden.",
                "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"
                  ],
                  "subject_reference_policy_cross_ref": "see id_policy.subject_reference_policy (subject_reference_policy SOT — policy enum 에 따라 ID/outlook 사용)"
                },
                "wide_medium_rules": {
                  "not_applicable": true,
                  "not_applicable_reason": "framing_scale=close — wide/medium 규칙은 이 샷에 해당하지 않는다"
                },
                "self_check_steps": [
                  "read framing_scale enum from render_strategy.framing_scale (do NOT infer 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."
            },
            "frame_spatial_contract": null,
            "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",
              "L02"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O01"
              },
              {
                "character_id": "C03",
                "outlook_id": "O03"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "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"
            },
            "reproduction_surface_rule": {
              "applies": true,
              "id_use": "forbidden — generic descriptor with demographic 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"
            },
            "subject_reference_policy": [],
            "constraints": [
              "ID format for visible subjects (C## bare, C##O## composite IDs, or common-noun descriptor) is determined per-subject by the id_policy per-subject policy array — see constraints[2] below",
              "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 generic descriptor with demographic instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the generic-descriptor rule applies only to a face reproduction, not to the prop entity.",
              "per-subject identity reference policy is provided in id_policy.subject_reference_policy[] — apply each subject's policy when emitting references: (a) id_and_outlook_required ⇒ subject must use C##O## composite ID; (b) base_id_required ⇒ subject must use bare C## only (no outlook); (c) generic_descriptor_allowed ⇒ subject must NOT use C## or C##O##, render with a common/demographic descriptor only. omitted subjects default to id_and_outlook_required",
              "when an ID is required for a subject (policy = id_and_outlook_required or base_id_required) and a C##O## (or C##) is introduced 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. does not apply to generic_descriptor_allowed subjects (no ID emitted)",
              "when an ID is required (policy = id_and_outlook_required or base_id_required), 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). does not apply to generic_descriptor_allowed subjects — those use a common/demographic descriptor with no C##/C##O##"
            ]
          },
          "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",
              "describe the close subject and its immediate surroundings without relying on the absent background reference",
              "use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "old_man_close_state",
                "element_type": "character_state",
                "character_name": "Old man",
                "description": "The elderly man, in his 70s, lies on his side with his eyes firmly closed, displaying an unconscious state. He wears a grey coat visible around his upper body. His face rests near the wet pavement, remaining completely still.",
                "applies_to_shots": [
                  3,
                  6
                ],
                "element_scope": "close",
                "source_facts": [
                  "노인(70대)",
                  "눈을 감은 채 옆으로 누워 있다",
                  "회색 외투 차림으로",
                  "gaze=closed_eyes, state=unconscious"
                ],
                "visual_inferences": [
                  "The side of his face remains resting heavily near or against the pavement.",
                  "His facial muscles are completely relaxed due to his unconscious condition."
                ],
                "creative_decisions": [
                  "The grey coat texture contrasts slightly with the wet ground, anchoring his upper body within the tight framing."
                ],
                "confidence": "high"
              },
              {
                "element_id": "wet_intersection_environment",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The paved surface of the four-way intersection is noticeably wet, forming reflective areas across the ground. An unstable, flickering streetlamp casts stretched, subdued evening illumination over the damp pavement.",
                "applies_to_shots": [
                  1,
                  3,
                  6
                ],
                "element_scope": "full",
                "source_facts": [
                  "길이 네 갈래로 갈라지는 사거리",
                  "가로등 하나가 깜빡인다",
                  "젖은 바닥에 불빛이 길게 번져 있다",
                  "flickering streetlamp casts unstable evening light"
                ],
                "visual_inferences": [
                  "The wet asphalt acts as a dark mirror, reflecting the faint ambient light and the streetlamp's beam."
                ],
                "creative_decisions": [
                  "A desaturated, low-key color palette is maintained to enhance the cold, desolate, and solitary evening atmosphere."
