[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6BY0Z94MAC4HN67NYG9ADC7":3,"skill-same:01M6BY0Z94MAC4HN67NYG9ADC7":68},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":40,"linkOnly":41,"markdownContent":42,"markdownBody":43,"frontMatter":44,"refName":46,"commitSha":47,"skillDirectory":48,"forksCount":49,"repositoryHtmlUrl":50,"repositoryHomepageUrl":51,"ownerHtmlUrl":52,"ownerBlogUrl":12,"ownerDescription":53,"updatedAt":54,"install":55,"resources":67},"01M6BY0Z94MAC4HN67NYG9ADC7","motion-graphics",null,"Build and review a NodeTool timeline piece — the motion agent's entry point. Covers the build loop, choosing between the code pack and edit ops, the revise-by-editing-the-script loop, and the review pass. Use for title cards, kinetic text, lower thirds, clip entrances and exits, and layered motion.","nodetool-ai\u002Fnodetool","nodetool-ai","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F164873569?v=4","01M6BXPFDW36K9K1MR8ZH63HG2","nodetool.ai","01M6BXPFJ893XM219GQN60B0WK",635,"AGPL-3.0","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fmotion-graphics\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39],"graphics","motion graphics","motion","point","review","title","layered","for","true","revise","build","clip","use","ops","edit","editing","nodetool","text","loop","timeline","code","2026-11-10 10:12:30",false,"---\nname: motion-graphics\ndescription: Build and review a NodeTool timeline piece — the motion agent's entry point. Covers the build loop, choosing between the code pack and edit ops, the revise-by-editing-the-script loop, and the review pass. Use for title cards, kinetic text, lower thirds, clip entrances and exits, and layered motion.\nfeatured: true\n---\n\n# Motion Graphics → Timeline Agent\n\n## Before the first edit of a new piece\n\nWhen starting a piece from a brief, load `motion-direction` and\n`motion-principles` with `load_skill` before the first timeline edit. Use them\nto set the motion language and timing for the whole piece. Do not start\nanimating from this file alone.\n\nIf the piece has a title layout or an end card, also load `frame-composition`\nbefore that first edit. If it has an end card, load `logo-reveal` too. These\nloads are required even when this skill was the only one selected from the\ncatalog.\n\nFor showcase, hero, launch, or \"best\" briefs, follow `motion-direction`'s\ndepth plan and example timelines before writing scenes.\n\nMotion is built, then checked. Author with the pack or with ops (below), then\nlook at the frames with `preview_timeline_frame`. A change you have not looked\nat is not done.\n\n## Load other craft skills for the job\n\nThis file is the build loop and the entry point, not the op contract. After\nthe required loads above, load other relevant craft skills for the job and\ncome back here to pick the authoring surface.\n\n| The question | Skill |\n|---|---|\n| Where do the cuts land, how do I pace it, how do I ramp a hit | `beat-sync-editing` |\n| Which colours, in what order do I grade, what can actually move | `color-motion` |\n| A preset is close but not right; overshoot, decay, wiggle, arc, draw-on | `motion-curves` |\n| An ambient bed behind the content | `motion-background` |\n\nThree are neighbours rather than layers. `caption-titles` decides what text\nsays and when it appears. `video-audio-continuity` decides, before the first\nclip renders, whether a multi-scene piece's sound comes from one generated clip\nor from a track of your own — a decision this file's ops cannot undo. The\nboard skills (`explainer-storyboard`, `commercial-beat-sheet`,\n`launch-commercial`, `music-video-treatment`, `trailer-template`) decide the\nshots before there is a timeline to animate; `assemble_storyboard_timeline` is\nwhere their board becomes this file's document.