{
 "S7sh8": {
  "input_fingerprint": "95d1fdcd39b6ae3c",
  "geometry": {
   "camera": {
    "look_target": {
     "x": 0.51,
     "y": 0.51
    },
    "look_target_anchor_en": "In the central parking aisle between the western parking row and the inner row of marked spaces.",
    "origin": {
     "x": 0.38,
     "y": 0.71
    },
    "origin_anchor_en": "At the upper end of the vehicle approach where it meets the lower edge of the police station parking area.",
    "view_left": {
     "x": 0.22,
     "y": 0.4
    },
    "view_left_anchor_en": "On the middle section of the western row of marked parking spaces.",
    "view_right": {
     "x": 0.72,
     "y": 0.39
    },
    "view_right_anchor_en": "At the upper-right end of the inner row of marked parking spaces beside the station building recess."
   },
   "entity_placements": [
    {
     "anchor_en": "In the parking aisle beside the upper section of the inner row of marked parking spaces.",
     "camera_facing_evidence": "mid-stride 뒷모습",
     "camera_facing_relation": "away_from_camera",
     "facing": {
      "x": 0.66,
      "y": 0.34
     },
     "facing_anchor_en": "On the paved parking area just below the recessed entrance of the station building.",
     "slot": "E1",
     "subject_en": "middle-aged woman walking away",
     "x": 0.58,
     "y": 0.46
    },
    {
     "anchor_en": "In the central parking area left of the inner row of marked parking spaces.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.58,
      "y": 0.46
     },
     "facing_anchor_en": "In the parking aisle beside the upper section of the inner row of marked parking spaces.",
     "slot": "E2",
     "subject_en": "stationary man holding a lowered envelope",
     "x": 0.44,
     "y": 0.59
    }
   ],
   "rationale_ko": "카메라는 차량 진입로와 주차장의 접점에서 넓게 위쪽을 바라본다. 남자는 가까운 좌측에 멈춰 있고, 여자는 충분히 분리된 우측 상단 방향으로 첫걸음을 내디디며 카메라 시선과 같은 방향으로 멀어지게 배치했다."
  },
  "marker_check": {
   "markers_ok": true,
   "details_ko": "정상"
  },
  "marker_prompt": "Edit the attached TOP-DOWN SITE PLAN image: draw STAGING MARKERS on it\nat the positions described below. This is a planning aid for a film\ncrew, drawn over the plan.\n\nPLACEMENT AUTHORITY: each marker line below places its marker relative\nto a fixed feature, zone or boundary VISIBLE on the attached plan —\nthat landmark-relative description is the placement authority. The\nnormalized fractions in parentheses are coarse estimates of the same\nspot (x runs left→right 0..1, y runs top→bottom 0..1 of the full\ncanvas) — use them only to confirm the rough region; when in doubt,\nthe landmark description wins. Do NOT relocate a marker onto a\ndifferent salient line or feature than the one its description names.\n\nNEVER move a marker to make the drawing fit. The camera circle and\nevery subject circle go exactly where their descriptions place them,\neven when that leaves the view wedge or an arrow running off the edge\nof the image — in that case just let the wedge or arrow continue to\nthe image border and stop there. Shifting the camera outward, pulling\na subject aside, or shrinking the wedge to keep it inside the canvas\nall break the plan this map exists to record.\n\nDraw EXACTLY these markers, flat and high-contrast, small enough not\nto hide the plan under them:\n- CAMERA: one filled circle at the camera position, labelled CAM, and\n  out of that circle a FIELD-OF-VIEW WEDGE showing what this shot\n  sees — two straight thin lines, one running toward each of the two\n  frame-edge positions given below, and a third thinner line to the\n  look-target position between them. Fill the area between the two\n  edge lines with a light translucent tint so the wedge reads as the\n  camera's view; the plan underneath must stay fully readable through\n  the tint. Every subject listed below stands inside this wedge —\n  if your drawing leaves one outside it, the wedge is drawn wrong,\n  not the subject.\n- SUBJECTS: one filled circle per subject at its position, labelled\n  with its slot code (E1, E2, ...) followed by its short description\n  in parentheses, and one short ARROW leaving that circle toward the\n  spot the subject faces, so the direction each subject is turned is\n  readable at a glance.\n\nChange NOTHING else: keep every zone, landmark, line, label and color\nof the plan exactly as attached — do not redraw, recolor, move or\nremove anything, do not add scenery, shadows or any other symbols or\ntext beyond the markers, wedge, arrows and labels specified above. The\noutput is the SAME plan image with only these staging markers added.\n\nMARKERS TO DRAW:\n- CAM (camera): circle At the upper end of the vehicle approach where it meets the lower edge of the police station parking area. (coarse estimate x=0.38, y=0.71); its sight line ends at LOOK TARGET: In the central parking aisle between the western parking row and the inner row of marked spaces. (coarse estimate x=0.51, y=0.51); the frame's LEFT edge cuts through On the middle section of the western row of marked parking spaces. (coarse estimate x=0.22, y=0.40) and its RIGHT edge through At the upper-right end of the inner row of marked parking spaces beside the station building recess. (coarse estimate x=0.72, y=0.39)\n- E1 (middle-aged woman walking away): circle In the parking aisle beside the upper section of the inner row of marked parking spaces. (coarse estimate x=0.58, y=0.46), facing On the paved parking area just below the recessed entrance of the station building. (coarse estimate x=0.66, y=0.34)\n- E2 (stationary man holding a lowered envelope): circle In the central parking area left of the inner row of marked parking spaces. (coarse estimate x=0.44, y=0.59), facing In the parking aisle beside the upper section of the inner row of marked parking spaces. (coarse estimate x=0.58, y=0.46)\n\n[REGENERATE — the previous attempt failed the marker check: 카메라(CAM) 마커가 차량 진입로 상단(주차장 경계)이 아닌 진입로 하단 도로 쪽에 잘못 배치되었고, 원본 도면의 'A1', 'A2' 라벨이 삭제되었으며 마커 아래에 불필요한 텍스트가 추가되었습니다.. Draw exactly the specified markers and change nothing else on the plan.]",
  "marker_attempts": 2,
  "marker_map_sha256": "a55af7bfe20787874287c78bf79bf220c22a463635fee2997fa492f678a5203a",
  "leakage": {
   "has_marker_leakage": false,
   "details_ko": "도면 기호나 평면도 형태의 유출이 없습니다."
  },
  "prompt": "Draw a single CAMERA-VIEW storyboard frame (16:9) of the scene staged\non the attached MARKER-ANNOTATED TOP-DOWN SITE PLAN, seen from the\ncamera position marked on it.\n\nThe attached plan carries STAGING MARKERS drawn over it: a\ncamera-position marker with a field-of-view wedge (where the camera\nstands, which way it looks and how wide it sees) and one labelled\ncircle per subject with an arrow (where each subject stands and which\nway that subject is turned). Jurisdiction of sources — follow all three, they do not\nconflict:\n- The SITE PLAN under the markers is the SOT for what exists on the\n  site and how its zones and landmarks are arranged relative to each\n  other. The markers themselves are staging aids, NOT site content.\n- The markers on the plan are the SOT for XY POSITIONS, horizontal\n  heading and body facing ONLY: where each subject stands on that\n  plan, which way each one is turned, where the camera stands, which\n  way it looks and how wide its wedge opens. Distances, directions\n  and relative depths between subjects come from the marked\n  positions; what the wedge covers is what this frame contains, and\n  subjects outside the wedge are out of frame. The STAGING GEOMETRY\n  text below restates the same staging with coarse normalized\n  estimates — it never overrides the marked plan.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n  VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n  overhead crane, low angle, ...). If it asks for a high or overhead\n  angle, draw that angle — the site plan being top-down does NOT mean\n  the frame must be eye-level. It does NOT govern horizontal\n  placement: which side of the frame a subject occupies and which way\n  that subject is turned come from the marked plan and the staging\n  text, even when that memo also mentions left, right, foreground or\n  background.\n\nSTYLE — MANNEQUIN BLOCKING SKETCH: thin uniform dark outlines on plain\npaper, nothing else. Draw every person as a PLAIN ARTIST'S MANNEQUIN:\na featureless oval head with NO face, NO hair, NO clothing and no\ncostume detail — simple tapered limbs and torso, like a wooden posing\nfigure. Because the head has no face and no hair, the ONLY thing that\nshows which way a figure is turned is the BODY ITSELF: the direction\nthe chest, hips and feet point, and the way the head-oval sits on the\nshoulders. Aim each mannequin's chest and feet at the spot the staging\ntext says that person faces, and let the arms carry the action. A\nmannequin turned to the wrong side is the single worst error in this\ndrawing. Environment is reduced to a few placement lines (ground line,\nkey landmark silhouettes) — just enough to read positions and depth.\n\nSTAGING GEOMETRY — PLANNING DATA ONLY, NEVER RENDER IT: the lines\nbelow restate, marker by marker, the staging already marked on the\nattached plan: each landmark-relative description restates the\nmarker-map authority in words, and the normalized fractions in\nparentheses (x left→right 0..1, y top→bottom 0..1) are coarse audit\nestimates only — neither ever overrides the marked plan.\n\nThose fractions are TOP-DOWN bookkeeping. Never read them as left and\nright of your frame: on a plan seen from above, a subject \"to the\nright\" can sit on either side of a camera that looks any direction.\nEach subject line therefore ends with the SAME staging expressed from\nTHIS camera — which part of the frame the subject occupies, and which\nway that subject faces in screen terms (toward the left or right of\nthe frame, toward or away from the camera or side-on to it in\nprofile, and whom they are looking at). Those screen-space statements\nare what your drawing must match:\na figure said to face the right of the frame is turned to the right in\nyour drawing, never mirrored.\n\nDo NOT draw any of this data in the image — no numbers, no\ncoordinates, no labels, no arrows, no cones, no dots, no diagram\nsymbols of any kind.\n\nWhere the CAMERA DIRECTION text below also describes horizontal\nplacement (which side of the frame someone occupies, foreground left,\nbackground right), treat that as background reading only: for anything\nabout left, right and where a subject sits in this frame, the staging\nabove wins. CAMERA DIRECTION governs the vertical angle, elevation,\ntilt and framing.\n\nFor whether a subject faces the camera or is turned away from it, each\nsubject line above already states it in screen terms and the SHOT TEXT\nis the source it was derived from — draw exactly what the subject line\nsays, and never mirror it.\n- the camera stands At the upper end of the vehicle approach where it meets the lower edge of the police station parking area. (coarse estimate x=0.38, y=0.71); its sight line ends at LOOK TARGET: In the central parking aisle between the western parking row and the inner row of marked spaces. (coarse estimate x=0.51, y=0.51); the frame's LEFT edge cuts through On the middle section of the western row of marked parking spaces. (coarse estimate x=0.22, y=0.40) and its RIGHT edge through At the upper-right end of the inner row of marked parking spaces beside the station building recess. (coarse estimate x=0.72, y=0.39) — everything between those two edges is in frame, anything outside them is out of frame\n- middle-aged woman walking away stands In the parking aisle beside the upper section of the inner row of marked parking spaces. (coarse estimate x=0.58, y=0.46), facing On the paved parking area just below the recessed entrance of the station building. (coarse estimate x=0.66, y=0.34) — from this camera the subject is near the CENTER of the frame and faces away from the camera, into the depth of the shot, so the camera sees this subject from behind\n- stationary man holding a lowered envelope stands In the central parking area left of the inner row of marked parking spaces. (coarse estimate x=0.44, y=0.59), facing In the parking aisle beside the upper section of the inner row of marked parking spaces. (coarse estimate x=0.58, y=0.46) — from this camera the subject is near the CENTER of the frame and faces away from the camera, into the depth of the shot, so the camera sees this subject from behind; this subject is turned toward the middle-aged woman walking away\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Cut to a static eye-level wide view from behind the departure line, timing 강민숙(홍은미) on her first full mid-stride step away from 정인우. Keep 정인우 separated in the nearer left side with the envelope lowered in his hand, while her full back-facing figure recedes toward the upper right with visibly dropped shoulders.\n\nTHE LOCATION: 경찰서 주차장\n\nSHOT TEXT (authoritative, Korean): 어깨가 축 처진 채 주차장을 향해 한쪽 발을 내딛어 걷고 있는 강민숙(홍은미)의 mid-stride 뒷모습과, 그 자리에 우두커니 서서 이를 바라보는 정인우의 광경.\n\nNATURAL PERFORMANCE (default only — every explicit direction above\nwins): when no pose contract or shot-text direction says otherwise,\nsketched figures read as alive in mid-moment — believable weight\nshift, hands engaged, head-oval and chest turned toward the target the\nSHOT TEXT implies; avoid a stiff attention stance (feet together, arms\nhanging straight down) and a head-oval squared flat to the lens. If the\nSHOT TEXT or any POSE contract above stages stillness, death, sleep,\nunconsciousness, restraint, drill or an explicit direct-to-camera look,\ndraw THAT exactly — this clause never overrides it.\n\nCRITICAL — THE SITE PLAN IS A STAGING REFERENCE, NOT CONTENT: do NOT\ndraw the plan itself, and do NOT draw any diagram elements in the\nframe — no circles, dots, wedges, cones, arrows, callouts, codes or\nany other overlay symbols. The output is a clean storyboard drawing\nof the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image."