                ],
                "confidence": "high"
              },
              {
                "element_id": "localized_flashlight_beam",
                "element_type": "environment_state",
                "character_name": "",
                "description": "A localized circle of bright flashlight illumination highlights the area near the ground. This distinct beam creates a restrained, low-key contrast against the surrounding darker, wet environment.",
                "applies_to_shots": [
                  3,
                  6
                ],
                "element_scope": "close",
                "source_facts": [
                  "연희가 손전등을 켜서 노인의 얼굴 쪽을 조심스럽게 비춘다",
                  "flashlight creates a localized circle across the elderly man's face",
                  "localized flashlight illumination retain restrained low-key contrast"
                ],
                "visual_inferences": [
                  "The distinct edge of the circular beam sharply separates the lit faces from the much darker background."
                ],
                "creative_decisions": [
                  "The stark practical light from the flashlight provides a dramatic focus, emphasizing the urgency of checking the unconscious man."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 2,
                "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'"
                ],
                "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy applies per-subject; policy enum (id_and_outlook_required / base_id_required / generic_descriptor_allowed) determines C##/C##O## use",
                "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. third-person full body and close-up of the same character must not be combined in a single prompt.",
              "third_person_handling": "describe persons with full or partial body in frame; single-camera consistency forbids combining a partial close-up and the same character's full body in one prompt",
              "close_up_handling": "an isolated framed area of a single subject must not coexist with that subject's full-body description in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "any framing-intent wording designates the area shown by the single camera position; it does not switch the view",
              "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy — subjects with policy = 'base_id_required' or 'generic_descriptor_allowed' must not use outlook reference (C##O## form); additionally, subjects with policy = 'generic_descriptor_allowed' must not use bare C## either — render with a common/demographic descriptor only",
              "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 — third-person full body and close-up of the same character must not be combined in a single shot. any framing-intent wording designates the area shown by the single camera position, not a view switch (id_policy.subject_reference_policy applies per-subject)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O01",
                "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": "fae443a24899c32e"
      },
      {
        "scene_index": 3,
        "heading": "#3. 정비소 안 – 밤",
        "beat_title": "작업대에 공간을 내어 눕힌 노인",
        "representative_moment": "밀려난 작업대 옆 바닥에 담요를 덮고 누운 C03을 C01이 쪼그려 앉아 살펴보고 있다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Slightly elevated oblique wide angle, slightly desaturated cool-toned glow, moderate low-key contrast",
            "t2i_prompt": "Photorealistic cinematic still. C01O01 in a dark worker uniform, an East Asian man in his 40s, is crouched in the lower-left foreground looking toward the elderly man beside him, while C03O03 in a gray coat, an East Asian elderly man in his 70s, lies in the middle-right midground, with the displaced workbench positioned in the upper-left background. Seen from a slightly elevated oblique angle looking diagonally down, the two figures and the cleared floor of the L01 interior fill the frame. C01O01 sits at a three-quarter angle from the right, attentive. C03O03 is seen in left profile, unconscious, slack and limp with eyes closed, his form covered by a blanket. An overhead fluorescent light emits a slightly desaturated, cool-toned glow, casting moderate low-key contrast across the floor. The displaced workbench's long side angles away, emphasizing the reorganized workspace.",
            "outfit_assignments": [
              {
                "character_id": "C01",
                "outlook_id": "O01"
              },
              {
                "character_id": "C03",
                "outlook_id": "O03"
              }
            ],
            "source_facts": [
              "작업대를 한쪽으로 밀어 자리를 냈다.",
              "형광등 불빛 아래 노인이 담요를 덮고 눈을 감은 채 누워 있다.",
              "민수가 노인 곁에 쪼그려 앉아 있다."
            ],
            "visual_inferences": [
              "fixed_element fluorescent_lighting_environment: The interior is illuminated by an overhead fluorescent light casting a slightly desaturated, cool-toned glow.",
              "C01 is positioned at a three-quarter angle from the right, checking on C03.",
              "C03 is in left profile, lying unconscious and limp under the blanket on the cleared floor."
            ],
            "creative_decisions": [
              "Use a slightly elevated wide framing to firmly anchor both figures on the shared cleared floor while keeping the displaced workbench visible in the background.",
              "Apply cool-toned, moderate low-key contrast lighting from the fluorescent light to convey the stark night atmosphere and the vulnerability of the scene."