\n\nAnything generated for the timeline — a bed, a sting, a voice, a clip — has a\nmodel line with its own guide. `find_model` returns `prompting_skill` on a\nroute it covers; `load_skill` it before writing that prompt.\n\n## The build loop\n\nBuild a piece in four moves: **direct** (fix the motion language and the\nscene list before writing anything — `motion-direction` and, when there is a\ntitle or end card, `frame-composition`\u002F`logo-reveal`), **build** (write the\nscenes in code, one `v.scene` per beat), **save** (`v.save()` attaches the code; read\n`saved.code` and `saved.warnings`), **review** (the\npass below). Revise by editing that script — see below — and repeat the\nlast two steps. A piece that renders clean on the first save has not been\nlooked at yet.\n\n## Choose the authoring surface\n\nTwo ways to touch a timeline document. Pick by what changed, not by habit.\n\n| Situation | Surface |\n|---|---|\n| A new cut, or a scene with more than a few clips | `@nodetool-ai\u002Fsandbox-timeline` (code) |\n| A few edits to a cut that already exists — retiming a clip, swapping a preset, nudging a stagger | `edit_timeline` ops |\n\n**Code, for a new cut.** In a JS script or a Code node, `import { video } from\n\"@nodetool-ai\u002Fsandbox-timeline\"`. Its `video()` builds a document of scenes\n(`v.scene(name, seconds, fn)`, local time in seconds), lays them out with\n`v.series([...])`, and `v.save(nodetool.timelines, {...})` validates, creates\nthe timeline, writes the whole document and validates again. It throws on any\nvalidation error and returns `{timeline_id, warnings}` plus `code`. A cut\nwith hundreds of keyframed clips costs a handful of API calls this way,\ninstead of one `edit_timeline` op per clip. NodeTool's shipped example\ntimelines (Cadence, Kite, Prism, Serein, Voltra, Tidewater, T minus 30) are built with it. Its\nfull signatures, options and gotchas are the pack's own skill — read\n`nodetool.packs.docs(\"@nodetool-ai\u002Fsandbox-timeline\")` before writing scenes,\nnot this file. `frame-composition` covers how to lay out what goes inside a\nscene; this file covers the build loop and the ops surface below.\n\n**Ops, for an edit.** `edit_timeline` takes a timeline ID and an `ops` array,\nrun in order, up to 60 per call:\n\n```json\n{\"timeline_id\":\"\u003Cid>\",\"ops\":[{\"op\":\"animate_clip\",\"target\":\"Title\",\"animations\":[{\"role\":\"in\",\"preset\":\"fade\",\"durationMs\":400}]}]}\n```\n\nIts full op shapes, what each op refuses, and when a change instead needs a\nwhole-document `set_timeline_document` write are `timeline-edit-ops` — load it\nbefore a batch of ops, not this file.\n\n## A compact two-scene piece\n\nThe build loop end to end: a backdrop, a laid-out title block, a counter, a\ntransition, a finish pass, save, then read the warnings.\n\n```js\nimport { video } from \"@nodetool-ai\u002Fsandbox-timeline\";\n\nconst v = video({\n  width: 1920, height: 1080, fps: 30,\n  palette: { ink: \"#06110e\", ink2: \"#0c2a21\", text: \"#f4fbf8\", accent: \"#34d399\" },\n  fonts: { display: \"Space Grotesk\", body: \"Inter\" }\n});\n\nconst hello = v.scene(\"Hello\", 2.5, (s) => {\n  s.backdrop();\n  const kicker = s.kicker(\"LIVE\");\n  const title = s.text(\"Round-ups\", { size: 120, weight: 700, font: \"display\" });\n  const sub = s.text(\"on every card\", { size: 40, color: \"#8fb3a6\" });\n  s.stack([kicker, title, sub], { gap: 20 }); \u002F\u002F frame-composition: a title block\n  title.enter({ from: { scale: 1.2, opacity: 0 }, dur: 0.3, ease: \"outExpo\" });\n  sub.enter({ from: { offsetY: 24, opacity: 0 }, at: 0.15, dur: 0.25 });\n});\n\nconst outro = v.scene(\"Outro\", 1.8, (s) => {\n  s.backdrop({ colors: [\"#0f3b2e\", \"#06110e\"] });\n  const counter = s.text(\"$0.00\", { size: 96, font: \"display\" });\n  counter.count({ from: 0, to: 1.6, dur: 1.2, prefix: \"$\", decimals: 2 });\n  s.finish({ vignette: 0.2, grain: 0.05 });\n});\n\nv.series([hello, v.transition(\"wipe\", 0.3, { direction: \"left\" }), outro]);\n\nconst saved = await v.save(nodetool.timelines, { name: \"Round-ups\", showcase: true });\nif (saved.warnings.length) {\n  \u002F\u002F read them before calling this done — see Review pass below\n}\n```\n\n## Author and revise with code\n\n`v.save(nodetool.timelines, {...