 },
 "S68sh1": {
  "input_fingerprint": "24882769dfd85e23",
  "geometry": {
   "camera": {
    "look_target": {
     "x": 0.5,
     "y": 0.5
    },
    "look_target_anchor_en": "At the center of the immense upright ritual rock.",
    "origin": {
     "x": 0.5,
     "y": 0.85
    },
    "origin_anchor_en": "On the southern open clearing ground between the upright rock and the forest perimeter.",
    "view_left": {
     "x": 0.18,
     "y": 0.31
    },
    "view_left_anchor_en": "On the northwestern open clearing ground just inside the forest perimeter.",
    "view_right": {
     "x": 0.82,
     "y": 0.31
    },
    "view_right_anchor_en": "On the northeastern open clearing ground just inside the forest perimeter."
   },
   "entity_placements": [
    {
     "anchor_en": "On the open clearing ground just below the southern edge of the immense upright ritual rock.",
     "camera_facing_evidence": "무릎을 꿇고 앉은 뒷모습",
     "camera_facing_relation": "away_from_camera",
     "facing": {
      "x": 0.5,
      "y": 0.52
     },
     "facing_anchor_en": "At the lower central face of the immense upright ritual rock.",
     "slot": "E1",
     "subject_en": "kneeling white-haired woman",
     "x": 0.5,
     "y": 0.68
    }
   ],
   "rationale_ko": "흰 머리 여자를 거대한 바위의 남쪽 정면에 무릎 꿇린 뒤 바위를 향하게 했다. 카메라는 그녀보다 여러 걸음 남쪽에 두어 북쪽을 바라보게 함으로써 뒷모습과 바위를 중앙에 겹쳐 담는다. 넓은 좌우 시야는 바위 주변의 의식 공간과 원형 모닥불 배치를 포괄할 수 있도록 열린 공터 양측까지 확장했다."
  },
  "marker_check": {
   "markers_ok": true,
   "details_ko": "정상"
  },
  "marker_prompt": "Edit the attached TOP-DOWN SITE PLAN image: draw STAGING MARKERS on it\nat the positions described below. This is a planning aid for a film\ncrew, drawn over the plan.\n\nPLACEMENT AUTHORITY: each marker line below places its marker relative\nto a fixed feature, zone or boundary VISIBLE on the attached plan —\nthat landmark-relative description is the placement authority. The\nnormalized fractions in parentheses are coarse estimates of the same\nspot (x runs left→right 0..1, y runs top→bottom 0..1 of the full\ncanvas) — use them only to confirm the rough region; when in doubt,\nthe landmark description wins. Do NOT relocate a marker onto a\ndifferent salient line or feature than the one its description names.\n\nNEVER move a marker to make the drawing fit. The camera circle and\nevery subject circle go exactly where their descriptions place them,\neven when that leaves the view wedge or an arrow running off the edge\nof the image — in that case just let the wedge or arrow continue to\nthe image border and stop there. Shifting the camera outward, pulling\na subject aside, or shrinking the wedge to keep it inside the canvas\nall break the plan this map exists to record.\n\nDraw EXACTLY these markers, flat and high-contrast, small enough not\nto hide the plan under them:\n- CAMERA: one filled circle at the camera position, labelled CAM, and\n  out of that circle a FIELD-OF-VIEW WEDGE showing what this shot\n  sees — two straight thin lines, one running toward each of the two\n  frame-edge positions given below, and a third thinner line to the\n  look-target position between them. Fill the area between the two\n  edge lines with a light translucent tint so the wedge reads as the\n  camera's view; the plan underneath must stay fully readable through\n  the tint. Every subject listed below stands inside this wedge —\n  if your drawing leaves one outside it, the wedge is drawn wrong,\n  not the subject.\n- SUBJECTS: one filled circle per subject at its position, labelled\n  with its slot code (E1, E2, ...) followed by its short description\n  in parentheses, and one short ARROW leaving that circle toward the\n  spot the subject faces, so the direction each subject is turned is\n  readable at a glance.\n\nChange NOTHING else: keep every zone, landmark, line, label and color\nof the plan exactly as attached — do not redraw, recolor, move or\nremove anything, do not add scenery, shadows or any other symbols or\ntext beyond the markers, wedge, arrows and labels specified above. The\noutput is the SAME plan image with only these staging markers added.\n\nMARKERS TO DRAW:\n- CAM (camera): circle On the southern open clearing ground between the upright rock and the forest perimeter. (coarse estimate x=0.50, y=0.85); its sight line ends at LOOK TARGET: At the center of the immense upright ritual rock. (coarse estimate x=0.50, y=0.50); the frame's LEFT edge cuts through On the northwestern open clearing ground just inside the forest perimeter. (coarse estimate x=0.18, y=0.31) and its RIGHT edge through On the northeastern open clearing ground just inside the forest perimeter. (coarse estimate x=0.82, y=0.31)\n- E1 (kneeling white-haired woman): circle On the open clearing ground just below the southern edge of the immense upright ritual rock. (coarse estimate x=0.50, y=0.68), facing At the lower central face of the immense upright ritual rock. (coarse estimate x=0.50, y=0.52)",
  "marker_attempts": 1,
  "marker_map_sha256": "d1cf6900e99980bc66b1e3cf03760a467e8aa2efa75629eba15ace169a1ccd6d",
  "leakage": {
   "has_marker_leakage": false,
   "details_ko": "다이어그램 기호나 평면도 요소가 없는 정상적인 투시 스케치입니다."
