research-wiki-motion

Agent skill for semantic graph motion work in the research-wiki vault UI.

This is not a general animation-library skill. It captures the project's motion contract for semantic graph demos, Instagram/Reels capture shots, and Reels/Capture Mode implementation review:

• motion must explain graph relationships
• screen-level scan/glitch decoration is banned
• node/link/camera/HUD motion has explicit budgets
• candidate directions should be scored before deep implementation
• Playwright visual verification is part of the definition of done

What It Is For

Use this skill when working on:

• research-wiki graph UI motion
• Reels/Capture Mode (?capture=1)
• semantic graph animations on nodes, links, camera, or HUD
• Instagram demo shots that need attention-grabbing but restrained motion
• review of graph motion regressions, visual overlap, or unverified polish

Do not use it as a substitute for a full animation-library reference such as GSAP, Framer Motion, Motion One, or Anime.js.

Install

npx skills add https://github.com/moonweave/research-wiki-motion-skill -g -y --copy

The installed skill should appear as research-wiki-motion:

npx skills list -g

Use

Example prompts:

Use $research-wiki-motion to compare three capture-mode motion directions.

Use $research-wiki-motion to review the graph animation for Reels recording.

The skill guides the agent toward restrained, meaning-bound motion and requires browser screenshots or runtime checks before claiming visual success.

Repository Layout

.
├── SKILL.md              # Agent-facing skill instructions
├── agents/openai.yaml    # UI metadata for skill listings
├── LICENSE               # MIT license
└── README.md             # Public repo overview

Verification

Validate the skill package:

python /Users/choemun-yeong/.codex/skills/.system/skill-creator/scripts/quick_validate.py .

For research-wiki app changes made with this skill, follow the verification checklist inside SKILL.md.

License

MIT