})` attaches the code that rebuilds the\ntimeline as its source — no separate `code.set` call. The stored code is the\nretained program: the scene callbacks and the helpers they use, with every\nresearch result, generated asset id and loop value written in as a literal.\nThe calls that produced those values ran once, in your action, and a rebake\nnever repeats them. `saved.code.embedded` is false, with the reason in\n`saved.code.warnings`, when a scene callback calls `nodetool.*`, uses\n`Math.random()`\u002F`Date.now()`, or reads a value that is not plain data. Do the\nresearch and generation at the top level, then build the scenes from the\nresults. To attach code after the fact,\n`nodetool.timelines.code.set(timeline_id, code)` bakes the code, which must\nbuild the timeline without a capability call.\n\nRevising a code-backed timeline is: `code.get(id)` → `code.edit(id, edits)`\nwith a few small exact string replacements (each `old` matches exactly once)\n→ read the `conflicts` and `warnings` it answers → look at frames. This is\nthe review → revise → rerun loop from the build loop above, made cheap — a\ntwo-line diff instead of a 500-line resend. `edit_timeline` ops still apply\nfor a timeline with no code, or a deliberate hand tweak — `timeline-edit-ops`\nowns that contract and what it costs a code-backed scene. The pack's own\nskill (`@nodetool-ai\u002Fsandbox-timeline`) covers what code-backed identity\nmeans for scene names and clip ids; `api-timelines` owns the full `code.*`\ncall contract.\n\n## Review pass\n\nFor a showcase, hero, launch, or \"best\" piece, critique like a director before\ncalling it done. A render with no errors is not a finished piece — this pass is\npart of the job, not a step to skip when the frames look clean. Reserve rounds\nfor it: the run budget is shared with authoring, so budget for at least one\nfull pass plus one revision, not just the build.\n\n1. **Contact sheet.** `preview_timeline_frame` with `sheet: true`, sampling\n   every scene at its entrance, its hero hold, and its exit.\n2. **Per-scene checklist**, answered in writing before you touch an op:\n   - What does the eye follow first, and is that the intended hero?\n   - What moves in the background? A static bed behind the action is a\n     finding, not a pass.\n   - Does any text touch or get crossed by a shape?\n   - How much of the frame is actually used? Look yourself — the validator's\n     `showcase_frame_underfilled` check is a bounding-box proxy and can miss a\n     thin outline or ring that reads as sparse against an empty backdrop.\n   - Does every number that is supposed to count actually count, and finish\n     while it is still on screen?\n   - How does the scene exit and the next one enter? A default crossfade on a\n     showcase brief is a finding to name, not the absence of one.\n3. **Compare against a reference.** Pick the closest scene with\n   `examples.list()` \u002F `examples.get(slug)`. Render matching-beat sheets of\n   your piece and that example with `preview_timeline_frame`, `sheet: true`,\n   mapping beats across differing durations and sampling the middle of a\n   move. Call `view_image` on both with this explicit question: \"Compared\n   with the example, what is missing in layer count, secondary motion,\n   texture and light, and camera depth? Name the beat and the gap.