            ],
            "confidence": "high",
            "applied_frame_spatial_constraint_ids": [
              "fsc_001",
              "fsc_002",
              "fsc_003"
            ],
            "reference_phrase_kinds": [
              "character",
              "background"
            ],
            "owned_object_usage": [
              {
                "owned_token": "bicycle",
                "usage_kind": "absent",
                "source_phrase": ""
              },
              {
                "owned_token": "blanket",
                "usage_kind": "anchor",
                "source_phrase": "his form covered by a blanket"
              },
              {
                "owned_token": "door",
                "usage_kind": "absent",
                "source_phrase": ""
              },
              {
                "owned_token": "flashlight",
                "usage_kind": "absent",
                "source_phrase": ""
              },
              {
                "owned_token": "fluorescent light",
                "usage_kind": "anchor",
                "source_phrase": "An overhead fluorescent light emits a slightly desaturated, cool-toned glow"
              },
              {
                "owned_token": "tool",
                "usage_kind": "absent",
                "source_phrase": ""
              },
              {
                "owned_token": "workbench",
                "usage_kind": "anchor",
                "source_phrase": "with the displaced workbench positioned in the upper-left background"
              }
            ],
            "owned_validation": {
              "schema_version": 2,
              "t2i_prompt_hash": "4247f046e966831b",
              "owned_hash": "9e0bdda2b88a1da7",
              "camera_direction_hash": "0951d75aab3e9f56",
              "owned_usage_hash": "1f458878745096bb",
              "validator": "scene_detail_owned_objects.v1",
              "violations": [
                {
                  "owned_object": "blanket",
                  "violating_phrase": "his form covered by a blanket",
                  "reason": "The prompt describes the character being covered by the background blanket, which is a visual state change explicitly required by the shot_intent ('담요를 덮고 누운 C03').",
                  "verdict": "allowed_visual_state_change"
                },
                {
                  "owned_object": "fluorescent light",
                  "violating_phrase": "An overhead fluorescent light emits a slightly desaturated, cool-toned glow",
                  "reason": "The prompt refers to the light emitted by the existing fluorescent light to set the scene's lighting, which is an anchor reference with no redraw intent.",
                  "verdict": "anchor_reference"
                },
                {
                  "owned_object": "workbench",
                  "violating_phrase": "with the displaced workbench positioned in the upper-left background",
                  "reason": "The prompt references the position and perspective angle of the existing workbench as a spatial anchor, matching the producer's echo.",
                  "verdict": "anchor_reference"
                }
              ]
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 1,
        "dependency_reason": "scene 1 shot 4의 톤 앤 매너(밤, 차가운 분위기, 인공광)를 atmosphere_reference로 참조",
        "_shot_index": 1,
        "visible_entities": [
          "C01",
          "C03",
          "L01"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 3,
            "shot_index": 1
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "wide",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "From beside the displaced workbench at a slightly elevated oblique angle, hold a medium-wide view looking diagonally down across crouching Minsu toward the elderly man under the fluorescent light. Place Minsu in the lower foreground and the blanket-covered elderly man across the midground, preserving enough cleared floor around them to show how the workspace has been rearranged.",
            "lighting_mood": "The fluorescent light gives the night interior a restrained, slightly desaturated illumination with moderate low-key contrast and a tense but compassionate stillness.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction expresses a vertical camera height or viewpoint position 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, wide, insert} from `shot_staging.framing_scale` enum (render_strategy.framing_scale, production SOT). LLM consumes the enum only; keyword classification is forbidden.",
                "close_framing_rules": {
                  "not_applicable": true,
                  "not_applicable_reason": "framing_scale=wide — close 규칙은 이 샷에 해당하지 않는다"
                },
                "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",
                  "subject_reference_policy_cross_ref": "see id_policy.subject_reference_policy (subject_reference_policy SOT — policy enum 에 따라 ID/outlook 사용)"
                },
                "self_check_steps": [
                  "read framing_scale enum from render_strategy.framing_scale (do NOT infer 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."
