Update to Picsart
Snapshot Nov 11, 2026 · 00:08 UTC · version 2.5.0
Collection source: downloaded plugin package.
Instructions updated for motion-review
Full description — Now says: move, and before reporting that your own render, character, rig or acceptance video works.
Observed in instructions or declared skills. Runtime behavior has not been tested.
Up to 3 plugins free · no card required.
All fields and original values.
Full description
Nov 7, 2026 · 00:09 UTC
move.
Nov 11, 2026 · 00:08 UTC
move, and before reporting that your own render, character, rig or acceptance video works.
Skill instructions
Nov 7, 2026 · 00:09 UTC
move. ported-from: motion-review 1.0 at 55.75.0
Nov 11, 2026 · 00:08 UTC
move, and before reporting that your own render, character, rig or acceptance video works. ported-from: motion-review 1.0 at 55.128.2 ## Accept your own render before reporting it Apply this before you call any rendered scene, characte...
Compare saved observations
Download comparison JSONSKILL.md text changes
− removed · + added
--- before +++ after @@ -1,10 +1,10 @@ --- name: motion-review -description: Review an existing scene for motion that is confusing, mistimed, distracting, broken at a seam, or unnecessarily expensive. Use when asked to critique or improve a finished animation, audit a multi-beat scene, prioritize motion fixes, or decide whether a static element should move. +description: Review an existing scene for motion that is confusing, mistimed, distracting, broken at a seam, or unnecessarily expensive. Use when asked to critique or improve a finished animation, audit a multi-beat scene, prioritize motion fixes, or decide whether a static element should move, and before reporting that your own render, character, rig or acceptance video works. metadata: category: workflow version: '1.0' - ported-from: motion-review 1.0 at 55.75.0 + ported-from: motion-review 1.0 at 55.128.2 --- # Review motion in an existing scene @@ -104,3 +104,30 @@ Separate **measured**, **visually inspected**, and **not yet verified** claims. List rejected motion additions when the user asked where to animate, so stillness is an explicit design choice. Keep a short ordered set of actionable changes; do not manufacture polish findings to fill a list. + +## Accept your own render before reporting it + +Apply this before you call any rendered scene, character, rig or acceptance video "working". The +author of a render tends to confirm it; look for the ways it fails. + +1. **Name the intended action in one sentence** ("both hands hold the ball at chest height"). Judge + the picture against that sentence, not against your previous draft or a metric. +2. **Look at the whole window.** Make a contact sheet of at least 8 frames spread across the clip, + including interior frames (`picsart_scene_contact_sheet` for a scene, `picsart_media_video_stills` for a video), and + open it yourself. One still or the endpoints hide mid-motion tears, drift and pops. For sync between + two things (mouth and voice, hand and prop), track both per frame (`picsart_media_track_motion`); stills hid a mouth + drift that per-frame tracking caught. +3. **Check what the viewer sees, not what you built.** A metric on the mechanism can pass while the + picture is wrong: a grip socket landed within 0.55 px of the ball while the drawn hand stopped a + hand-width short. Check the drawn result: contact between silhouettes, limbs bending the + anatomical way, nothing passing through a body or prop, near and far parts in the right draw + order, garments following the body without tears or flares. A mean pixel diff cannot see text + or small parts. +4. **List defects first**, each with the frame or time where it shows, then what works. One passing + gate is not success. A first iteration may carry minor labelled defects but no bad mistake: a + broken anatomy, a missed contact or a stated rule broken blocks it. +5. **Deliver the moving result.** Send the labelled final video for motion, not a sheet of + measurements; the sheet is your evidence, the video is the deliverable. + +When the work matters (a change others will build on, a result shared with others), have someone other than the author +review it with fresh eyes, applying this section.
Full technical diff · 2 changed fields
changed /description
"Review an existing scene for motion that is confusing, mistimed, distracting, broken at a seam, or unnecessarily expensive. Use when asked to critique or improve a finished animation, audit a multi-beat scene, prioritize motion fixes, or decide whether a static element should move."