\" Read how\n   the example scene was actually built with `examples.source(slug)` when the\n   clip excerpt alone does not explain a moment — a repeater's phase offsets,\n   a two-pass text effect, a shared id scheme. `compare_timeline_frames`\n   cannot target a shipped example — both its sides must be your own timeline\n   (an id, version or document) — so use it after a revision to prove which\n   frames actually changed, not to judge against a reference.\n4. **Validate at the showcase tier.** `validate_timeline` with\n   `{\"timeline_id\": \"\u003Cid>\", \"tier\": \"showcase\"}` (in code,\n   `nodetool.timelines.validate(id, {tier: \"showcase\"})`) returns `errors`\n   and `warnings` — there is no `issues` field.\n   Every `showcase_*` warning is a hard stop: resolve it, or state the\n   specific artistic exclusion to the user before rendering. Expect\n   `showcase_scene_groups_missing`, `showcase_custom_motion_missing`,\n   `showcase_finish_missing`, `showcase_camera_missing`,\n   `showcase_layer_density_low`, `showcase_scene_group_density_low`,\n   `showcase_custom_animation_density_low`, `showcase_effect_variety_low`,\n   `showcase_clip_density_low`, `showcase_motion_blur_missing`, plus\n   `showcase_frame_underfilled`, `showcase_text_collision`,\n   `showcase_motion_range_narrow` and `showcase_dead_motion`. Each names a\n   shipped example to study. Zero warnings is a structural floor, not a\n   craft verdict — the medians in the warning text are the design target.\n5. **Revise, then re-run this pass.** Make at least one revision that changes\n   the composition itself, not only a fix for a validator error, before\n   calling the piece done. Then repeat steps 1–4 on the result. Stopping at\n   the first clean render is the failure this pass exists to catch.\n","# Motion Graphics → Timeline Agent\n\n## Before the first edit of a new piece\n\nWhen starting a piece from a brief, load `motion-direction` and\n`motion-principles` with `load_skill` before the first timeline edit. Use them\nto set the motion language and timing for the whole piece. Do not start\nanimating from this file alone.\n\nIf the piece has a title layout or an end card, also load `frame-composition`\nbefore that first edit. If it has an end card, load `logo-reveal` too. These\nloads are required even when this skill was the only one selected from the\ncatalog.\n\nFor showcase, hero, launch, or \"best\" briefs, follow `motion-direction`'s\ndepth plan and example timelines before writing scenes.\n\nMotion is built, then checked. Author with the pack or with ops (below), then\nlook at the frames with `preview_timeline_frame`. A change you have not looked\nat is not done.\n\n## Load other craft skills for the job\n\nThis file is the build loop and the entry point, not the op contract. After\nthe required loads above, load other relevant craft skills for the job and\ncome back here to pick the authoring surface.\n\n| The question | Skill |\n|---|---|\n| Where do the cuts land, how do I pace it, how do I ramp a hit | `beat-sync-editing` |\n| Which colours, in what order do I grade, what can actually move | `color-motion` |\n| A preset is close but not right; overshoot, decay, wiggle, arc, draw-on | `motion-curves` |\n| An ambient bed behind the content | `motion-background` |\n\nThree are neighbours rather than layers. `caption-titles` decides what text\nsays and when it appears. `video-audio-continuity` decides, before the first\nclip renders, whether a multi-scene piece's sound comes from one generated clip\nor from a track of your own — a decision this file's ops cannot undo. The\nboard skills (`explainer-storyboard`, `commercial-beat-sheet`,\n`launch-commercial`, `music-video-treatment`, `trailer-template`) decide the\nshots before there is a timeline to animate; `assemble_storyboard_timeline` is\nwhere their board becomes this file's document.\n\nAnything generated for the timeline — a bed, a sting, a voice, a clip — has a\nmodel line with its own guide. `find_model` returns `prompting_skill` on a\nroute it covers; `load_skill` it before writing that prompt.