            },
            "frame_spatial_contract": {
              "reason": "shared_space_relation",
              "constraints": [
                {
                  "constraint_id": "fsc_001",
                  "target_kind": "background",
                  "target_id": "",
                  "label": "displaced workbench",
                  "screen_zone": "upper_left",
                  "depth_plane": "background",
                  "gesture_action": "none",
                  "gesture_target_label": ""
                },
                {
                  "constraint_id": "fsc_002",
                  "target_kind": "character",
                  "target_id": "C01",
                  "label": "민수",
                  "screen_zone": "lower_left",
                  "depth_plane": "foreground",
                  "gesture_action": "looks_toward",
                  "gesture_target_label": "노인 beside him"
                },
                {
                  "constraint_id": "fsc_003",
                  "target_kind": "character",
                  "target_id": "C03",
                  "label": "노인",
                  "screen_zone": "middle_right",
                  "depth_plane": "midground",
                  "gesture_action": "none",
                  "gesture_target_label": ""
                }
              ]
            },
            "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",
              "L01"
            ],
            "allowed_outlook_pairs": [
              {
                "character_id": "C01",
                "outlook_id": "O01"
              },
              {
                "character_id": "C03",
                "outlook_id": "O03"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "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"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — generic descriptor with demographic 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"
            },
            "subject_reference_policy": [],
            "constraints": [
              "ID format for visible subjects (C## bare, C##O## composite IDs, or common-noun descriptor) is determined per-subject by the id_policy per-subject policy array — see constraints[2] below",
              "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 generic descriptor with demographic instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the generic-descriptor rule applies only to a face reproduction, not to the prop entity.",
              "per-subject identity reference policy is provided in id_policy.subject_reference_policy[] — apply each subject's policy when emitting references: (a) id_and_outlook_required ⇒ subject must use C##O## composite ID; (b) base_id_required ⇒ subject must use bare C## only (no outlook); (c) generic_descriptor_allowed ⇒ subject must NOT use C## or C##O##, render with a common/demographic descriptor only. omitted subjects default to id_and_outlook_required",
              "when an ID is required for a subject (policy = id_and_outlook_required or base_id_required) and a C##O## (or C##) is introduced 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. does not apply to generic_descriptor_allowed subjects (no ID emitted)",
              "when an ID is required (policy = id_and_outlook_required or base_id_required), 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). does not apply to generic_descriptor_allowed subjects — those use a common/demographic descriptor with no C##/C##O##"
            ]
          },
          "background_binding": {
            "mode": "background_ref_attached",
            "bg_id": "L01B02",
            "reference_usage": "exact_background",
            "owned_objects": [
              "bicycle",
              "blanket",
              "door",
              "flashlight",
              "fluorescent light",
              "tool",
              "workbench"
            ],
            "camera_reference": {
              "camera_position": "near number 6 (entrance door), facing inward across number 3 (open floor zone) toward number 9 (fluorescent fixture), with number 7 (workbench) shifted toward the side through the state indicated by number 11",
              "camera_height": "eye-level standing, approximately 1.6m",
              "lens_hint": "32mm wide angle",
              "framing_notes": "Center number 12 beneath number 9 in the enlarged floor space, retain the doorway position for the waiting visitor with number 13, and show the sideward displacement of the workbench clearly without changing the permanent room geometry."
            },
            "close_framing_skips_background_ref": false,
            "constraints": [
              "owned_objects lists all environment objects already drawn in the background reference — do not create new ones",
              "if focusing on an owned object, declare its reference-kind via the reference_phrase_kinds sidecar (producer contract); do not enumerate forbidden phrasings here",
              "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": "fluorescent_lighting_environment",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The interior of the repair shop is illuminated by an overhead fluorescent light. The lighting emits a slightly desaturated, cool-toned glow that casts moderate low-key contrast against the dark night environment. This illumination defines the stark atmosphere of the interior space and spills outward toward the doorway.",
                "applies_to_shots": [
                  1,
                  2
                ],
                "element_scope": "full",
                "source_facts": [
                  "형광등 불빛 아래 노인이 담요를 덮고 눈을 감은 채 누워 있다.",
                  "The fluorescent light gives the night interior a restrained, slightly desaturated illumination",
                  "Fluorescent interior light outlines Yeonhee against the dark alley"
                ],
                "visual_inferences": [
                  "The primary light source is positioned overhead, casting a pale and cool glow typical of utilitarian fluorescent tubes.",
                  "The lighting creates a stark contrast between the illuminated interior and the deep darkness outside."
