motion-craft
Skill5 starsChanged 5 months agoMITView on GitHub
Implement professional web animations in a target repo against an approved design blueprint or an explicit motion direction. Use when asked to "add animations", "improve the motion", "make it feel alive", "implement the motion plan", or when dispatched by the upgrade-site skill. Plans by default; modifies code only on explicit implement authority. For broad visual/reference research, see design-dna instead.
6 other projects use a version of this.
Adopted by
Most recently taken up by
- XuHo-IT/Content-Agent-Kittheir own version, first seen Nov 8, 2026
- EnamTiga/motion-crafttheir own version, first seen Nov 8, 2026
- sithumdgomez/claude-skillstheir own version, first seen Oct 11, 2026
- aegntic/video-shotcrafttheir own version, first seen Oct 9, 2026
- jakubsuplicki/codumenttheir own version, first seen Oct 4, 2026
- propiter/motion-crafttheir own version, first seen Jun 15, 2026
Other versions6 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 6
- jakubsuplicki/codument★ 48same name
- XuHo-IT/Content-Agent-Kit★ 3same name
- aegntic/video-shotcraft★ 1same name
- EnamTiga/motion-craft★ 1same name
- propiter/motion-craft★ 1same name
- sithumdgomez/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository28 other files it gives its agents.
- AGENTS.mdAGENTS.md
- analyze-variationSkill
- arxiv-searchSkill
- blakean-sceneSkill
- codex-reviewSkill
- create-explanationSkill
- critic-loopSkill
- deepen-taskSkill
- deliver-changeSkill
- design-dnaSkill
- gloss-drift-auditSkill
- image-to-sceneSkill
- marketing-claritySkill
- market-pulseSkill
- media-kitSkill
- nextjs-i18n-seoSkill
- oss-licence-auditSkill
- plan-review-loopSkill
- shaughnessy-ichingSkill
- swapping-pptx-templatesSkill
- ui-ux-architectSkill
- unslopSkill
- upgrade-siteSkill
- vercel-build-speedSkill
- verify-changeSkill
- verse-to-promptSkill
- video-to-shortsSkill
- voice-calibrationSkill
The file
We split this skill into the rules it gives an agent
9 rules. 0 are common, 5 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 6 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 57
Prefer stack-native techniques.
- Distinctive: line 61
Never add a new animation dependency without asking.
- Very specific: line 73
Write per-section specs into <target_repo>/docs/MOTION.md inside managed sections — replace a matching section on re-run, never append a …
- Very specific: line 83
If docs/MOTION.md does not exist yet, create it: a minimal top-level heading (e.g.
- Very specific: line 87
Every data row must have exactly 13 cells matching the header.
- Distinctive: line 103
If the working tree is dirty with unrelated changes: stop and report.
- Distinctive: line 104
Prefer a dedicated branch.
- Distinctive: line 105
Commit per section only when the worktree was clean at start and that section's gates passed.
- Very specific: line 111
Every implemented section passes references/verification-gates.md before you move to the next.
“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Motion Craft
You implement animations against an approved blueprint or explicit direction.
You NEVER perform broad reference research — that is design-dna's job.
Authority mode
mode =
inherited from upgrade-site when dispatched
else explicit implement request ("implement", "apply", "make the changes") → implement
else planPlan mode writes only docs/MOTION.md sections and reports. Implement mode
may modify source, subject to Git safety below.
Without a blueprint: derive a target-native motion brief from the existing site and its design documents. Do not research references.
Target resolution
target_repo = explicit path → current git root → cwdCanonicalize; refuse / or $HOME as roots; all writes beneath the target;
state the resolved target up front.
Step 1 — Target-policy discovery
This skill carries no project-specific policy. Before planning, read the
target's AGENTS.md, CLAUDE.md, docs/DESIGN.md, docs/MOTION.md, route
conventions, and package metadata. Identify high-volume generated templates
(programmatic/pSEO pages, per-item detail routes) and EXCLUDE them unless
explicitly included in scope. Record discovered policy at the top of your
plan output — i.e. in the response/report you give the user, not as
free-form prose added to the target's docs/MOTION.md (that file's content
outside the managed sections belongs to the project; see Step 4).
Step 2 — Stack detection
Detect the animation library and version, its import convention, the CSS framework, and the rendering model (server/client components, hydration). Prefer stack-native techniques. Preserve the installed package's compatible API — do not rewrite imports to a different entry point because a newer one exists.
Never add a new animation dependency without asking. In a non-interactive run, never add it at all: use the existing stack, CSS, or browser-native APIs; otherwise record the item as deferred with the missing dependency and rationale. The pipeline must not stall.
Step 3 — Motion audit
What animates today, what is static, obvious jank (layout-affecting transitions, unthrottled scroll handlers, missing reduced-motion handling).
Step 4 — Choreography plan (upsert)
Write per-section specs into <target_repo>/docs/MOTION.md inside managed
sections — replace a matching section on re-run, never append a duplicate:
<!-- inkstone:motion-craft:start section-id=homepage-hero -->
| Element/pattern | Trigger | Initial state | Final state | Scroll range/threshold | Duration | Delay/stagger | Easing | Scrubbed or discrete | Mobile variation | Reduced-motion fallback | Evidence | Confidence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
<!-- inkstone:motion-craft:end section-id=homepage-hero -->If docs/MOTION.md does not exist yet, create it: a minimal top-level
heading (e.g. # Motion) followed by the managed section above — do not
invent additional unmanaged content.
Every data row must have exactly 13 cells matching the header. Never put a
literal | inside a cell (it silently shifts every later column); wrap
inline code or punctuation instead, and count cells before writing when a
row's content is long or generated programmatically.
Content outside managed sections belongs to the project — never touch it.
Step 5 — Implement (implement mode only)
One section at a time, following references/implementation-rules.md
(client-boundary, property, bundle, accessibility, and i18n rules — read it
before writing any code).
Git safety
- Confirm the target is a git repository; refuse implement mode otherwise.
- If the working tree is dirty with unrelated changes: stop and report.
- Prefer a dedicated branch.
- Commit per section only when the worktree was clean at start and that section's gates passed; otherwise leave changes uncommitted and provide a proposed commit breakdown.
Step 6 — Verify
Every implemented section passes references/verification-gates.md before
you move to the next. Evidence goes to
<target_repo>/.artifacts/site-upgrade/<run-id>/.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