"Review an existing scene for motion that is confusing, mistimed, distracting, broken at a seam, or unnecessarily expensive. Use when asked to critique or improve a finished animation, audit a multi-beat scene, prioritize motion fixes, or decide whether a static element should move, and before reporting that your own render, character, rig or acceptance video works."
changed /skill_md_contents
"---\nname: motion-review\ndescription: Review an existing scene for motion that is confusing, mistimed, distracting, broken at a seam, or unnecessarily expensive. Use when asked to critique or improve a finished animation, audit a multi-beat scene, prioritize motion fixes, or decide whether a static element should move.\nmetadata:\n category: workflow\n version: '1.0'\n ported-from: motion-review 1.0 at 55.75.0\n---\n\n# Review motion in an existing scene\n\nReview the authored scene and the moving picture together. A valid document and a good still frame\ncannot establish that a beat reads correctly in sequence. Find the few changes that would most\nimprove comprehension, then verify each change in the interval where it matters.\n\nThis recipe reviews an existing document. For a new beat, start with motion-intent (fetch with\n`picsart_scene_recipe`, name `motion-intent`). For scene syntax and worked constructions, use\nscene-animation-reference and motion-pattern-library, fetched the same way.\n\n## Establish what the scene is trying to say\n\nRead the brief, storyboard, reference, or editor notes when supplied. Without one, describe the\napparent message as an **inference** and avoid treating your preferred style as the intended one.\n\nMake a short timeline map from the scene itself: composition duration and frame rate, layer\nwindows, track clips and transitions, motion tracks, looks, text animations, and nested\n`scene_ref` content. Use `picsart_scene_inspect` for a large document; it shows authored values\nand addresses, not evaluated root-time positions. Identify each visible beat's leader, the change\nit should communicate, and the interval in which that change must be read. Mark every seam and\nevery long hold for closer inspection.\n\n**Do not add motion simply because a layer is static.** Stillness may support reading, contrast, or\nan intentional reset. A proposed addition needs a communication job and a viewer-visible change it\nwould clarify. Reject it when it only repeats another layer's gesture or competes with the leader.\n\n## Gather evidence at the right scale\n\n1. Run `picsart_scene_validate` on the document under review. Repair an invalid scene before\n judging subtle timing. A successful validation proves the document's contract, not the rendered\n beat.\n2. Use `picsart_scene_query_layout` at the start, interior, landing, and hold of each suspect\n motion. Query both sides of a seam. Map nested or layer-local clocks to root time before\n comparing values. If there is an explicit geometric expectation, use\n `picsart_scene_check_contracts` over its intended window, and `picsart_scene_check_motion` for\n frame-complete motion expectations.\n3. Use `picsart_scene_contact_sheet` with explicit `times` for decisive frames, including an\n interior frame rather than endpoints alone. Inspect pixels for visibility, crop, blur, text\n readability, and overlapping attention.\n4. Watch the **continuous rendered interval** at its intended viewing speed when the issue is\n pacing, acceleration, causality, or a seam. Stills can suggest where to look but cannot settle\n those questions. If only stills are available, mark the temporal judgment unverified.\n\nFor reference matching, compare the reference and candidate over the same window using\n`picsart_media_track_motion_diff`, `picsart_media_flow_diff`, or `picsart_media_inventory_diff` as\nappropriate. These answer different questions: trajectory, flow, or visible inventory. None\njudges whether an original motion choice is expressive.\n\n## Diagnose before changing a value\n\nName the observed failure in scene terms, then identify the smallest responsible span or layer.\n\n| Observation | First thing to inspect | Next decision |\n|---|---|---|\n| The viewer cannot tell what changed | leader, contrast, arrival order, hold | remove competing motion or reveal the needed result more clearly |\n| A move feels abrupt or drifts after landing | interior pose, easing, expression tail, layer clock | correct its progress or settle at the intended root time |\n| A button or pointer action feels disconnected | pointer arrival, response start, result start | align the causal event and inspect the full widget, not one child |\n| The seam feels like a new slide | carrier identity, direction, overlapping windows, transition source frames | choose continuation, replacement, or an intentional reset before choosing a transition |\n| Motion stutters or misses frame budget | active video count, effect area, blur, live layer windows | reduce the heaviest live layers in that window and re-check the target's playback |\n| A static interval seems dead | reading demand, next action, intended pause | shorten the hold or add one purposeful change; a pause may be correct |\n\nThese are hypotheses, not automatic fixes. Check the relevant recipe before changing a\nrepresentation (fetch with `picsart_scene_recipe`): app-ui-animation for a simulated interaction,\nmotion-scene-continuity for a seam, linked-element-motion for reflow. A preset cannot repair a\nwrong semantic relationship or an incorrect layer lifetime.