\n\n## The build loop\n\nBuild a piece in four moves: **direct** (fix the motion language and the\nscene list before writing anything — `motion-direction` and, when there is a\ntitle or end card, `frame-composition`\u002F`logo-reveal`), **build** (write the\nscenes in code, one `v.scene` per beat), **save** (`v.save()` attaches the code; read\n`saved.code` and `saved.warnings`), **review** (the\npass below). Revise by editing that script — see below — and repeat the\nlast two steps. A piece that renders clean on the first save has not been\nlooked at yet.\n\n## Choose the authoring surface\n\nTwo ways to touch a timeline document. Pick by what changed, not by habit.\n\n| Situation | Surface |\n|---|---|\n| A new cut, or a scene with more than a few clips | `@nodetool-ai\u002Fsandbox-timeline` (code) |\n| A few edits to a cut that already exists — retiming a clip, swapping a preset, nudging a stagger | `edit_timeline` ops |\n\n**Code, for a new cut.** In a JS script or a Code node, `import { video } from\n\"@nodetool-ai\u002Fsandbox-timeline\"`. Its `video()` builds a document of scenes\n(`v.scene(name, seconds, fn)`, local time in seconds), lays them out with\n`v.series([...])`, and `v.save(nodetool.timelines, {...})` validates, creates\nthe timeline, writes the whole document and validates again. It throws on any\nvalidation error and returns `{timeline_id, warnings}` plus `code`. A cut\nwith hundreds of keyframed clips costs a handful of API calls this way,\ninstead of one `edit_timeline` op per clip. NodeTool's shipped example\ntimelines (Cadence, Kite, Prism, Serein, Voltra, Tidewater, T minus 30) are built with it. Its\nfull signatures, options and gotchas are the pack's own skill — read\n`nodetool.packs.docs(\"@nodetool-ai\u002Fsandbox-timeline\")` before writing scenes,\nnot this file. `frame-composition` covers how to lay out what goes inside a\nscene; this file covers the build loop and the ops surface below.\n\n**Ops, for an edit.** `edit_timeline` takes a timeline ID and an `ops` array,\nrun in order, up to 60 per call:\n\n```json\n{\"timeline_id\":\"\u003Cid>\",\"ops\":[{\"op\":\"animate_clip\",\"target\":\"Title\",\"animations\":[{\"role\":\"in\",\"preset\":\"fade\",\"durationMs\":400}]}]}\n```\n\nIts full op shapes, what each op refuses, and when a change instead needs a\nwhole-document `set_timeline_document` write are `timeline-edit-ops` — load it\nbefore a batch of ops, not this file.\n\n## A compact two-scene piece\n\nThe build loop end to end: a backdrop, a laid-out title block, a counter, a\ntransition, a finish pass, save, then read the warnings.\n\n```js\nimport { video } from \"@nodetool-ai\u002Fsandbox-timeline\";\n\nconst v = video({\n  width: 1920, height: 1080, fps: 30,\n  palette: { ink: \"#06110e\", ink2: \"#0c2a21\", text: \"#f4fbf8\", accent: \"#34d399\" },\n  fonts: { display: \"Space Grotesk\", body: \"Inter\" }\n});\n\nconst hello = v.scene(\"Hello\", 2.5, (s) => {\n  s.backdrop();\n  const kicker = s.kicker(\"LIVE\");\n  const title = s.text(\"Round-ups\", { size: 120, weight: 700, font: \"display\" });\n  const sub = s.text(\"on every card\", { size: 40, color: \"#8fb3a6\" });\n  s.stack([kicker, title, sub], { gap: 20 }); \u002F\u002F frame-composition: a title block\n  title.enter({ from: { scale: 1.2, opacity: 0 }, dur: 0.3, ease: \"outExpo\" });\n  sub.enter({ from: { offsetY: 24, opacity: 0 }, at: 0.15, dur: 0.25 });\n});\n\nconst outro = v.scene(\"Outro\", 1.8, (s) => {\n  s.backdrop({ colors: [\"#0f3b2e\", \"#06110e\"] });\n  const counter = s.text(\"$0.00\", { size: 96, font: \"display\" });\n  counter.count({ from: 0, to: 1.6, dur: 1.2, prefix: \"$\", decimals: 2 });\n  s.finish({ vignette: 0.2, grain: 0.05 });\n});\n\nv.series([hello, v.transition(\"wipe\", 0.3, { direction: \"left\" }), outro]);\n\nconst saved = await v.save(nodetool.timelines, { name: \"Round-ups\", showcase: true });\nif (saved.warnings.length) {\n  \u002F\u002F read them before calling this done — see Review pass below\n}\n```\n\n## Author and revise with code\n\n`v.save(nodetool.timelines, {...