awesome-motion-design-agent-skills
AGENTS.md111 starsChanged 11 days agoCC0-1.0View on GitHub
This repo is the public motion-agent discovery and authority layer for FrankX, Starlight, and Arcanea. For any site, app, landing page, dashboard, visual identity, brand, motion, media, social, or frontend task, apply the maintainer's Design Taste Kernel (design taste, web experience standard, motion taste rubric, multi-agent design council, visual QA gate) before handoff. Those documents live in the maintainer's private estate and are intentionally not referenced by path here. When motion, scroll, generated media, GIF/video, or premium polish matters, route…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 32,671 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository6 other files it gives its agents.
- .github/copilot-instructions.mdCopilot instructions
- generative-media-directorSkill
- motion-directorSkill
- remotion-video-agentSkill
- site-motion-excellence-directorSkill
- web-motion-auditorSkill
The file
Show its rules and text
We split this AGENTS.md into the rules it gives an agent
17 rules. About 1 is common, 10 are distinctive, and 6 are very specific to this project.
≈1
Common · estimate
Many files like this have a rule of this kind.
≈10
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Prefer official sources, active repos, clear licenses, and examples that improve shipped output.
- Common: line 9
Keep this repo public-safe: no private brand strategy, internal credentials, paid assets, client work, non-public memory, or local machine …
- Distinctive: line 15
Every new tool entry should answer: when to use it, why it matters, agent fit, risk, and source/license notes.
- Distinctive: line 16
Avoid giant unsorted lists.
- Distinctive: line 21
Prefer primary links: official docs, official GitHub repos, vendor MCP docs, and standards.
- Very specific: line 22
Keep skills-lock.json aligned with third-party skills actually referenced by the repo.
- Distinctive: line 23
Do not commit installed third-party skill folders unless the source license and purpose are explicit.
- Distinctive: line 27
Before publishing, scan for broken Markdown links, stale dates, placeholders, and private/internal paths.
- Distinctive: line 28
If adding setup commands, verify they are current and make clear whether they are global, repo-local, or agent-environment-specific.
- Very specific: line 34
When motion, scroll, generated media, GIF/video, or premium polish matters, route through the skills in skills/ and verify the result …
- Distinctive: line 44
When the private Premium Web OS is present in the operator environment, read it first and treat it as the source of truth for taste, …
- Very specific: line 45
Use /pwo or the premium-web-os skill for full builds.
- Very specific: line 46
Use /pwo review-pr before absorbing another agent's PR or branch.
- Very specific: line 47
Use /pwo absorb-assets before using external, generated, scientific, audio, video, or 3D assets.
- Very specific: line 48
Use /pwo motion-score before shipping cinematic scroll, sound-paired motion, or complex choreography.
- Distinctive: line 51
Do not copy reference sites or agencies.
- Distinctive: line 52
Do not ship without responsive, accessibility, performance, reduced-motion, and visual QA checks appropriate to the change.
≈ “Common” is an estimate: at least 1 in 5 of the 41,581 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Repository Instructions
This repo is the public motion-agent discovery and authority layer for FrankX, Starlight, and Arcanea.
Scope
- Maintain curated rankings, playbooks, skills, MCP notes, and rubrics for AI agents doing motion design.
- Prefer official sources, active repos, clear licenses, and examples that improve shipped output.
- Keep this repo public-safe: no private brand strategy, internal credentials, paid assets, client work, non-public memory, or local machine paths (CI enforces the last one).
Quality Bar
- Rank by usefulness to real agent workflows, not by hype or stars alone.
- Separate product UI motion, creative web motion, interactive assets, video generation, capture/QA, and MCP/design-tool context.
- Every new tool entry should answer: when to use it, why it matters, agent fit, risk, and source/license notes.
- Avoid giant unsorted lists. Curate aggressively.
Maintenance
- Refresh external links and rankings at least monthly.
- Prefer primary links: official docs, official GitHub repos, vendor MCP docs, and standards.
- Keep
skills-lock.jsonaligned with third-party skills actually referenced by the repo. - Do not commit installed third-party skill folders unless the source license and purpose are explicit.
Verification
- Before publishing, scan for broken Markdown links, stale dates, placeholders, and private/internal paths.
- If adding setup commands, verify they are current and make clear whether they are global, repo-local, or agent-environment-specific.
Design Taste Kernel
For any site, app, landing page, dashboard, visual identity, brand, motion, media, social, or frontend task, apply the maintainer's Design Taste Kernel (design taste, web experience standard, motion taste rubric, multi-agent design council, visual QA gate) before handoff. Those documents live in the maintainer's private estate and are intentionally not referenced by path here.
When motion, scroll, generated media, GIF/video, or premium polish matters, route through the skills in skills/ and verify the result visually.
<!-- PREMIUM-WEB-OS:START -->Premium Intelligence Web OS Adoption
This repo participates in the Starlight Premium Intelligence Web OS.
For any website, app, landing page, dashboard, brand surface, visual asset, motion system, 3D/WebGL scene, generated media, or public-facing UI work:
- When the private Premium Web OS is present in the operator environment, read it first and treat it as the source of truth for taste, motion, WebGL, copy, assets, and quality gates.
- Use
/pwoor thepremium-web-osskill for full builds; use/madfor a design council pass. - Use
/pwo review-prbefore absorbing another agent's PR or branch. - Use
/pwo absorb-assetsbefore using external, generated, scientific, audio, video, or 3D assets. - Use
/pwo motion-scorebefore shipping cinematic scroll, sound-paired motion, or complex choreography. - Build static composition first, add Track A local motion second, add Track B GSAP/Lenis scroll only when earned, and add 3D only with fallback and reduced-motion behavior.
- Record asset provenance, rights, and publication status for every generated or external asset.
- Do not copy reference sites or agencies. Deconstruct principles and create original execution.
- Do not ship without responsive, accessibility, performance, reduced-motion, and visual QA checks appropriate to the change.
Repo-local instructions remain authoritative when stricter.
<!-- PREMIUM-WEB-OS:END -->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.