\n\n## Choose the next fix\n\nRank findings by the viewer's likely misunderstanding, how much of the scene they affect, and the\nconfidence of the evidence. Resolve structural failures and missing or incorrect content first,\nthen causality and continuity, then timing and emphasis, then polish. When two changes could\nexplain the same symptom, try the smaller one and re-render its window before adding more.\n\nFor each proposed change, record:\n\n- the layer or scoped target and root-time window;\n- the intended viewer-visible result;\n- the current observation and its evidence source;\n- one concrete edit to test, including removing motion when it has no job;\n- a numeric check where a numeric expectation exists, plus the interval to watch.\n\nAvoid universal timing verdicts. A 0.2-second gesture can be a clear UI tap and an unreadable\ntitle entrance, depending on frame rate, travel, density, and context. Treat timing tables as\nstarting trials. A rendered scene must be judged at its delivery frame rate and viewing size.\n\nAfter an edit, validate again, recheck its entire beat, and inspect the neighboring seams. A\nsuccessful midpoint frame does not prove the entrance, landing, or exit. When the deliverable has\nmultiple output targets, verify the affected interval on each; report a target's translation\ndiagnostic instead of assuming equivalent output.\n\n## Report a review\n\nLead with the highest-impact confirmed issue, or say that none was found. For each finding give a\nscene location, root-time interval, what the viewer sees, the evidence, and the proposed edit.\nSeparate **measured**, **visually inspected**, and **not yet verified** claims. List rejected\nmotion additions when the user asked where to animate, so stillness is an explicit design choice.\nKeep a short ordered set of actionable changes; do not manufacture polish findings to fill a list.\n"
"---\nname: motion-review\ndescription: Review an existing scene for motion that is confusing, mistimed, distracting, broken at a seam, or unnecessarily expensive. Use when asked to critique or improve a finished animation, audit a multi-beat scene, prioritize motion fixes, or decide whether a static element should move, and before reporting that your own render, character, rig or acceptance video works.\nmetadata:\n category: workflow\n version: '1.0'\n ported-from: motion-review 1.0 at 55.128.2\n---\n\n# Review motion in an existing scene\n\nReview the authored scene and the moving picture together. A valid document and a good still frame\ncannot establish that a beat reads correctly in sequence. Find the few changes that would most\nimprove comprehension, then verify each change in the interval where it matters.\n\nThis recipe reviews an existing document. For a new beat, start with motion-intent (fetch with\n`picsart_scene_recipe`, name `motion-intent`). For scene syntax and worked constructions, use\nscene-animation-reference and motion-pattern-library, fetched the same way.\n\n## Establish what the scene is trying to say\n\nRead the brief, storyboard, reference, or editor notes when supplied. Without one, describe the\napparent message as an **inference** and avoid treating your preferred style as the intended one.\n\nMake a short timeline map from the scene itself: composition duration and frame rate, layer\nwindows, track clips and transitions, motion tracks, looks, text animations, and nested\n`scene_ref` content. Use `picsart_scene_inspect` for a large document; it shows authored values\nand addresses, not evaluated root-time positions. Identify each visible beat's leader, the change\nit should communicate, and the interval in which that change must be read. Mark every seam and\nevery long hold for closer inspection.\n\n**Do not add motion simply because a layer is static.** Stillness may support reading, contrast, or\nan intentional reset. A proposed addition needs a communication job and a viewer-visible change it\nwould clarify. Reject it when it only repeats another layer's gesture or competes with the leader.\n\n## Gather evidence at the right scale\n\n1. Run `picsart_scene_validate` on the document under review. Repair an invalid scene before\n judging subtle timing. A successful validation proves the document's contract, not the rendered\n beat.\n2. Use `picsart_scene_query_layout` at the start, interior, landing, and hold of each suspect\n motion. Query both sides of a seam. Map nested or layer-local clocks to root time before\n comparing values. If there is an explicit geometric expectation, use\n `picsart_scene_check_contracts` over its intended window, and `picsart_scene_check_motion` for\n frame-complete motion expectations.\n3. Use `picsart_scene_contact_sheet` with explicit `times` for decisive frames, including an\n interior frame rather than endpoints alone. Inspect pixels for visibility, crop, blur, text\n readability, and overlapping attention.