})` attaches the code that rebuilds the\ntimeline as its source — no separate `code.set` call. The stored code is the\nretained program: the scene callbacks and the helpers they use, with every\nresearch result, generated asset id and loop value written in as a literal.\nThe calls that produced those values ran once, in your action, and a rebake\nnever repeats them. `saved.code.embedded` is false, with the reason in\n`saved.code.warnings`, when a scene callback calls `nodetool.*`, uses\n`Math.random()`\u002F`Date.now()`, or reads a value that is not plain data. Do the\nresearch and generation at the top level, then build the scenes from the\nresults. To attach code after the fact,\n`nodetool.timelines.code.set(timeline_id, code)` bakes the code, which must\nbuild the timeline without a capability call.\n\nRevising a code-backed timeline is: `code.get(id)` → `code.edit(id, edits)`\nwith a few small exact string replacements (each `old` matches exactly once)\n→ read the `conflicts` and `warnings` it answers → look at frames. This is\nthe review → revise → rerun loop from the build loop above, made cheap — a\ntwo-line diff instead of a 500-line resend. `edit_timeline` ops still apply\nfor a timeline with no code, or a deliberate hand tweak — `timeline-edit-ops`\nowns that contract and what it costs a code-backed scene. The pack's own\nskill (`@nodetool-ai\u002Fsandbox-timeline`) covers what code-backed identity\nmeans for scene names and clip ids; `api-timelines` owns the full `code.*`\ncall contract.\n\n## Review pass\n\nFor a showcase, hero, launch, or \"best\" piece, critique like a director before\ncalling it done. A render with no errors is not a finished piece — this pass is\npart of the job, not a step to skip when the frames look clean. Reserve rounds\nfor it: the run budget is shared with authoring, so budget for at least one\nfull pass plus one revision, not just the build.\n\n1. **Contact sheet.** `preview_timeline_frame` with `sheet: true`, sampling\n   every scene at its entrance, its hero hold, and its exit.\n2. **Per-scene checklist**, answered in writing before you touch an op:\n   - What does the eye follow first, and is that the intended hero?\n   - What moves in the background? A static bed behind the action is a\n     finding, not a pass.\n   - Does any text touch or get crossed by a shape?\n   - How much of the frame is actually used? Look yourself — the validator's\n     `showcase_frame_underfilled` check is a bounding-box proxy and can miss a\n     thin outline or ring that reads as sparse against an empty backdrop.\n   - Does every number that is supposed to count actually count, and finish\n     while it is still on screen?\n   - How does the scene exit and the next one enter? A default crossfade on a\n     showcase brief is a finding to name, not the absence of one.\n3. **Compare against a reference.** Pick the closest scene with\n   `examples.list()` \u002F `examples.get(slug)`. Render matching-beat sheets of\n   your piece and that example with `preview_timeline_frame`, `sheet: true`,\n   mapping beats across differing durations and sampling the middle of a\n   move. Call `view_image` on both with this explicit question: \"Compared\n   with the example, what is missing in layer count, secondary motion,\n   texture and light, and camera depth? Name the beat and the gap.\" Read how\n   the example scene was actually built with `examples.source(slug)` when the\n   clip excerpt alone does not explain a moment — a repeater's phase offsets,\n   a two-pass text effect, a shared id scheme. `compare_timeline_frames`\n   cannot target a shipped example — both its sides must be your own timeline\n   (an id, version or document) — so use it after a revision to prove which\n   frames actually changed, not to judge against a reference.