                ],
                "creative_decisions": [
                  "Emphasize the slightly desaturated, cool tones to reflect the tense, cold night while maintaining a focus on the subjects.",
                  "Use the lighting to unify the wide interior view and the close-up at the doorway."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 1,
                "shot_index": 4,
                "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'"
                ],
                "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy applies per-subject; policy enum (id_and_outlook_required / base_id_required / generic_descriptor_allowed) determines C##/C##O## use",
                "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. third-person full body and close-up of the same character must not be combined in a single prompt.",
              "third_person_handling": "describe persons with full or partial body in frame; single-camera consistency forbids combining a partial close-up and the same character's full body in one prompt",
              "close_up_handling": "an isolated framed area of a single subject must not coexist with that subject's full-body description in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "any framing-intent wording designates the area shown by the single camera position; it does not switch the view",
              "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy — subjects with policy = 'base_id_required' or 'generic_descriptor_allowed' must not use outlook reference (C##O## form); additionally, subjects with policy = 'generic_descriptor_allowed' must not use bare C## either — render with a common/demographic descriptor only",
              "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 — third-person full body and close-up of the same character must not be combined in a single shot. any framing-intent wording designates the area shown by the single camera position, not a view switch (id_policy.subject_reference_policy applies per-subject)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C01O01",
                "policy": "required"
              },
              {
                "kind": "background",
                "id": "L01B02",
                "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": "4992b25ad0d5ed57"
      },
      {
        "scene_index": 3,
        "heading": "정비소 안 – 밤",
        "beat_title": "구조를 기다리는 초조함",
        "representative_moment": "연희(C02)가 문가에 서서 손전등 스위치에 손가락을 얹은 채 어두운 골목 밖을 주시하고 있다.",
        "t2i_variations": [
          {
            "variant_label": "var_1",
            "camera_effect": "Observer shoulder-height close framing, slightly desaturated low-key fluorescent lighting",
            "t2i_prompt": "Photorealistic cinematic still. C02O02 in a yellow raincoat, a Korean woman in her 20s with natural East Asian facial features and dark hair, stands in the middle-left foreground, looking toward the alley beyond the open doorway in the middle-right background. Seen from shoulder height and slightly behind her, the figure in right profile occupies the left half of the close frame from the chest up. Her weight leans forward, and her thumb is poised over the switch of a small flashlight in her hand, her eyes fixed off-screen into the deep darkness. The same room environment from the prior shot is preserved. An overhead fluorescent tube casts a restrained, slightly desaturated cool-toned glow with moderate low-key contrast, outlining her rim-lit silhouette against the dark night. The interior edge of the open doorway runs vertically beside her, revealing only a narrow view of the dark alleyway on the right side of the frame.",
            "outfit_assignments": [
              {
                "character_id": "C02",
                "outlook_id": "O02"
              }
            ],
            "source_facts": [
              "연희는 문가에 서서 손전등을 껐다 켰다 하며 골목 쪽을 내다본다.",
              "The fluorescent light gives the night interior a restrained, slightly desaturated illumination",
              "Fluorescent interior light outlines Yeonhee against the dark alley"
            ],
            "visual_inferences": [
              "fixed_element fluorescent_lighting_environment: overhead light casting a cool-toned glow and stark contrast",
              "close framing focuses entirely on Yeonhee's profile and the immediate doorway, omitting the rest of the repair shop"
            ],
            "creative_decisions": [
              "Shoulder-height observer camera positioned slightly behind her to emphasize her eyeline channeling into the dark alley",
              "Right profile framing on the left half of the screen creates spatial tension with the narrow exterior view on the right"
            ],
            "confidence": "high",
            "applied_frame_spatial_constraint_ids": [
              "fsc_001",
              "fsc_002"
            ],
            "reference_phrase_kinds": [
              "character"
            ],
            "owned_object_usage": [
              {
                "owned_token": "bicycle",
                "usage_kind": "absent",
                "source_phrase": ""
              },
              {
                "owned_token": "blanket",
                "usage_kind": "absent",
                "source_phrase": ""
              },
              {
                "owned_token": "door",
                "usage_kind": "absent",
                "source_phrase": ""
              },
              {
                "owned_token": "flashlight",
                "usage_kind": "absent",
                "source_phrase": ""
              },
              {
                "owned_token": "fluorescent light",