\n4. Watch the **continuous rendered interval** at its intended viewing speed when the issue is\n pacing, acceleration, causality, or a seam. Stills can suggest where to look but cannot settle\n those questions. If only stills are available, mark the temporal judgment unverified.\n\nFor reference matching, compare the reference and candidate over the same window using\n`picsart_media_track_motion_diff`, `picsart_media_flow_diff`, or `picsart_media_inventory_diff` as\nappropriate. These answer different questions: trajectory, flow, or visible inventory. None\njudges whether an original motion choice is expressive.\n\n## Diagnose before changing a value\n\nName the observed failure in scene terms, then identify the smallest responsible span or layer.\n\n| Observation | First thing to inspect | Next decision |\n|---|---|---|\n| The viewer cannot tell what changed | leader, contrast, arrival order, hold | remove competing motion or reveal the needed result more clearly |\n| A move feels abrupt or drifts after landing | interior pose, easing, expression tail, layer clock | correct its progress or settle at the intended root time |\n| A button or pointer action feels disconnected | pointer arrival, response start, result start | align the causal event and inspect the full widget, not one child |\n| The seam feels like a new slide | carrier identity, direction, overlapping windows, transition source frames | choose continuation, replacement, or an intentional reset before choosing a transition |\n| Motion stutters or misses frame budget | active video count, effect area, blur, live layer windows | reduce the heaviest live layers in that window and re-check the target's playback |\n| A static interval seems dead | reading demand, next action, intended pause | shorten the hold or add one purposeful change; a pause may be correct |\n\nThese are hypotheses, not automatic fixes. Check the relevant recipe before changing a\nrepresentation (fetch with `picsart_scene_recipe`): app-ui-animation for a simulated interaction,\nmotion-scene-continuity for a seam, linked-element-motion for reflow. A preset cannot repair a\nwrong semantic relationship or an incorrect layer lifetime.\n\n## Choose the next fix\n\nRank findings by the viewer's likely misunderstanding, how much of the scene they affect, and the\nconfidence of the evidence. Resolve structural failures and missing or incorrect content first,\nthen causality and continuity, then timing and emphasis, then polish. When two changes could\nexplain the same symptom, try the smaller one and re-render its window before adding more.\n\nFor each proposed change, record:\n\n- the layer or scoped target and root-time window;\n- the intended viewer-visible result;\n- the current observation and its evidence source;\n- one concrete edit to test, including removing motion when it has no job;\n- a numeric check where a numeric expectation exists, plus the interval to watch.\n\nAvoid universal timing verdicts. A 0.2-second gesture can be a clear UI tap and an unreadable\ntitle entrance, depending on frame rate, travel, density, and context. Treat timing tables as\nstarting trials. A rendered scene must be judged at its delivery frame rate and viewing size.\n\nAfter an edit, validate again, recheck its entire beat, and inspect the neighboring seams. A\nsuccessful midpoint frame does not prove the entrance, landing, or exit. When the deliverable has\nmultiple output targets, verify the affected interval on each; report a target's translation\ndiagnostic instead of assuming equivalent output.\n\n## Report a review\n\nLead with the highest-impact confirmed issue, or say that none was found. For each finding give a\nscene location, root-time interval, what the viewer sees, the evidence, and the proposed edit.\nSeparate **measured**, **visually inspected**, and **not yet verified** claims. List rejected\nmotion additions when the user asked where to animate, so stillness is an explicit design choice.\nKeep a short ordered set of actionable changes; do not manufacture polish findings to fill a list.\n\n## Accept your own render before reporting it\n\nApply this before you call any rendered scene, character, rig or acceptance video \"working\". The\nauthor of a render tends to confirm it; look for the ways it fails.\n\n1. **Name the intended action in one sentence** (\"both hands hold the ball at chest height\"). Judge\n the picture against that sentence, not against your previous draft or a metric.\n2. **Look at the whole window.** Make a contact sheet of at least 8 frames spread across the clip,\n including interior frames (`picsart_scene_contact_sheet` for a scene, `picsart_media_video_stills` for a video), and\n open it yourself. One still or the endpoints hide mid-motion tears, drift and pops. For sync between\n two things (mouth and voice, hand and prop), track both per frame (`picsart_media_track_motion`); stills hid a mouth\n drift that per-frame tracking caught.\n3. **Check what the viewer sees, not what you built.** A metric on the mechanism can pass while the\n picture is wrong: a grip socket landed within 0.55 px of the ball while the drawn hand stopped a\n hand-width short. Check the drawn result: contact between silhouettes, limbs bending the\n anatomical way, nothing passing through a body or prop, near and far parts in the right draw\n order, garments following the body without tears or flares. A mean pixel diff cannot see text\n or small parts.