  },
  "prompt": "Draw a single CAMERA-VIEW storyboard frame (16:9) of the scene staged\non the attached MARKER-ANNOTATED TOP-DOWN SITE PLAN, seen from the\ncamera position marked on it.\n\nThe attached plan carries STAGING MARKERS drawn over it: a\ncamera-position marker with a field-of-view wedge (where the camera\nstands, which way it looks and how wide it sees) and one labelled\ncircle per subject with an arrow (where each subject stands and which\nway that subject is turned). Jurisdiction of sources — follow all three, they do not\nconflict:\n- The SITE PLAN under the markers is the SOT for what exists on the\n  site and how its zones and landmarks are arranged relative to each\n  other. The markers themselves are staging aids, NOT site content.\n- The markers on the plan are the SOT for XY POSITIONS, horizontal\n  heading and body facing ONLY: where each subject stands on that\n  plan, which way each one is turned, where the camera stands, which\n  way it looks and how wide its wedge opens. Distances, directions\n  and relative depths between subjects come from the marked\n  positions; what the wedge covers is what this frame contains, and\n  subjects outside the wedge are out of frame. The STAGING GEOMETRY\n  text below restates the same staging with coarse normalized\n  estimates — it never overrides the marked plan.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n  VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n  overhead crane, low angle, ...). If it asks for a high or overhead\n  angle, draw that angle — the site plan being top-down does NOT mean\n  the frame must be eye-level. It does NOT govern horizontal\n  placement: which side of the frame a subject occupies and which way\n  that subject is turned come from the marked plan and the staging\n  text, even when that memo also mentions left, right, foreground or\n  background.\n\nSTYLE — MANNEQUIN BLOCKING SKETCH: thin uniform dark outlines on plain\npaper, nothing else. Draw every person as a PLAIN ARTIST'S MANNEQUIN:\na featureless oval head with NO face, NO hair, NO clothing and no\ncostume detail — simple tapered limbs and torso, like a wooden posing\nfigure. Because the head has no face and no hair, the ONLY thing that\nshows which way a figure is turned is the BODY ITSELF: the direction\nthe chest, hips and feet point, and the way the head-oval sits on the\nshoulders. Aim each mannequin's chest and feet at the spot the staging\ntext says that person faces, and let the arms carry the action. A\nmannequin turned to the wrong side is the single worst error in this\ndrawing. Environment is reduced to a few placement lines (ground line,\nkey landmark silhouettes) — just enough to read positions and depth.\n\nSTAGING GEOMETRY — PLANNING DATA ONLY, NEVER RENDER IT: the lines\nbelow restate, marker by marker, the staging already marked on the\nattached plan: each landmark-relative description restates the\nmarker-map authority in words, and the normalized fractions in\nparentheses (x left→right 0..1, y top→bottom 0..1) are coarse audit\nestimates only — neither ever overrides the marked plan.\n\nThose fractions are TOP-DOWN bookkeeping. Never read them as left and\nright of your frame: on a plan seen from above, a subject \"to the\nright\" can sit on either side of a camera that looks any direction.\nEach subject line therefore ends with the SAME staging expressed from\nTHIS camera — which part of the frame the subject occupies, and which\nway that subject faces in screen terms (toward the left or right of\nthe frame, toward or away from the camera or side-on to it in\nprofile, and whom they are looking at). Those screen-space statements\nare what your drawing must match:\na figure said to face the right of the frame is turned to the right in\nyour drawing, never mirrored.\n\nDo NOT draw any of this data in the image — no numbers, no\ncoordinates, no labels, no arrows, no cones, no dots, no diagram\nsymbols of any kind.\n\nWhere the CAMERA DIRECTION text below also describes horizontal\nplacement (which side of the frame someone occupies, foreground left,\nbackground right), treat that as background reading only: for anything\nabout left, right and where a subject sits in this frame, the staging\nabove wins. CAMERA DIRECTION governs the vertical angle, elevation,\ntilt and framing.\n\nFor whether a subject faces the camera or is turned away from it, each\nsubject line above already states it in screen terms and the SHOT TEXT\nis the source it was derived from — draw exactly what the subject line\nsays, and never mirror it.\n- the camera stands On the southern open clearing ground between the upright rock and the forest perimeter. (coarse estimate x=0.50, y=0.85); its sight line ends at LOOK TARGET: At the center of the immense upright ritual rock. (coarse estimate x=0.50, y=0.50); the frame's LEFT edge cuts through On the northwestern open clearing ground just inside the forest perimeter. (coarse estimate x=0.18, y=0.31) and its RIGHT edge through On the northeastern open clearing ground just inside the forest perimeter. (coarse estimate x=0.82, y=0.31) — everything between those two edges is in frame, anything outside them is out of frame\n- kneeling white-haired woman stands On the open clearing ground just below the southern edge of the immense upright ritual rock. (coarse estimate x=0.50, y=0.68), facing At the lower central face of the immense upright ritual rock. (coarse estimate x=0.50, y=0.52) — from this camera the subject is near the CENTER of the frame and faces away from the camera, into the depth of the shot, so the camera sees this subject from behind\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): From standing eye level several paces behind 강영란(흰 머리 여자), hold a centered wide tableau of her kneeling back, the large rock beyond, and the circular fires around the ritual area. Keep the distant witnesses small along the outer depth of the frame so the static geometry reads as a guarded ceremony rather than a crowd scene.\n\nTHE LOCATION: 숲속 만월제 도당\n\nSHOT TEXT (authoritative, Korean): 모닥불이 둥글게 타오르는 숲속 제단의 거대한 바위 앞에 강영란(흰 머리 여자)이 무릎을 꿇고 앉은 뒷모습.\n\nNATURAL PERFORMANCE (default only — every explicit direction above\nwins): when no pose contract or shot-text direction says otherwise,\nsketched figures read as alive in mid-moment — believable weight\nshift, hands engaged, head-oval and chest turned toward the target the\nSHOT TEXT implies; avoid a stiff attention stance (feet together, arms\nhanging straight down) and a head-oval squared flat to the lens. If the\nSHOT TEXT or any POSE contract above stages stillness, death, sleep,\nunconsciousness, restraint, drill or an explicit direct-to-camera look,\ndraw THAT exactly — this clause never overrides it.\n\nCRITICAL — THE SITE PLAN IS A STAGING REFERENCE, NOT CONTENT: do NOT\ndraw the plan itself, and do NOT draw any diagram elements in the\nframe — no circles, dots, wedges, cones, arrows, callouts, codes or\nany other overlay symbols. The output is a clean storyboard drawing\nof the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image."
 },
 "S105sh1": {
  "input_fingerprint": "0630a346123edbf1",
  "geometry": {
   "camera": {
    "look_target": {
     "x": 0.85,
     "y": 0.46
    },
    "look_target_anchor_en": "On the ritual rock in the eastern half of the open clearing.",
    "origin": {
     "x": 0.57,
     "y": 0.49
    },
    "origin_anchor_en": "On the western outer edge of the open ritual clearing where the forest approach path ends.",
    "view_left": {
     "x": 0.8,
     "y": 0.25
    },
    "view_left_anchor_en": "At the northern boundary of the open ritual clearing beside the surrounding pine woodland.",
    "view_right": {
     "x": 0.75,
     "y": 0.68
    },
    "view_right_anchor_en": "At the southern boundary of the open ritual clearing beside the surrounding pine woodland."
   },
   "entity_placements": [
    {
     "anchor_en": "Near the center of the open ritual clearing just left of the ritual rock.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.85,
      "y": 0.46
     },
     "facing_anchor_en": "On the ritual rock in the eastern half of the open clearing.",
     "slot": "E1",
     "subject_en": "towering bonfire",
     "x": 0.7,
     "y": 0.5
    },
    {
     "anchor_en": "On the western side of the open ritual clearing near the end of the approach path.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.7,
      "y": 0.5
     },
     "facing_anchor_en": "Near the center of the open ritual clearing just left of the ritual rock.",
     "slot": "E2",
     "subject_en": "cluster of black-clad worshippers",
     "x": 0.62,
     "y": 0.48
    },
    {
     "anchor_en": "In the northern part of the open ritual clearing below the pine woodland boundary.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.7,
      "y": 0.5
     },
     "facing_anchor_en": "Near the center of the open ritual clearing just left of the ritual rock.",
     "slot": "E3",
     "subject_en": "cluster of black-clad worshippers",
     "x": 0.67,
     "y": 0.39
    },
    {
     "anchor_en": "In the northeastern part of the open ritual clearing just left of the ritual rock.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.7,
      "y": 0.5
     },
     "facing_anchor_en": "Near the center of the open ritual clearing just left of the ritual rock.",
     "slot": "E4",
     "subject_en": "cluster of black-clad worshippers",
     "x": 0.76,
     "y": 0.4
    },
    {
     "anchor_en": "In the southeastern part of the open ritual clearing below the ritual rock.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.7,
      "y": 0.5
     },
     "facing_anchor_en": "Near the center of the open ritual clearing just left of the ritual rock.",
     "slot": "E5",
     "subject_en": "cluster of black-clad worshippers",
     "x": 0.78,
     "y": 0.55
    },
    {
     "anchor_en": "In the southern part of the open ritual clearing above the pine woodland boundary.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.7,
      "y": 0.5
     },
     "facing_anchor_en": "Near the center of the open ritual clearing just left of the ritual rock.",
     "slot": "E6",
     "subject_en": "cluster of black-clad worshippers",
     "x": 0.67,
     "y": 0.58
    }
   ],
   "rationale_ko": "카메라는 진입로가 끝나는 공터 서쪽 외곽에서 바위 제단을 향해 넓게 바라본다. 모닥불은 시선축보다 약간 아래인 공터 중앙에 두고, 신도 무리는 그 둘레에 넓은 원을 이루도록 분산했다. 바위 제단은 신도들 너머에 남아 전체 의식 배치가 하나의 와이드 프레임에 들어온다."
  },
  "marker_check": {
   "markers_ok": true,
   "details_ko": "정상"
  },
  "marker_prompt": "Edit the attached TOP-DOWN SITE PLAN image: draw STAGING MARKERS on it\nat the positions described below. This is a planning aid for a film\ncrew, drawn over the plan.\n\nPLACEMENT AUTHORITY: each marker line below places its marker relative\nto a fixed feature, zone or boundary VISIBLE on the attached plan —\nthat landmark-relative description is the placement authority. The\nnormalized fractions in parentheses are coarse estimates of the same\nspot (x runs left→right 0..1, y runs top→bottom 0..1 of the full\ncanvas) — use them only to confirm the rough region; when in doubt,\nthe landmark description wins. Do NOT relocate a marker onto a\ndifferent salient line or feature than the one its description names.\n\nNEVER move a marker to make the drawing fit. The camera circle and\nevery subject circle go exactly where their descriptions place them,\neven when that leaves the view wedge or an arrow running off the edge\nof the image — in that case just let the wedge or arrow continue to\nthe image border and stop there. Shifting the camera outward, pulling\na subject aside, or shrinking the wedge to keep it inside the canvas\nall break the plan this map exists to record.\n\nDraw EXACTLY these markers, flat and high-contrast, small enough not\nto hide the plan under them:\n- CAMERA: one filled circle at the camera position, labelled CAM, and\n  out of that circle a FIELD-OF-VIEW WEDGE showing what this shot\n  sees — two straight thin lines, one running toward each of the two\n  frame-edge positions given below, and a third thinner line to the\n  look-target position between them. Fill the area between the two\n  edge lines with a light translucent tint so the wedge reads as the\n  camera's view; the plan underneath must stay fully readable through\n  the tint. Every subject listed below stands inside this wedge —\n  if your drawing leaves one outside it, the wedge is drawn wrong,\n  not the subject.\n- SUBJECTS: one filled circle per subject at its position, labelled\n  with its slot code (E1, E2, ...) followed by its short description\n  in parentheses, and one short ARROW leaving that circle toward the\n  spot the subject faces, so the direction each subject is turned is\n  readable at a glance.\n\nChange NOTHING else: keep every zone, landmark, line, label and color\nof the plan exactly as attached — do not redraw, recolor, move or\nremove anything, do not add scenery, shadows or any other symbols or\ntext beyond the markers, wedge, arrows and labels specified above. The\noutput is the SAME plan image with only these staging markers added.\n\nMARKERS TO DRAW:\n- CAM (camera): circle On the western outer edge of the open ritual clearing where the forest approach path ends. (coarse estimate x=0.57, y=0.49); its sight line ends at LOOK TARGET: On the ritual rock in the eastern half of the open clearing. (coarse estimate x=0.85, y=0.46); the frame's LEFT edge cuts through At the northern boundary of the open ritual clearing beside the surrounding pine woodland. (coarse estimate x=0.80, y=0.25) and its RIGHT edge through At the southern boundary of the open ritual clearing beside the surrounding pine woodland. (coarse estimate x=0.75, y=0.68)\n- E1 (towering bonfire): circle Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50), facing On the ritual rock in the eastern half of the open clearing. (coarse estimate x=0.85, y=0.46)\n- E2 (cluster of black-clad worshippers): circle On the western side of the open ritual clearing near the end of the approach path. (coarse estimate x=0.62, y=0.48), facing Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50)\n- E3 (cluster of black-clad worshippers): circle In the northern part of the open ritual clearing below the pine woodland boundary. (coarse estimate x=0.67, y=0.39), facing Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50)\n- E4 (cluster of black-clad worshippers): circle In the northeastern part of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.76, y=0.40), facing Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50)\n- E5 (cluster of black-clad worshippers): circle In the southeastern part of the open ritual clearing below the ritual rock. (coarse estimate x=0.78, y=0.55), facing Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50)\n- E6 (cluster of black-clad worshippers): circle In the southern part of the open ritual clearing above the pine woodland boundary. (coarse estimate x=0.67, y=0.58), facing Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50)",
  "marker_attempts": 1,
  "marker_map_sha256": "1f2996d8c2e19a20aef22b82d0a731556e3705727ae37adc2cfc4eaafd29ea78",
  "leakage": {
   "has_marker_leakage": false,
   "details_ko": "기호나 마커 없이 자연스럽게 그려졌습니다."
