agentleFS
augchan42augchan42/inkstone/skills/motion-craft/SKILL.md

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

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
  1. Distinctive:

    Prefer stack-native techniques.

    line 57
  2. Distinctive:

    Never add a new animation dependency without asking.

    line 61
  3. Very specific:

    Write per-section specs into <target_repo>/docs/MOTION.md inside managed sections — replace a matching section on re-run, never append a …

    line 73
  4. Very specific:

    If docs/MOTION.md does not exist yet, create it: a minimal top-level heading (e.g.

    line 83
  5. Very specific:

    Every data row must have exactly 13 cells matching the header.

    line 87
  6. Distinctive:

    If the working tree is dirty with unrelated changes: stop and report.

    line 103
  7. Distinctive:

    Prefer a dedicated branch.

    line 104
  8. Distinctive:

    Commit per section only when the worktree was clean at start and that section's gates passed.

    line 105
  9. Very specific:

    Every implemented section passes references/verification-gates.md before you move to the next.

    line 111

“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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

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 plan

Plan 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 → cwd

Canonicalize; 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

  1. Confirm the target is a git repository; refuse implement mode otherwise.
  2. If the working tree is dirty with unrelated changes: stop and report.
  3. Prefer a dedicated branch.
  4. 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>/.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.