\n4. **List defects first**, each with the frame or time where it shows, then what works. One passing\n gate is not success. A first iteration may carry minor labelled defects but no bad mistake: a\n broken anatomy, a missed contact or a stated rule broken blocks it.\n5. **Deliver the moving result.** Send the labelled final video for motion, not a sheet of\n measurements; the sheet is your evidence, the video is the deliverable.\n\nWhen the work matters (a change others will build on, a result shared with others), have someone other than the author\nreview it with fresh eyes, applying this section.\n"
Full snapshot data
{
"description": "Review an existing scene for motion that is confusing, mistimed, distracting, broken at a seam, or unnecessarily expensive. Use when asked to critique or improve a finished animation, audit a multi-beat scene, prioritize motion fixes, or decide whether a static element should move, and before reporting that your own render, character, rig or acceptance video works.",
"included_files": [],
"name": "motion-review",
"skill_md_contents": "---\nname: motion-review\ndescription: Review an existing scene for motion that is confusing, mistimed, distracting, broken at a seam, or unnecessarily expensive. Use when asked to critique or improve a finished animation, audit a multi-beat scene, prioritize motion fixes, or decide whether a static element should move, and before reporting that your own render, character, rig or acceptance video works.\nmetadata:\n category: workflow\n version: '1.0'\n ported-from: motion-review 1.0 at 55.128.2\n---\n\n# Review motion in an existing scene\n\nReview the authored scene and the moving picture together. A valid document and a good still frame\ncannot establish that a beat reads correctly in sequence. Find the few changes that would most\nimprove comprehension, then verify each change in the interval where it matters.\n\nThis recipe reviews an existing document. For a new beat, start with motion-intent (fetch with\n`picsart_scene_recipe`, name `motion-intent`). For scene syntax and worked constructions, use\nscene-animation-reference and motion-pattern-library, fetched the same way.\n\n## Establish what the scene is trying to say\n\nRead the brief, storyboard, reference, or editor notes when supplied. Without one, describe the\napparent message as an **inference** and avoid treating your preferred style as the intended one.\n\nMake a short timeline map from the scene itself: composition duration and frame rate, layer\nwindows, track clips and transitions, motion tracks, looks, text animations, and nested\n`scene_ref` content. Use `picsart_scene_inspect` for a large document; it shows authored values\nand addresses, not evaluated root-time positions. Identify each visible beat's leader, the change\nit should communicate, and the interval in which that change must be read. Mark every seam and\nevery long hold for closer inspection.\n\n**Do not add motion simply because a layer is static.** Stillness may support reading, contrast, or\nan intentional reset. A proposed addition needs a communication job and a viewer-visible change it\nwould clarify. Reject it when it only repeats another layer's gesture or competes with the leader.\n\n## Gather evidence at the right scale\n\n1. Run `picsart_scene_validate` on the document under review. Repair an invalid scene before\n judging subtle timing. A successful validation proves the document's contract, not the rendered\n beat.\n2. Use `picsart_scene_query_layout` at the start, interior, landing, and hold of each suspect\n motion. Query both sides of a seam. Map nested or layer-local clocks to root time before\n comparing values. If there is an explicit geometric expectation, use\n `picsart_scene_check_contracts` over its intended window, and `picsart_scene_check_motion` for\n frame-complete motion expectations.\n3. Use `picsart_scene_contact_sheet` with explicit `times` for decisive frames, including an\n interior frame rather than endpoints alone. Inspect pixels for visibility, crop, blur, text\n readability, and overlapping attention.\n4. Watch the **continuous rendered interval** at its intended viewing speed when the issue is\n pacing, acceleration, causality, or a seam. Stills can suggest where to look but cannot settle\n those questions. If only stills are available, mark the temporal judgment unverified.\n\nFor reference matching, compare the reference and candidate over the same window using\n`picsart_media_track_motion_diff`, `picsart_media_flow_diff`, or `picsart_media_inventory_diff` as\nappropriate. These answer different questions: trajectory, flow, or visible inventory. None\njudges whether an original motion choice is expressive.\n\n## Diagnose before changing a value\n\nName the observed failure in scene terms, then identify the smallest responsible span or layer.