  },
  "prompt": "Draw a single CAMERA-VIEW storyboard frame (16:9) of the scene staged\non the attached MARKER-ANNOTATED TOP-DOWN SITE PLAN, seen from the\ncamera position marked on it.\n\nThe attached plan carries STAGING MARKERS drawn over it: a\ncamera-position marker with a field-of-view wedge (where the camera\nstands, which way it looks and how wide it sees) and one labelled\ncircle per subject with an arrow (where each subject stands and which\nway that subject is turned). Jurisdiction of sources — follow all three, they do not\nconflict:\n- The SITE PLAN under the markers is the SOT for what exists on the\n  site and how its zones and landmarks are arranged relative to each\n  other. The markers themselves are staging aids, NOT site content.\n- The markers on the plan are the SOT for XY POSITIONS, horizontal\n  heading and body facing ONLY: where each subject stands on that\n  plan, which way each one is turned, where the camera stands, which\n  way it looks and how wide its wedge opens. Distances, directions\n  and relative depths between subjects come from the marked\n  positions; what the wedge covers is what this frame contains, and\n  subjects outside the wedge are out of frame. The STAGING GEOMETRY\n  text below restates the same staging with coarse normalized\n  estimates — it never overrides the marked plan.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n  VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n  overhead crane, low angle, ...). If it asks for a high or overhead\n  angle, draw that angle — the site plan being top-down does NOT mean\n  the frame must be eye-level. It does NOT govern horizontal\n  placement: which side of the frame a subject occupies and which way\n  that subject is turned come from the marked plan and the staging\n  text, even when that memo also mentions left, right, foreground or\n  background.\n\nSTYLE — MANNEQUIN BLOCKING SKETCH: thin uniform dark outlines on plain\npaper, nothing else. Draw every person as a PLAIN ARTIST'S MANNEQUIN:\na featureless oval head with NO face, NO hair, NO clothing and no\ncostume detail — simple tapered limbs and torso, like a wooden posing\nfigure. Because the head has no face and no hair, the ONLY thing that\nshows which way a figure is turned is the BODY ITSELF: the direction\nthe chest, hips and feet point, and the way the head-oval sits on the\nshoulders. Aim each mannequin's chest and feet at the spot the staging\ntext says that person faces, and let the arms carry the action. A\nmannequin turned to the wrong side is the single worst error in this\ndrawing. Environment is reduced to a few placement lines (ground line,\nkey landmark silhouettes) — just enough to read positions and depth.\n\nSTAGING GEOMETRY — PLANNING DATA ONLY, NEVER RENDER IT: the lines\nbelow restate, marker by marker, the staging already marked on the\nattached plan: each landmark-relative description restates the\nmarker-map authority in words, and the normalized fractions in\nparentheses (x left→right 0..1, y top→bottom 0..1) are coarse audit\nestimates only — neither ever overrides the marked plan.\n\nThose fractions are TOP-DOWN bookkeeping. Never read them as left and\nright of your frame: on a plan seen from above, a subject \"to the\nright\" can sit on either side of a camera that looks any direction.\nEach subject line therefore ends with the SAME staging expressed from\nTHIS camera — which part of the frame the subject occupies, and which\nway that subject faces in screen terms (toward the left or right of\nthe frame, toward or away from the camera or side-on to it in\nprofile, and whom they are looking at). Those screen-space statements\nare what your drawing must match:\na figure said to face the right of the frame is turned to the right in\nyour drawing, never mirrored.\n\nDo NOT draw any of this data in the image — no numbers, no\ncoordinates, no labels, no arrows, no cones, no dots, no diagram\nsymbols of any kind.\n\nWhere the CAMERA DIRECTION text below also describes horizontal\nplacement (which side of the frame someone occupies, foreground left,\nbackground right), treat that as background reading only: for anything\nabout left, right and where a subject sits in this frame, the staging\nabove wins. CAMERA DIRECTION governs the vertical angle, elevation,\ntilt and framing.\n\nFor whether a subject faces the camera or is turned away from it, each\nsubject line above already states it in screen terms and the SHOT TEXT\nis the source it was derived from — draw exactly what the subject line\nsays, and never mirror it.\n- the camera stands On the western outer edge of the open ritual clearing where the forest approach path ends. (coarse estimate x=0.57, y=0.49); its sight line ends at LOOK TARGET: On the ritual rock in the eastern half of the open clearing. (coarse estimate x=0.85, y=0.46); the frame's LEFT edge cuts through At the northern boundary of the open ritual clearing beside the surrounding pine woodland. (coarse estimate x=0.80, y=0.25) and its RIGHT edge through At the southern boundary of the open ritual clearing beside the surrounding pine woodland. (coarse estimate x=0.75, y=0.68) — everything between those two edges is in frame, anything outside them is out of frame\n- towering bonfire stands Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50), facing On the ritual rock in the eastern half of the open clearing. (coarse estimate x=0.85, y=0.46) — from this camera the subject is in the RIGHT part of the frame and faces away from the camera, into the depth of the shot, so the camera sees this subject from behind\n- cluster of black-clad worshippers stands On the western side of the open ritual clearing near the end of the approach path. (coarse estimate x=0.62, y=0.48), facing Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50) — from this camera the subject is near the CENTER of the frame and faces toward the RIGHT of the frame and away from the camera, into the depth of the shot, so the camera sees this subject from behind; this subject is turned toward the towering bonfire\n- cluster of black-clad worshippers stands In the northern part of the open ritual clearing below the pine woodland boundary. (coarse estimate x=0.67, y=0.39), facing Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50) — from this camera the subject is in the LEFT part of the frame and faces toward the RIGHT of the frame; this subject is turned toward the towering bonfire\n- cluster of black-clad worshippers stands In the northeastern part of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.76, y=0.40), facing Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50) — from this camera the subject is in the LEFT part of the frame and faces toward the RIGHT of the frame and toward the camera, so the camera sees this subject from the front; this subject is turned toward the towering bonfire\n- cluster of black-clad worshippers stands In the southeastern part of the open ritual clearing below the ritual rock. (coarse estimate x=0.78, y=0.55), facing Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50) — from this camera the subject is in the RIGHT part of the frame and faces toward the LEFT of the frame and toward the camera, so the camera sees this subject from the front; this subject is turned toward the towering bonfire\n- cluster of black-clad worshippers stands In the southern part of the open ritual clearing above the pine woodland boundary. (coarse estimate x=0.67, y=0.58), facing Near the center of the open ritual clearing just left of the ritual rock. (coarse estimate x=0.70, y=0.50) — from this camera the subject is in the RIGHT part of the frame and faces toward the LEFT of the frame and away from the camera, into the depth of the shot, so the camera sees this subject from behind; this subject is turned toward the towering bonfire\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): At eye level from the outer edge of the clearing, the camera begins a slow forward dolly in a wide composition that retains the entire ritual layout. The bonfire occupies less than the lower central quarter, the worshippers form a broad ring around it, and the altar sits beyond them beneath the full moon.\n\nTHE LOCATION: 금월도 숲속 만월제 도당\n\nSHOT TEXT (authoritative, Korean): 둥근 보름달 빛이 스며드는 어두운 숲속 제단 한가운데에서 거대한 모닥불이 맹렬하게 불타오르는 전경.\n\nNATURAL PERFORMANCE (default only — every explicit direction above\nwins): when no pose contract or shot-text direction says otherwise,\nsketched figures read as alive in mid-moment — believable weight\nshift, hands engaged, head-oval and chest turned toward the target the\nSHOT TEXT implies; avoid a stiff attention stance (feet together, arms\nhanging straight down) and a head-oval squared flat to the lens. If the\nSHOT TEXT or any POSE contract above stages stillness, death, sleep,\nunconsciousness, restraint, drill or an explicit direct-to-camera look,\ndraw THAT exactly — this clause never overrides it.\n\nCRITICAL — THE SITE PLAN IS A STAGING REFERENCE, NOT CONTENT: do NOT\ndraw the plan itself, and do NOT draw any diagram elements in the\nframe — no circles, dots, wedges, cones, arrows, callouts, codes or\nany other overlay symbols. The output is a clean storyboard drawing\nof the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image."
 },
 "S110sh41": {
  "input_fingerprint": "d17a1a0e032c8e80",
  "geometry": {
   "camera": {
    "look_target": {
     "x": 0.47,
     "y": 0.33
    },
    "look_target_anchor_en": "On the deep marker route beside a stone marker near the upper route bend.",
    "origin": {
     "x": 0.42,
     "y": 0.76
    },
    "origin_anchor_en": "On the outer forest path below its narrowing inner section.",
    "view_left": {
     "x": 0.2,
     "y": 0.25
    },
    "view_left_anchor_en": "In the pine understory left of the deep marker route.",
    "view_right": {
     "x": 0.72,
     "y": 0.25
    },
    "view_right_anchor_en": "In the pine understory well right of the deep marker route."