                "usage_kind": "absent",
                "source_phrase": ""
              },
              {
                "owned_token": "tool",
                "usage_kind": "absent",
                "source_phrase": ""
              },
              {
                "owned_token": "workbench",
                "usage_kind": "absent",
                "source_phrase": ""
              }
            ],
            "owned_validation": {
              "schema_version": 2,
              "t2i_prompt_hash": "2ea7045a9a9ef69a",
              "owned_hash": "9e0bdda2b88a1da7",
              "camera_direction_hash": "daca14bf6381e939",
              "owned_usage_hash": "a4ac5996dc687ea6",
              "validator": "scene_detail_owned_objects.v1.close_skip",
              "violations": []
            }
          }
        ],
        "scene_type": "normal",
        "dependent_scene_index": 1,
        "dependency_reason": "zoom_in_detail ref_usage requires preserving the same exact background room environment and lighting from the prior shot, focusing on a sub-region.",
        "_shot_index": 2,
        "visible_entities": [
          "C02"
        ],
        "render_prompt_card": {
          "schema_version": 1,
          "shot_key": {
            "scene_index": 3,
            "shot_index": 2
          },
          "render_strategy": {
            "mode": "direct",
            "primary_subject": "",
            "framing_scale": "close",
            "moment_lock": "single still moment from the selected shot only",
            "camera_direction": "The track settles just inside the open doorway at Yeonhee's shoulder height, close to her interior-side profile and slightly behind her, with her face occupying the left half and a narrow corridor of the alley remaining visible on the right. Her thumb pauses over the flashlight switch while her weight inclines toward the opening, but her eyes remain fixed beyond the doorway rather than toward the camera.",
            "lighting_mood": "Fluorescent interior light outlines Yeonhee against the dark alley in a restrained, slightly desaturated low-key composition.",
            "perception_mode": "direct",
            "spatial_consistency": {
              "camera_frame_rule": {
                "applies_when": "camera_direction expresses a vertical camera height or viewpoint position 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": {
                "not_applicable": true,
                "not_applicable_reason": "이 샷의 visible_entities 에 인물이 1명이라 두 인물 사이의 공유 anchor 규칙은 성립하지 않는다",
                "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)"
              },
              "primary_framing_rule": {
                "applies_when": "framing_scale ∈ {close, medium, wide, insert} from `shot_staging.framing_scale` enum (render_strategy.framing_scale, production SOT). LLM consumes the enum only; keyword classification is forbidden.",
                "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"
                  ],
                  "subject_reference_policy_cross_ref": "see id_policy.subject_reference_policy (subject_reference_policy SOT — policy enum 에 따라 ID/outlook 사용)"
                },
                "wide_medium_rules": {
                  "not_applicable": true,
                  "not_applicable_reason": "framing_scale=close — wide/medium 규칙은 이 샷에 해당하지 않는다"
                },
                "self_check_steps": [
                  "read framing_scale enum from render_strategy.framing_scale (do NOT infer 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."
            },
            "frame_spatial_contract": {
              "reason": "looks_to_anchor",
              "constraints": [
                {
                  "constraint_id": "fsc_001",
                  "target_kind": "background",
                  "target_id": "",
                  "label": "alley beyond the open doorway",
                  "screen_zone": "middle_right",
                  "depth_plane": "background",
                  "gesture_action": "none",
                  "gesture_target_label": ""
                },
                {
                  "constraint_id": "fsc_002",
                  "target_kind": "character",
                  "target_id": "C02",
                  "label": "연희",
                  "screen_zone": "middle_left",
                  "depth_plane": "foreground",
                  "gesture_action": "looks_toward",
                  "gesture_target_label": "alley beyond the open doorway"
                }
              ]
            },
            "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",
              "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": "O02"
              }
            ],
            "must_use_composite_character_ids": true,
            "common_noun_required_when": [
              "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"
            },
            "reproduction_surface_rule": {
              "applies": false,
              "id_use": "forbidden — generic descriptor with demographic 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"
            },
            "subject_reference_policy": [],
            "constraints": [
              "ID format for visible subjects (C## bare, C##O## composite IDs, or common-noun descriptor) is determined per-subject by the id_policy per-subject policy array — see constraints[2] below",
              "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 generic descriptor with demographic instead. NOTE: the physical photo/poster/document/picture-frame/map/key prop itself MUST use its P## ID — the generic-descriptor rule applies only to a face reproduction, not to the prop entity.",