\n4. **Validate at the showcase tier.** `validate_timeline` with\n   `{\"timeline_id\": \"\u003Cid>\", \"tier\": \"showcase\"}` (in code,\n   `nodetool.timelines.validate(id, {tier: \"showcase\"})`) returns `errors`\n   and `warnings` — there is no `issues` field.\n   Every `showcase_*` warning is a hard stop: resolve it, or state the\n   specific artistic exclusion to the user before rendering. Expect\n   `showcase_scene_groups_missing`, `showcase_custom_motion_missing`,\n   `showcase_finish_missing`, `showcase_camera_missing`,\n   `showcase_layer_density_low`, `showcase_scene_group_density_low`,\n   `showcase_custom_animation_density_low`, `showcase_effect_variety_low`,\n   `showcase_clip_density_low`, `showcase_motion_blur_missing`, plus\n   `showcase_frame_underfilled`, `showcase_text_collision`,\n   `showcase_motion_range_narrow` and `showcase_dead_motion`. Each names a\n   shipped example to study. Zero warnings is a structural floor, not a\n   craft verdict — the medians in the warning text are the design target.\n5. **Revise, then re-run this pass.** Make at least one revision that changes\n   the composition itself, not only a fix for a validator error, before\n   calling the piece done. Then repeat steps 1–4 on the result. Stopping at\n   the first clean render is the failure this pass exists to catch.\n",{"name":5,"featured":45,"description":7},true,"main","c7226baa82daca42a19a5855c116544f97aa8bac","packages\u002Fsystem-skills\u002Fmotion-graphics",91,"https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool","https:\u002F\u002Fnodetool.ai\u002F","https:\u002F\u002Fgithub.com\u002Fnodetool-ai","No-code AI development","2026-11-15 03:27:49",{"source":8,"skillName":5,"commands":56,"prompt":66,"zipAvailable":45},[57,60,63],{"tool":58,"command":59},"npx","npx skills add nodetool-ai\u002Fnodetool --skill motion-graphics",{"tool":61,"command":62},"bunx","bunx skills add nodetool-ai\u002Fnodetool --skill motion-graphics",{"tool":64,"command":65},"pnpm","pnpm dlx skills add nodetool-ai\u002Fnodetool --skill motion-graphics","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fmotion-graphics\u002FSKILL.md",[],[69,78,113,142,156,172],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":73,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":77},"01M742MX3Y2ZYK884ZXTRFKB9P","humanizer","Rewrite AI-sounding text so it reads like the writer without changing what it says.\nUse when editing or reviewing prose for AI tells: not-X-but-Y contrasts, one-line\nclosers, staged openers, forced triads, dashes everywhere, inflated claims, sales\nlanguage, stock AI words, bold labels, or filler.","MIT","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fhumanizer\u002FSKILL.md",[],[],"2026-11-11 15:11:14",{"skillNo":79,"name":35,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":93,"publishedAt":40},"01M6BY4D93R0Q9400ZHJ04DXP1","Use the NodeTool MCP server to build, validate, run, and debug visual AI workflows, and to work with assets, collections, and jobs. Use when the user mentions NodeTool or asks for a workflow graph, a node, or a media generation run on their NodeTool server.","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fplugins\u002Fchatgpt\u002Fnodetool\u002Fskills\u002Fnodetool\u002FSKILL.md",[83,88],{"categoryNo":84,"categoryName":85,"taxonomyType":86,"taxonomyTypeDesc":87},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",{"categoryNo":89,"categoryName":90,"taxonomyType":91,"taxonomyTypeDesc":92},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[35,94,95,96,97,98,99,100,101,31,102,26,103,104,105,106,107,108,109,110,111,112],"MCP","Media","workflows","when","generation","media