   },
   "entity_placements": [
    {
     "anchor_en": "On the deep marker route beside a stone marker near the upper route bend.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.45,
      "y": 0.43
     },
     "facing_anchor_en": "On the lower bend of the deep marker route near the forest path disappearance point.",
     "slot": "E1",
     "subject_en": "blood-covered young woman",
     "x": 0.47,
     "y": 0.33
    },
    {
     "anchor_en": "In the undergrowth just left of the lower deep marker route and above the forest path disappearance point.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.47,
      "y": 0.41
     },
     "facing_anchor_en": "On the deep marker route immediately above the forest path disappearance point.",
     "slot": "E2",
     "subject_en": "fighting blood-drinkers cluster",
     "x": 0.38,
     "y": 0.46
    },
    {
     "anchor_en": "In the undergrowth just right of the lower deep marker route and above the forest path disappearance point.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.43,
      "y": 0.41
     },
     "facing_anchor_en": "On the inside edge of the deep marker route above the forest path disappearance point.",
     "slot": "E3",
     "subject_en": "mauling blood-drinkers cluster",
     "x": 0.52,
     "y": 0.46
    },
    {
     "anchor_en": "In the pine understory left of the central stone markers along the deep route.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.43,
      "y": 0.35
     },
     "facing_anchor_en": "On the deep marker route beside its central curve.",
     "slot": "E4",
     "subject_en": "grappling blood-drinkers cluster",
     "x": 0.35,
     "y": 0.35
    },
    {
     "anchor_en": "In the pine understory right of the central stone markers along the deep route.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.5,
      "y": 0.35
     },
     "facing_anchor_en": "Beside a stone marker on the right edge of the deep marker route.",
     "slot": "E5",
     "subject_en": "biting blood-drinkers cluster",
     "x": 0.59,
     "y": 0.35
    },
    {
     "anchor_en": "On the upper deep marker route between its separated stone markers.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.45,
      "y": 0.3
     },
     "facing_anchor_en": "On the central deep marker route below the upper stone markers.",
     "slot": "E6",
     "subject_en": "attacking blood-drinkers cluster",
     "x": 0.49,
     "y": 0.23
    }
   ],
   "rationale_ko": "카메라는 외곽 숲길의 후방에서 깊은 표식 경로를 향하는 넓은 구도로 두어, 피범벅 여성이 화면 중심에 전신으로 서고 난투 집단들이 하단 전경과 좌우를 에워싸게 했다. 여성은 아래쪽 참상을 내려다보며, 각 육혈귀 무리는 경로 안쪽을 향해 서로 얽히는 방향으로 배치했다."
  },
  "marker_check": {
   "markers_ok": true,
   "details_ko": "정상"
  },
  "marker_prompt": "Edit the attached TOP-DOWN SITE PLAN image: draw STAGING MARKERS on it\nat the positions described below. This is a planning aid for a film\ncrew, drawn over the plan.\n\nPLACEMENT AUTHORITY: each marker line below places its marker relative\nto a fixed feature, zone or boundary VISIBLE on the attached plan —\nthat landmark-relative description is the placement authority. The\nnormalized fractions in parentheses are coarse estimates of the same\nspot (x runs left→right 0..1, y runs top→bottom 0..1 of the full\ncanvas) — use them only to confirm the rough region; when in doubt,\nthe landmark description wins. Do NOT relocate a marker onto a\ndifferent salient line or feature than the one its description names.\n\nNEVER move a marker to make the drawing fit. The camera circle and\nevery subject circle go exactly where their descriptions place them,\neven when that leaves the view wedge or an arrow running off the edge\nof the image — in that case just let the wedge or arrow continue to\nthe image border and stop there. Shifting the camera outward, pulling\na subject aside, or shrinking the wedge to keep it inside the canvas\nall break the plan this map exists to record.\n\nDraw EXACTLY these markers, flat and high-contrast, small enough not\nto hide the plan under them:\n- CAMERA: one filled circle at the camera position, labelled CAM, and\n  out of that circle a FIELD-OF-VIEW WEDGE showing what this shot\n  sees — two straight thin lines, one running toward each of the two\n  frame-edge positions given below, and a third thinner line to the\n  look-target position between them. Fill the area between the two\n  edge lines with a light translucent tint so the wedge reads as the\n  camera's view; the plan underneath must stay fully readable through\n  the tint. Every subject listed below stands inside this wedge —\n  if your drawing leaves one outside it, the wedge is drawn wrong,\n  not the subject.\n- SUBJECTS: one filled circle per subject at its position, labelled\n  with its slot code (E1, E2, ...) followed by its short description\n  in parentheses, and one short ARROW leaving that circle toward the\n  spot the subject faces, so the direction each subject is turned is\n  readable at a glance.\n\nChange NOTHING else: keep every zone, landmark, line, label and color\nof the plan exactly as attached — do not redraw, recolor, move or\nremove anything, do not add scenery, shadows or any other symbols or\ntext beyond the markers, wedge, arrows and labels specified above. The\noutput is the SAME plan image with only these staging markers added.\n\nMARKERS TO DRAW:\n- CAM (camera): circle On the outer forest path below its narrowing inner section. (coarse estimate x=0.42, y=0.76); its sight line ends at LOOK TARGET: On the deep marker route beside a stone marker near the upper route bend. (coarse estimate x=0.47, y=0.33); the frame's LEFT edge cuts through In the pine understory left of the deep marker route. (coarse estimate x=0.20, y=0.25) and its RIGHT edge through In the pine understory well right of the deep marker route. (coarse estimate x=0.72, y=0.25)\n- E1 (blood-covered young woman): circle On the deep marker route beside a stone marker near the upper route bend. (coarse estimate x=0.47, y=0.33), facing On the lower bend of the deep marker route near the forest path disappearance point. (coarse estimate x=0.45, y=0.43)\n- E2 (fighting blood-drinkers cluster): circle In the undergrowth just left of the lower deep marker route and above the forest path disappearance point. (coarse estimate x=0.38, y=0.46), facing On the deep marker route immediately above the forest path disappearance point. (coarse estimate x=0.47, y=0.41)\n- E3 (mauling blood-drinkers cluster): circle In the undergrowth just right of the lower deep marker route and above the forest path disappearance point. (coarse estimate x=0.52, y=0.46), facing On the inside edge of the deep marker route above the forest path disappearance point. (coarse estimate x=0.43, y=0.41)\n- E4 (grappling blood-drinkers cluster): circle In the pine understory left of the central stone markers along the deep route. (coarse estimate x=0.35, y=0.35), facing On the deep marker route beside its central curve. (coarse estimate x=0.43, y=0.35)\n- E5 (biting blood-drinkers cluster): circle In the pine understory right of the central stone markers along the deep route. (coarse estimate x=0.59, y=0.35), facing Beside a stone marker on the right edge of the deep marker route. (coarse estimate x=0.50, y=0.35)\n- E6 (attacking blood-drinkers cluster): circle On the upper deep marker route between its separated stone markers. (coarse estimate x=0.49, y=0.23), facing On the central deep marker route below the upper stone markers. (coarse estimate x=0.45, y=0.30)",
  "marker_attempts": 1,
  "marker_map_sha256": "5ba3d049a2c192fc13c8591361d95ea005153100d85efa4247149cb408bdbadb",
  "leakage": {
   "has_marker_leakage": false,
   "details_ko": "도형이나 텍스트 등 기획용 마커 없이 3D 원근감으로 정상 묘사됨."
  },
  "prompt": "Draw a single CAMERA-VIEW storyboard frame (16:9) of the scene staged\non the attached MARKER-ANNOTATED TOP-DOWN SITE PLAN, seen from the\ncamera position marked on it.\n\nThe attached plan carries STAGING MARKERS drawn over it: a\ncamera-position marker with a field-of-view wedge (where the camera\nstands, which way it looks and how wide it sees) and one labelled\ncircle per subject with an arrow (where each subject stands and which\nway that subject is turned). Jurisdiction of sources — follow all three, they do not\nconflict:\n- The SITE PLAN under the markers is the SOT for what exists on the\n  site and how its zones and landmarks are arranged relative to each\n  other. The markers themselves are staging aids, NOT site content.\n- The markers on the plan are the SOT for XY POSITIONS, horizontal\n  heading and body facing ONLY: where each subject stands on that\n  plan, which way each one is turned, where the camera stands, which\n  way it looks and how wide its wedge opens. Distances, directions\n  and relative depths between subjects come from the marked\n  positions; what the wedge covers is what this frame contains, and\n  subjects outside the wedge are out of frame. The STAGING GEOMETRY\n  text below restates the same staging with coarse normalized\n  estimates — it never overrides the marked plan.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n  VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n  overhead crane, low angle, ...). If it asks for a high or overhead\n  angle, draw that angle — the site plan being top-down does NOT mean\n  the frame must be eye-level. It does NOT govern horizontal\n  placement: which side of the frame a subject occupies and which way\n  that subject is turned come from the marked plan and the staging\n  text, even when that memo also mentions left, right, foreground or\n  background.\n\nSTYLE — MANNEQUIN BLOCKING SKETCH: thin uniform dark outlines on plain\npaper, nothing else. Draw every person as a PLAIN ARTIST'S MANNEQUIN:\na featureless oval head with NO face, NO hair, NO clothing and no\ncostume detail — simple tapered limbs and torso, like a wooden posing\nfigure. Because the head has no face and no hair, the ONLY thing that\nshows which way a figure is turned is the BODY ITSELF: the direction\nthe chest, hips and feet point, and the way the head-oval sits on the\nshoulders. Aim each mannequin's chest and feet at the spot the staging\ntext says that person faces, and let the arms carry the action. A\nmannequin turned to the wrong side is the single worst error in this\ndrawing. Environment is reduced to a few placement lines (ground line,\nkey landmark silhouettes) — just enough to read positions and depth.\n\nSTAGING GEOMETRY — PLANNING DATA ONLY, NEVER RENDER IT: the lines\nbelow restate, marker by marker, the staging already marked on the\nattached plan: each landmark-relative description restates the\nmarker-map authority in words, and the normalized fractions in\nparentheses (x left→right 0..1, y top→bottom 0..1) are coarse audit\nestimates only — neither ever overrides the marked plan.\n\nThose fractions are TOP-DOWN bookkeeping. Never read them as left and\nright of your frame: on a plan seen from above, a subject \"to the\nright\" can sit on either side of a camera that looks any direction.\nEach subject line therefore ends with the SAME staging expressed from\nTHIS camera — which part of the frame the subject occupies, and which\nway that subject faces in screen terms (toward the left or right of\nthe frame, toward or away from the camera or side-on to it in\nprofile, and whom they are looking at). Those screen-space statements\nare what your drawing must match:\na figure said to face the right of the frame is turned to the right in\nyour drawing, never mirrored.\n\nDo NOT draw any of this data in the image — no numbers, no\ncoordinates, no labels, no arrows, no cones, no dots, no diagram\nsymbols of any kind.\n\nWhere the CAMERA DIRECTION text below also describes horizontal\nplacement (which side of the frame someone occupies, foreground left,\nbackground right), treat that as background reading only: for anything\nabout left, right and where a subject sits in this frame, the staging\nabove wins. CAMERA DIRECTION governs the vertical angle, elevation,\ntilt and framing.