              "per-subject identity reference policy is provided in id_policy.subject_reference_policy[] — apply each subject's policy when emitting references: (a) id_and_outlook_required ⇒ subject must use C##O## composite ID; (b) base_id_required ⇒ subject must use bare C## only (no outlook); (c) generic_descriptor_allowed ⇒ subject must NOT use C## or C##O##, render with a common/demographic descriptor only. omitted subjects default to id_and_outlook_required",
              "when an ID is required for a subject (policy = id_and_outlook_required or base_id_required) and a C##O## (or C##) is introduced 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. does not apply to generic_descriptor_allowed subjects (no ID emitted)",
              "when an ID is required (policy = id_and_outlook_required or base_id_required), 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). does not apply to generic_descriptor_allowed subjects — those use a common/demographic descriptor with no C##/C##O##"
            ]
          },
          "background_binding": {
            "mode": "skipped_close_framing",
            "bg_id": "L01B02",
            "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",
              "describe the close subject and its immediate surroundings without relying on the absent background reference",
              "use [L##: ...] block for frame-edge surface description"
            ]
          },
          "continuity_elements_used": {
            "fixed_elements": [
              {
                "element_id": "fluorescent_lighting_environment",
                "element_type": "environment_state",
                "character_name": "",
                "description": "The interior of the repair shop is illuminated by an overhead fluorescent light. The lighting emits a slightly desaturated, cool-toned glow that casts moderate low-key contrast against the dark night environment. This illumination defines the stark atmosphere of the interior space and spills outward toward the doorway.",
                "applies_to_shots": [
                  1,
                  2
                ],
                "element_scope": "full",
                "source_facts": [
                  "형광등 불빛 아래 노인이 담요를 덮고 눈을 감은 채 누워 있다.",
                  "The fluorescent light gives the night interior a restrained, slightly desaturated illumination",
                  "Fluorescent interior light outlines Yeonhee against the dark alley"
                ],
                "visual_inferences": [
                  "The primary light source is positioned overhead, casting a pale and cool glow typical of utilitarian fluorescent tubes.",
                  "The lighting creates a stark contrast between the illuminated interior and the deep darkness outside."
                ],
                "creative_decisions": [
                  "Emphasize the slightly desaturated, cool tones to reflect the tense, cold night while maintaining a focus on the subjects.",
                  "Use the lighting to unify the wide interior view and the close-up at the doorway."
                ],
                "confidence": "high"
              }
            ],
            "previous_shot_refs": [
              {
                "scene_index": 1,
                "shot_index": 2,
                "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'"
                ],
                "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy applies per-subject; policy enum (id_and_outlook_required / base_id_required / generic_descriptor_allowed) determines C##/C##O## use",
                "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. third-person full body and close-up of the same character must not be combined in a single prompt.",
              "third_person_handling": "describe persons with full or partial body in frame; single-camera consistency forbids combining a partial close-up and the same character's full body in one prompt",
              "close_up_handling": "an isolated framed area of a single subject must not coexist with that subject's full-body description in the same prompt",
              "focus_on_is_focus_area_not_view_switch": "any framing-intent wording designates the area shown by the single camera position; it does not switch the view",
              "subject_reference_policy_cross_ref": "id_policy.subject_reference_policy — subjects with policy = 'base_id_required' or 'generic_descriptor_allowed' must not use outlook reference (C##O## form); additionally, subjects with policy = 'generic_descriptor_allowed' must not use bare C## either — render with a common/demographic descriptor only",
              "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 — third-person full body and close-up of the same character must not be combined in a single shot. any framing-intent wording designates the area shown by the single camera position, not a view switch (id_policy.subject_reference_policy applies per-subject)",
              "props held by characters face one direction only — toward camera OR toward character, never both"
            ]
          },
          "asset_requirements": {
            "required_refs": [
              {
                "kind": "character_outlook",
                "id": "C02O02",
                "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": "7138b4256c227783"
      }
    ],
    "consumed_bg_catalog_hash": "e57da9cd79270448",
    "consumed_shot_binding_hash": "5f0e329afd105ec9"
  },
  "step_id": "scene_detail",
  "run_id": "b728b94a-665b-4b08-a126-9982725ef9bd",
  "resolved_model": "gemini-pro",
  "updated_at": "2026-08-26T09:12:20.129199+00:00",
  "project_config_snapshot": {}
}