\n\n| Observation | First thing to inspect | Next decision |\n|---|---|---|\n| The viewer cannot tell what changed | leader, contrast, arrival order, hold | remove competing motion or reveal the needed result more clearly |\n| A move feels abrupt or drifts after landing | interior pose, easing, expression tail, layer clock | correct its progress or settle at the intended root time |\n| A button or pointer action feels disconnected | pointer arrival, response start, result start | align the causal event and inspect the full widget, not one child |\n| The seam feels like a new slide | carrier identity, direction, overlapping windows, transition source frames | choose continuation, replacement, or an intentional reset before choosing a transition |\n| Motion stutters or misses frame budget | active video count, effect area, blur, live layer windows | reduce the heaviest live layers in that window and re-check the target's playback |\n| A static interval seems dead | reading demand, next action, intended pause | shorten the hold or add one purposeful change; a pause may be correct |\n\nThese are hypotheses, not automatic fixes. Check the relevant recipe before changing a\nrepresentation (fetch with `picsart_scene_recipe`): app-ui-animation for a simulated interaction,\nmotion-scene-continuity for a seam, linked-element-motion for reflow. A preset cannot repair a\nwrong semantic relationship or an incorrect layer lifetime.\n\n## Choose the next fix\n\nRank findings by the viewer's likely misunderstanding, how much of the scene they affect, and the\nconfidence of the evidence. Resolve structural failures and missing or incorrect content first,\nthen causality and continuity, then timing and emphasis, then polish. When two changes could\nexplain the same symptom, try the smaller one and re-render its window before adding more.\n\nFor each proposed change, record:\n\n- the layer or scoped target and root-time window;\n- the intended viewer-visible result;\n- the current observation and its evidence source;\n- one concrete edit to test, including removing motion when it has no job;\n- a numeric check where a numeric expectation exists, plus the interval to watch.\n\nAvoid universal timing verdicts. A 0.2-second gesture can be a clear UI tap and an unreadable\ntitle entrance, depending on frame rate, travel, density, and context. Treat timing tables as\nstarting trials. A rendered scene must be judged at its delivery frame rate and viewing size.\n\nAfter an edit, validate again, recheck its entire beat, and inspect the neighboring seams. A\nsuccessful midpoint frame does not prove the entrance, landing, or exit. When the deliverable has\nmultiple output targets, verify the affected interval on each; report a target's translation\ndiagnostic instead of assuming equivalent output.\n\n## Report a review\n\nLead with the highest-impact confirmed issue, or say that none was found. For each finding give a\nscene location, root-time interval, what the viewer sees, the evidence, and the proposed edit.\nSeparate **measured**, **visually inspected**, and **not yet verified** claims. List rejected\nmotion additions when the user asked where to animate, so stillness is an explicit design choice.\nKeep a short ordered set of actionable changes; do not manufacture polish findings to fill a list.\n\n## Accept your own render before reporting it\n\nApply this before you call any rendered scene, character, rig or acceptance video \"working\". The\nauthor of a render tends to confirm it; look for the ways it fails.\n\n1. **Name the intended action in one sentence** (\"both hands hold the ball at chest height\"). Judge\n the picture against that sentence, not against your previous draft or a metric.\n2. **Look at the whole window.** Make a contact sheet of at least 8 frames spread across the clip,\n including interior frames (`picsart_scene_contact_sheet` for a scene, `picsart_media_video_stills` for a video), and\n open it yourself. One still or the endpoints hide mid-motion tears, drift and pops. For sync between\n two things (mouth and voice, hand and prop), track both per frame (`picsart_media_track_motion`); stills hid a mouth\n drift that per-frame tracking caught.\n3. **Check what the viewer sees, not what you built.** A metric on the mechanism can pass while the\n picture is wrong: a grip socket landed within 0.55 px of the ball while the drawn hand stopped a\n hand-width short. Check the drawn result: contact between silhouettes, limbs bending the\n anatomical way, nothing passing through a body or prop, near and far parts in the right draw\n order, garments following the body without tears or flares. A mean pixel diff cannot see text\n or small parts.\n4. **List defects first**, each with the frame or time where it shows, then what works. One passing\n gate is not success. A first iteration may carry minor labelled defects but no bad mistake: a\n broken anatomy, a missed contact or a stated rule broken blocks it.\n5. **Deliver the moving result.** Send the labelled final video for motion, not a sheet of\n measurements; the sheet is your evidence, the video is the deliverable.\n\nWhen the work matters (a change others will build on, a result shared with others), have someone other than the author\nreview it with fresh eyes, applying this section.\n"
}SHA-256 of public snapshot: 43948b64757888503d8a92646164f8ae0283f44e179323665f76706348bc7437