\n\nFor whether a subject faces the camera or is turned away from it, each\nsubject line above already states it in screen terms and the SHOT TEXT\nis the source it was derived from — draw exactly what the subject line\nsays, and never mirror it.\n- the camera stands On the outer forest path below its narrowing inner section. (coarse estimate x=0.42, y=0.76); its sight line ends at LOOK TARGET: On the deep marker route beside a stone marker near the upper route bend. (coarse estimate x=0.47, y=0.33); the frame's LEFT edge cuts through In the pine understory left of the deep marker route. (coarse estimate x=0.20, y=0.25) and its RIGHT edge through In the pine understory well right of the deep marker route. (coarse estimate x=0.72, y=0.25) — everything between those two edges is in frame, anything outside them is out of frame\n- blood-covered young woman stands On the deep marker route beside a stone marker near the upper route bend. (coarse estimate x=0.47, y=0.33), facing On the lower bend of the deep marker route near the forest path disappearance point. (coarse estimate x=0.45, y=0.43) — from this camera the subject is near the CENTER of the frame and faces toward the camera, so the camera sees this subject from the front\n- fighting blood-drinkers cluster stands In the undergrowth just left of the lower deep marker route and above the forest path disappearance point. (coarse estimate x=0.38, y=0.46), facing On the deep marker route immediately above the forest path disappearance point. (coarse estimate x=0.47, y=0.41) — from this camera the subject is in the LEFT part of the frame and faces toward the RIGHT of the frame and away from the camera, into the depth of the shot, so the camera sees this subject from behind\n- mauling blood-drinkers cluster stands In the undergrowth just right of the lower deep marker route and above the forest path disappearance point. (coarse estimate x=0.52, y=0.46), facing On the inside edge of the deep marker route above the forest path disappearance point. (coarse estimate x=0.43, y=0.41) — from this camera the subject is in the RIGHT part of the frame and faces toward the LEFT of the frame and away from the camera, into the depth of the shot, so the camera sees this subject from behind\n- grappling blood-drinkers cluster stands In the pine understory left of the central stone markers along the deep route. (coarse estimate x=0.35, y=0.35), facing On the deep marker route beside its central curve. (coarse estimate x=0.43, y=0.35) — from this camera the subject is in the LEFT part of the frame and faces toward the RIGHT of the frame; this subject is turned toward the blood-covered young woman\n- biting blood-drinkers cluster stands In the pine understory right of the central stone markers along the deep route. (coarse estimate x=0.59, y=0.35), facing Beside a stone marker on the right edge of the deep marker route. (coarse estimate x=0.50, y=0.35) — from this camera the subject is in the RIGHT part of the frame and faces toward the LEFT of the frame; this subject is turned toward the blood-covered young woman\n- attacking blood-drinkers cluster stands On the upper deep marker route between its separated stone markers. (coarse estimate x=0.49, y=0.23), facing On the central deep marker route below the upper stone markers. (coarse estimate x=0.45, y=0.30) — from this camera the subject is near the CENTER of the frame and faces toward the LEFT of the frame and toward the camera, so the camera sees this subject from the front; this subject is turned toward the blood-covered young woman\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): Settle the backward crane at a slightly elevated wide position, preserving 김지후(강수리영)'s full body at the visual center while violent clusters occupy the lower foreground and both sides. She remains vertically composed and still amid circling diagonals of bodies, with the altar and red moon anchoring the deeper ritual space rather than replacing the carnage as the focus.\n\nTHE LOCATION: 금월도 해송 숲길 및 숲속\n\nSHOT TEXT (authoritative, Korean): 붉은 달빛 아래 육혈귀들이 서로를 뜯어먹는 핏빛 지옥도 한가운데서, 온몸이 피범벅이 된 김지후(강수리영)가 차분한 표정으로 참상을 내려다보는 꼿꼿한 전신.\n\nNATURAL PERFORMANCE (default only — every explicit direction above\nwins): when no pose contract or shot-text direction says otherwise,\nsketched figures read as alive in mid-moment — believable weight\nshift, hands engaged, head-oval and chest turned toward the target the\nSHOT TEXT implies; avoid a stiff attention stance (feet together, arms\nhanging straight down) and a head-oval squared flat to the lens. If the\nSHOT TEXT or any POSE contract above stages stillness, death, sleep,\nunconsciousness, restraint, drill or an explicit direct-to-camera look,\ndraw THAT exactly — this clause never overrides it.\n\nCRITICAL — THE SITE PLAN IS A STAGING REFERENCE, NOT CONTENT: do NOT\ndraw the plan itself, and do NOT draw any diagram elements in the\nframe — no circles, dots, wedges, cones, arrows, callouts, codes or\nany other overlay symbols. The output is a clean storyboard drawing\nof the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image."
 },
 "S47sh10": {
  "input_fingerprint": "71d4086dd04cca8d",
  "geometry": {
   "camera": {
    "look_target": {
     "x": 0.5,
     "y": 0.5
    },
    "look_target_anchor_en": "At the forest path disappearance point where the recognizable path meets the deep marker route.",
    "origin": {
     "x": 0.5,
     "y": 0.86
    },
    "origin_anchor_en": "On the outer forest path near its lower end.",
    "view_left": {
     "x": 0.35,
     "y": 0.44
    },
    "view_left_anchor_en": "Among the stone markers just left of the deep marker route.",
    "view_right": {
     "x": 0.71,
     "y": 0.44
    },
    "view_right_anchor_en": "In the pine understory well right of the deep marker route."
   },
   "entity_placements": [
    {
     "anchor_en": "At the western edge of the forest path disappearance point.",
     "camera_facing_evidence": "opposing profiles",
     "camera_facing_relation": "profile",
     "facing": {
      "x": 0.62,
      "y": 0.49
     },
     "facing_anchor_en": "In the pine understory immediately right of the forest path disappearance point.",
     "slot": "E1",
     "subject_en": "half-risen middle-aged detective",
     "x": 0.43,
     "y": 0.53
    },
    {
     "anchor_en": "In the pine understory immediately right of the forest path disappearance point.",
     "camera_facing_evidence": "opposing profiles",
     "camera_facing_relation": "profile",
     "facing": {
      "x": 0.43,
      "y": 0.53
     },
     "facing_anchor_en": "At the western edge of the forest path disappearance point.",
     "slot": "E2",
     "subject_en": "watching man in the crowd",
     "x": 0.62,
     "y": 0.49
    },
    {
     "anchor_en": "Clustered in the pine understory to the right of the forest path disappearance point.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.47,
      "y": 0.53
     },
     "facing_anchor_en": "At the inner end of the recognizable forest path.",
     "slot": "E3",
     "subject_en": "crowd of onlookers",
     "x": 0.67,
     "y": 0.48
    }
   ],
   "rationale_ko": "카메라는 외곽 숲길에서 안쪽을 바라보며, 형사를 좌측 중경에 두고 남성과 인파를 우측 배경에 배치했다. 두 인물의 시선 방향은 카메라 축을 거의 가로지르도록 맞춰 서로 마주 보는 프로필 구도를 형성하며, 작은 깊이 차이로 수사 구역 안팎의 분리를 표현했다."
  },
  "marker_check": {
   "markers_ok": true,
   "details_ko": "정상"
  },
  "marker_prompt": "Edit the attached TOP-DOWN SITE PLAN image: draw STAGING MARKERS on it\nat the positions described below. This is a planning aid for a film\ncrew, drawn over the plan.\n\nPLACEMENT AUTHORITY: each marker line below places its marker relative\nto a fixed feature, zone or boundary VISIBLE on the attached plan —\nthat landmark-relative description is the placement authority. The\nnormalized fractions in parentheses are coarse estimates of the same\nspot (x runs left→right 0..1, y runs top→bottom 0..1 of the full\ncanvas) — use them only to confirm the rough region; when in doubt,\nthe landmark description wins. Do NOT relocate a marker onto a\ndifferent salient line or feature than the one its description names.\n\nNEVER move a marker to make the drawing fit. The camera circle and\nevery subject circle go exactly where their descriptions place them,\neven when that leaves the view wedge or an arrow running off the edge\nof the image — in that case just let the wedge or arrow continue to\nthe image border and stop there. Shifting the camera outward, pulling\na subject aside, or shrinking the wedge to keep it inside the canvas\nall break the plan this map exists to record.\n\nDraw EXACTLY these markers, flat and high-contrast, small enough not\nto hide the plan under them:\n- CAMERA: one filled circle at the camera position, labelled CAM, and\n  out of that circle a FIELD-OF-VIEW WEDGE showing what this shot\n  sees — two straight thin lines, one running toward each of the two\n  frame-edge positions given below, and a third thinner line to the\n  look-target position between them. Fill the area between the two\n  edge lines with a light translucent tint so the wedge reads as the\n  camera's view; the plan underneath must stay fully readable through\n  the tint. Every subject listed below stands inside this wedge —\n  if your drawing leaves one outside it, the wedge is drawn wrong,\n  not the subject.\n- SUBJECTS: one filled circle per subject at its position, labelled\n  with its slot code (E1, E2, ...) followed by its short description\n  in parentheses, and one short ARROW leaving that circle toward the\n  spot the subject faces, so the direction each subject is turned is\n  readable at a glance.\n\nChange NOTHING else: keep every zone, landmark, line, label and color\nof the plan exactly as attached — do not redraw, recolor, move or\nremove anything, do not add scenery, shadows or any other symbols or\ntext beyond the markers, wedge, arrows and labels specified above. The\noutput is the SAME plan image with only these staging markers added.\n\nMARKERS TO DRAW:\n- CAM (camera): circle On the outer forest path near its lower end. (coarse estimate x=0.50, y=0.86); its sight line ends at LOOK TARGET: At the forest path disappearance point where the recognizable path meets the deep marker route. (coarse estimate x=0.50, y=0.50); the frame's LEFT edge cuts through Among the stone markers just left of the deep marker route. (coarse estimate x=0.35, y=0.44) and its RIGHT edge through In the pine understory well right of the deep marker route. (coarse estimate x=0.71, y=0.44)\n- E1 (half-risen middle-aged detective): circle At the western edge of the forest path disappearance point. (coarse estimate x=0.43, y=0.53), facing In the pine understory immediately right of the forest path disappearance point. (coarse estimate x=0.62, y=0.49)\n- E2 (watching man in the crowd): circle In the pine understory immediately right of the forest path disappearance point. (coarse estimate x=0.62, y=0.49), facing At the western edge of the forest path disappearance point. (coarse estimate x=0.43, y=0.53)\n- E3 (crowd of onlookers): circle Clustered in the pine understory to the right of the forest path disappearance point. (coarse estimate x=0.67, y=0.48), facing At the inner end of the recognizable forest path. (coarse estimate x=0.47, y=0.53)\n\n[REGENERATE — the previous attempt failed the marker check: 피사체들이 시야각 영역(파란색 색칠된 부분) 밖에 배치되어 있으며, 왼쪽 시야각 경계선이 지정된 묘비 위치를 향하지 않고 너무 왼쪽으로 벗어나 있습니다.. Draw exactly the specified markers and change nothing else on the plan.]",
  "marker_attempts": 2,
  "marker_map_sha256": "88bfb87fa958841dd931b4977effef3eb155e391d4fb64c859851657fc57436d",
  "leakage": {
   "has_marker_leakage": false,
   "details_ko": "다이어그램 기호나 텍스트 등의 배치도 마커 누출 흔적이 없습니다."