generation","with","visual","collections","node","work","debug","graph","validate","assets","run","server","mcp server","workflow",{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":122,"publishedAt":40},"01M6BY4AEB05ZPBRVD078ATR2H","wan-2-6-prompting","Prompt Alibaba's Wan 2.6 video — text-to-video with a global style line plus timing-bracketed shots, and image-to-video that describes only what changes over time. Covers the shot-mode switch per provider, the prompt budget, negative prompts, motion intensity wording, audio input and prompt expansion. Use whenever the model id names a Wan 2.6 video route (alibaba\u002Fwan-2.6\u002Ftext-to-video, alibaba\u002Fwan-2.6\u002Fimage-to-video, wan-video\u002Fwan-2.6-t2v, wan-video\u002Fwan-2.6-i2v, wan-video\u002Fwan2.6-i2v-flash) on generate_video, animate_image or a TextToVideo node. Not for Wan 2.5 or 2.7, or the Wan 2.6 image routes.","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fwan-2-6-prompting\u002FSKILL.md",[119],{"categoryNo":120,"categoryName":121,"taxonomyType":86,"taxonomyTypeDesc":87},"01JBSEEDCAT000000000000016","图像与多媒体",[123,124,125,126,127,128,129,103,130,131,132,133,134,135,136,36,137,138,139,140,141],"prompting","wan 2 6 prompting","line","expansion","input","image-to-video","text-to-video","route","only","mode","time","flash","Budget","generate","image","animate","model","Audio","Video",{"skillNo":143,"name":144,"nameZh":6,"description":145,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":146,"categories":147,"keywords":150,"publishedAt":40},"01M6BY47T4748ZG5ZMTAR97ZWZ","video-workflow","Build a reusable NodeTool video workflow for batches or repeated inputs. For a single finished piece, use a storyboard skill.","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fvideo-workflow\u002FSKILL.md",[148,149],{"categoryNo":120,"categoryName":121,"taxonomyType":86,"taxonomyTypeDesc":87},{"categoryNo":84,"categoryName":85,"taxonomyType":86,"taxonomyTypeDesc":87},[112,151,141,152,153,29,154,31,155,35,26],"video workflow","reusable","storyboard","batches","single",{"skillNo":157,"name":158,"nameZh":6,"description":159,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":160,"categories":161,"keywords":163,"publishedAt":40},"01M6BY4587N2SYN315SXGEMZC3","video-clone","Analyze a reference video and recreate its shot structure with a new product or cast in NodeTool.","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fvideo-clone\u002FSKILL.md",[162],{"categoryNo":120,"categoryName":121,"taxonomyType":86,"taxonomyTypeDesc":87},[164,165,141,100,166,167,35,168,169,170,171],"clone","video clone","cast","new","analyze","product","structure","reference",{"skillNo":173,"name":174,"nameZh":6,"description":175,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":176,"categories":177,"keywords":179,"publishedAt":40},"01M6BY4257C2KQYYMD9EK6DK8E","video-audio-continuity","Keep sound continuous across a multi-scene piece cut from generated video — why one clip per scene hard-cuts the audio at every boundary, when to write all the scenes into a single generation instead, the sound-design vocabulary that survives a provider's copyright filter, and how to check the length, the aspect ratio and the mix of what came back. Use when a piece has more than one scene and the video model writes its own audio (Seedance 2, Veo 3, MiniMax H3, Kling 2.6 and later), or when a cut's sound drops out at the shot changes.","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fvideo-audio-continuity\u002FSKILL.md",[178],{"categoryNo":120,"categoryName":121,"taxonomyType":86,"taxonomyTypeDesc":87},[180,141,140,181,182,183,184,185,186,187,188,189,190,97,191,192,139,193,194,195,196],"video audio continuity","continuity","write","why","back","veo 3","at","cut","later","across","filter","generated","how to","mix","design","boundary","how"]