  },
  "prompt": "Draw a single CAMERA-VIEW storyboard frame (16:9) of the scene staged\non the attached MARKER-ANNOTATED TOP-DOWN SITE PLAN, seen from the\ncamera position marked on it.\n\nThe attached plan carries STAGING MARKERS drawn over it: a\ncamera-position marker with a field-of-view wedge (where the camera\nstands, which way it looks and how wide it sees) and one labelled\ncircle per subject with an arrow (where each subject stands and which\nway that subject is turned). Jurisdiction of sources — follow all three, they do not\nconflict:\n- The SITE PLAN under the markers is the SOT for what exists on the\n  site and how its zones and landmarks are arranged relative to each\n  other. The markers themselves are staging aids, NOT site content.\n- The markers on the plan are the SOT for XY POSITIONS, horizontal\n  heading and body facing ONLY: where each subject stands on that\n  plan, which way each one is turned, where the camera stands, which\n  way it looks and how wide its wedge opens. Distances, directions\n  and relative depths between subjects come from the marked\n  positions; what the wedge covers is what this frame contains, and\n  subjects outside the wedge are out of frame. The STAGING GEOMETRY\n  text below restates the same staging with coarse normalized\n  estimates — it never overrides the marked plan.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n  VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n  overhead crane, low angle, ...). If it asks for a high or overhead\n  angle, draw that angle — the site plan being top-down does NOT mean\n  the frame must be eye-level. It does NOT govern horizontal\n  placement: which side of the frame a subject occupies and which way\n  that subject is turned come from the marked plan and the staging\n  text, even when that memo also mentions left, right, foreground or\n  background.\n\nSTYLE — MANNEQUIN BLOCKING SKETCH: thin uniform dark outlines on plain\npaper, nothing else. Draw every person as a PLAIN ARTIST'S MANNEQUIN:\na featureless oval head with NO face, NO hair, NO clothing and no\ncostume detail — simple tapered limbs and torso, like a wooden posing\nfigure. Because the head has no face and no hair, the ONLY thing that\nshows which way a figure is turned is the BODY ITSELF: the direction\nthe chest, hips and feet point, and the way the head-oval sits on the\nshoulders. Aim each mannequin's chest and feet at the spot the staging\ntext says that person faces, and let the arms carry the action. A\nmannequin turned to the wrong side is the single worst error in this\ndrawing. Environment is reduced to a few placement lines (ground line,\nkey landmark silhouettes) — just enough to read positions and depth.\n\nSTAGING GEOMETRY — PLANNING DATA ONLY, NEVER RENDER IT: the lines\nbelow restate, marker by marker, the staging already marked on the\nattached plan: each landmark-relative description restates the\nmarker-map authority in words, and the normalized fractions in\nparentheses (x left→right 0..1, y top→bottom 0..1) are coarse audit\nestimates only — neither ever overrides the marked plan.\n\nThose fractions are TOP-DOWN bookkeeping. Never read them as left and\nright of your frame: on a plan seen from above, a subject \"to the\nright\" can sit on either side of a camera that looks any direction.\nEach subject line therefore ends with the SAME staging expressed from\nTHIS camera — which part of the frame the subject occupies, and which\nway that subject faces in screen terms (toward the left or right of\nthe frame, toward or away from the camera or side-on to it in\nprofile, and whom they are looking at). Those screen-space statements\nare what your drawing must match:\na figure said to face the right of the frame is turned to the right in\nyour drawing, never mirrored.\n\nDo NOT draw any of this data in the image — no numbers, no\ncoordinates, no labels, no arrows, no cones, no dots, no diagram\nsymbols of any kind.\n\nWhere the CAMERA DIRECTION text below also describes horizontal\nplacement (which side of the frame someone occupies, foreground left,\nbackground right), treat that as background reading only: for anything\nabout left, right and where a subject sits in this frame, the staging\nabove wins. CAMERA DIRECTION governs the vertical angle, elevation,\ntilt and framing.\n\nFor whether a subject faces the camera or is turned away from it, each\nsubject line above already states it in screen terms and the SHOT TEXT\nis the source it was derived from — draw exactly what the subject line\nsays, and never mirror it.\n- the camera stands On the outer forest path near its lower end. (coarse estimate x=0.50, y=0.86); its sight line ends at LOOK TARGET: At the forest path disappearance point where the recognizable path meets the deep marker route. (coarse estimate x=0.50, y=0.50); the frame's LEFT edge cuts through Among the stone markers just left of the deep marker route. (coarse estimate x=0.35, y=0.44) and its RIGHT edge through In the pine understory well right of the deep marker route. (coarse estimate x=0.71, y=0.44) — everything between those two edges is in frame, anything outside them is out of frame\n- half-risen middle-aged detective stands At the western edge of the forest path disappearance point. (coarse estimate x=0.43, y=0.53), facing In the pine understory immediately right of the forest path disappearance point. (coarse estimate x=0.62, y=0.49) — from this camera the subject is in the LEFT part of the frame and faces toward the RIGHT of the frame and side-on to the camera, neither toward it nor away from it; this subject is turned toward the watching man in the crowd\n- watching man in the crowd stands In the pine understory immediately right of the forest path disappearance point. (coarse estimate x=0.62, y=0.49), facing At the western edge of the forest path disappearance point. (coarse estimate x=0.43, y=0.53) — from this camera the subject is in the RIGHT part of the frame and faces toward the LEFT of the frame and side-on to the camera, neither toward it nor away from it; this subject is turned toward the half-risen middle-aged detective\n- crowd of onlookers stands Clustered in the pine understory to the right of the forest path disappearance point. (coarse estimate x=0.67, y=0.48), facing At the inner end of the recognizable forest path. (coarse estimate x=0.47, y=0.53) — from this camera the subject is in the RIGHT part of the frame and faces toward the LEFT of the frame; this subject is turned toward the half-risen middle-aged detective\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): At standing eye height beside the investigation area, a slightly tightened side-on composition places 조 형사 at left midground, caught halfway upright, and 정인우 at right background between members of the crowd. The police line cuts horizontally between their depth planes as their opposing profiles lock into a single uninterrupted sightline.\n\nTHE LOCATION: 금월도 해송 숲길 및 숲속\n\nSHOT TEXT (authoritative, Korean): 바닥에서 몸을 일으키던 조 형사가 폴리스 라인 너머 인파 속의 정인우와 똑바로 눈이 마주친 채 멈칫하는 순간.\n\nNATURAL PERFORMANCE (default only — every explicit direction above\nwins): when no pose contract or shot-text direction says otherwise,\nsketched figures read as alive in mid-moment — believable weight\nshift, hands engaged, head-oval and chest turned toward the target the\nSHOT TEXT implies; avoid a stiff attention stance (feet together, arms\nhanging straight down) and a head-oval squared flat to the lens. If the\nSHOT TEXT or any POSE contract above stages stillness, death, sleep,\nunconsciousness, restraint, drill or an explicit direct-to-camera look,\ndraw THAT exactly — this clause never overrides it.\n\nCRITICAL — THE SITE PLAN IS A STAGING REFERENCE, NOT CONTENT: do NOT\ndraw the plan itself, and do NOT draw any diagram elements in the\nframe — no circles, dots, wedges, cones, arrows, callouts, codes or\nany other overlay symbols. The output is a clean storyboard drawing\nof the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image."
 },
 "S42sh1": {
  "input_fingerprint": "57aa03ae41f46d97",
  "geometry": {
   "camera": {
    "look_target": {
     "x": 0.58,
     "y": 0.59
    },
    "look_target_anchor_en": "On the open fueling forecourt immediately right of the roadside fuel pump.",
    "origin": {
     "x": 0.5,
     "y": 0.87
    },
    "origin_anchor_en": "On the national road directly opposite the gas-station vehicle approach.",
    "view_left": {
     "x": 0.22,
     "y": 0.43
    },
    "view_left_anchor_en": "On the left side of the fueling forecourt inward from the road-facing edge.",
    "view_right": {
     "x": 0.78,
     "y": 0.43
    },
    "view_right_anchor_en": "On the right side of the fueling forecourt inward from the road-facing edge."
   },
   "entity_placements": [
    {
     "anchor_en": "In the open vehicle standing space just right of the roadside fuel pump.",
     "camera_facing_evidence": "",
     "camera_facing_relation": "unspecified",
     "facing": {
      "x": 0.76,
      "y": 0.6
     },
     "facing_anchor_en": "On the fueling forecourt farther right of the roadside fuel pump.",
     "slot": "E1",
     "subject_en": "parked old-model sedan",
     "x": 0.64,
     "y": 0.6
    }
   ],
   "rationale_ko": "구형 승용차를 주유기 오른쪽의 개방된 차량 공간에 가로 방향으로 배치했다. 카메라는 국도 쪽에서 여러 차 길이 떨어진 위치에 두고 차량과 주유기를 함께 향하게 했으며, 넓은 프레임 안에 두 요소와 주변 주유소 전경이 들어오도록 시야 쐐기를 설정했다."
  },
  "marker_check": {
   "markers_ok": true,
   "details_ko": "정상"
  },
  "marker_prompt": "Edit the attached TOP-DOWN SITE PLAN image: draw STAGING MARKERS on it\nat the positions described below. This is a planning aid for a film\ncrew, drawn over the plan.\n\nPLACEMENT AUTHORITY: each marker line below places its marker relative\nto a fixed feature, zone or boundary VISIBLE on the attached plan —\nthat landmark-relative description is the placement authority. The\nnormalized fractions in parentheses are coarse estimates of the same\nspot (x runs left→right 0..1, y runs top→bottom 0..1 of the full\ncanvas) — use them only to confirm the rough region; when in doubt,\nthe landmark description wins. Do NOT relocate a marker onto a\ndifferent salient line or feature than the one its description names.\n\nNEVER move a marker to make the drawing fit. The camera circle and\nevery subject circle go exactly where their descriptions place them,\neven when that leaves the view wedge or an arrow running off the edge\nof the image — in that case just let the wedge or arrow continue to\nthe image border and stop there. Shifting the camera outward, pulling\na subject aside, or shrinking the wedge to keep it inside the canvas\nall break the plan this map exists to record.\n\nDraw EXACTLY these markers, flat and high-contrast, small enough not\nto hide the plan under them:\n- CAMERA: one filled circle at the camera position, labelled CAM, and\n  out of that circle a FIELD-OF-VIEW WEDGE showing what this shot\n  sees — two straight thin lines, one running toward each of the two\n  frame-edge positions given below, and a third thinner line to the\n  look-target position between them. Fill the area between the two\n  edge lines with a light translucent tint so the wedge reads as the\n  camera's view; the plan underneath must stay fully readable through\n  the tint. Every subject listed below stands inside this wedge —\n  if your drawing leaves one outside it, the wedge is drawn wrong,\n  not the subject.\n- SUBJECTS: one filled circle per subject at its position, labelled\n  with its slot code (E1, E2, ...) followed by its short description\n  in parentheses, and one short ARROW leaving that circle toward the\n  spot the subject faces, so the direction each subject is turned is\n  readable at a glance.\n\nChange NOTHING else: keep every zone, landmark, line, label and color\nof the plan exactly as attached — do not redraw, recolor, move or\nremove anything, do not add scenery, shadows or any other symbols or\ntext beyond the markers, wedge, arrows and labels specified above. The\noutput is the SAME plan image with only these staging markers added.\n\nMARKERS TO DRAW:\n- CAM (camera): circle On the national road directly opposite the gas-station vehicle approach. (coarse estimate x=0.50, y=0.87); its sight line ends at LOOK TARGET: On the open fueling forecourt immediately right of the roadside fuel pump. (coarse estimate x=0.58, y=0.59); the frame's LEFT edge cuts through On the left side of the fueling forecourt inward from the road-facing edge. (coarse estimate x=0.22, y=0.43) and its RIGHT edge through On the right side of the fueling forecourt inward from the road-facing edge. (coarse estimate x=0.78, y=0.43)\n- E1 (parked old-model sedan): circle In the open vehicle standing space just right of the roadside fuel pump. (coarse estimate x=0.64, y=0.60), facing On the fueling forecourt farther right of the roadside fuel pump. (coarse estimate x=0.76, y=0.60)",
  "marker_attempts": 1,
  "marker_map_sha256": "e89a1fa1b3cafcff8816e905537768416552ff6a93574e1852389a296aefadc4",
  "leakage": {
   "has_marker_leakage": false,
   "details_ko": "도면 기호나 텍스트 누출이 없습니다."
  },
  "prompt": "Draw a single CAMERA-VIEW storyboard frame (16:9) of the scene staged\non the attached MARKER-ANNOTATED TOP-DOWN SITE PLAN, seen from the\ncamera position marked on it.\n\nThe attached plan carries STAGING MARKERS drawn over it: a\ncamera-position marker with a field-of-view wedge (where the camera\nstands, which way it looks and how wide it sees) and one labelled\ncircle per subject with an arrow (where each subject stands and which\nway that subject is turned). Jurisdiction of sources — follow all three, they do not\nconflict:\n- The SITE PLAN under the markers is the SOT for what exists on the\n  site and how its zones and landmarks are arranged relative to each\n  other. The markers themselves are staging aids, NOT site content.\n- The markers on the plan are the SOT for XY POSITIONS, horizontal\n  heading and body facing ONLY: where each subject stands on that\n  plan, which way each one is turned, where the camera stands, which\n  way it looks and how wide its wedge opens. Distances, directions\n  and relative depths between subjects come from the marked\n  positions; what the wedge covers is what this frame contains, and\n  subjects outside the wedge are out of frame. The STAGING GEOMETRY\n  text below restates the same staging with coarse normalized\n  estimates — it never overrides the marked plan.\n- The CAMERA DIRECTION text below (when present) is the SOT for the\n  VERTICAL angle, elevation, tilt and framing (eye-level, high angle,\n  overhead crane, low angle, ...). If it asks for a high or overhead\n  angle, draw that angle — the site plan being top-down does NOT mean\n  the frame must be eye-level. It does NOT govern horizontal\n  placement: which side of the frame a subject occupies and which way\n  that subject is turned come from the marked plan and the staging\n  text, even when that memo also mentions left, right, foreground or\n  background.\n\nSTYLE — MANNEQUIN BLOCKING SKETCH: thin uniform dark outlines on plain\npaper, nothing else. Draw every person as a PLAIN ARTIST'S MANNEQUIN:\na featureless oval head with NO face, NO hair, NO clothing and no\ncostume detail — simple tapered limbs and torso, like a wooden posing\nfigure. Because the head has no face and no hair, the ONLY thing that\nshows which way a figure is turned is the BODY ITSELF: the direction\nthe chest, hips and feet point, and the way the head-oval sits on the\nshoulders. Aim each mannequin's chest and feet at the spot the staging\ntext says that person faces, and let the arms carry the action. A\nmannequin turned to the wrong side is the single worst error in this\ndrawing. Environment is reduced to a few placement lines (ground line,\nkey landmark silhouettes) — just enough to read positions and depth.\n\nSTAGING GEOMETRY — PLANNING DATA ONLY, NEVER RENDER IT: the lines\nbelow restate, marker by marker, the staging already marked on the\nattached plan: each landmark-relative description restates the\nmarker-map authority in words, and the normalized fractions in\nparentheses (x left→right 0..1, y top→bottom 0..1) are coarse audit\nestimates only — neither ever overrides the marked plan.\n\nThose fractions are TOP-DOWN bookkeeping. Never read them as left and\nright of your frame: on a plan seen from above, a subject \"to the\nright\" can sit on either side of a camera that looks any direction.\nEach subject line therefore ends with the SAME staging expressed from\nTHIS camera — which part of the frame the subject occupies, and which\nway that subject faces in screen terms (toward the left or right of\nthe frame, toward or away from the camera or side-on to it in\nprofile, and whom they are looking at). Those screen-space statements\nare what your drawing must match:\na figure said to face the right of the frame is turned to the right in\nyour drawing, never mirrored.\n\nDo NOT draw any of this data in the image — no numbers, no\ncoordinates, no labels, no arrows, no cones, no dots, no diagram\nsymbols of any kind.\n\nWhere the CAMERA DIRECTION text below also describes horizontal\nplacement (which side of the frame someone occupies, foreground left,\nbackground right), treat that as background reading only: for anything\nabout left, right and where a subject sits in this frame, the staging\nabove wins. CAMERA DIRECTION governs the vertical angle, elevation,\ntilt and framing.\n\nFor whether a subject faces the camera or is turned away from it, each\nsubject line above already states it in screen terms and the SHOT TEXT\nis the source it was derived from — draw exactly what the subject line\nsays, and never mirror it.\n- the camera stands On the national road directly opposite the gas-station vehicle approach. (coarse estimate x=0.50, y=0.87); its sight line ends at LOOK TARGET: On the open fueling forecourt immediately right of the roadside fuel pump. (coarse estimate x=0.58, y=0.59); the frame's LEFT edge cuts through On the left side of the fueling forecourt inward from the road-facing edge. (coarse estimate x=0.22, y=0.43) and its RIGHT edge through On the right side of the fueling forecourt inward from the road-facing edge. (coarse estimate x=0.78, y=0.43) — everything between those two edges is in frame, anything outside them is out of frame\n- parked old-model sedan stands In the open vehicle standing space just right of the roadside fuel pump. (coarse estimate x=0.64, y=0.60), facing On the fueling forecourt farther right of the roadside fuel pump. (coarse estimate x=0.76, y=0.60) — from this camera the subject is in the RIGHT part of the frame and faces toward the RIGHT of the frame and away from the camera, into the depth of the shot, so the camera sees this subject from behind\n\nCAMERA DIRECTION (vertical angle / elevation / framing SOT): From the roadside side at chest height and several car lengths away, the camera begins a measured dolly toward the old-model Avante parked beside the fuel pump. The vehicle sits across the middle of the wide frame with the pump offset beside it, leaving the roadside station context visible around both.\n\nTHE LOCATION: 국도변 주유소\n\nSHOT TEXT (authoritative, Korean): 밝은 대낮의 국도변 주유소 주유기 옆에 세워진 구형 아반테 차량의 외부 전경.\n\nNATURAL PERFORMANCE (default only — every explicit direction above\nwins): when no pose contract or shot-text direction says otherwise,\nsketched figures read as alive in mid-moment — believable weight\nshift, hands engaged, head-oval and chest turned toward the target the\nSHOT TEXT implies; avoid a stiff attention stance (feet together, arms\nhanging straight down) and a head-oval squared flat to the lens. If the\nSHOT TEXT or any POSE contract above stages stillness, death, sleep,\nunconsciousness, restraint, drill or an explicit direct-to-camera look,\ndraw THAT exactly — this clause never overrides it.\n\nCRITICAL — THE SITE PLAN IS A STAGING REFERENCE, NOT CONTENT: do NOT\ndraw the plan itself, and do NOT draw any diagram elements in the\nframe — no circles, dots, wedges, cones, arrows, callouts, codes or\nany other overlay symbols. The output is a clean storyboard drawing\nof the scene as seen by the camera only.\n\nNo text, no letters, no numbers, no captions anywhere in the image."
 }
}