agentleFS
muzimu217muzimu217/ui-design-agent-kit/AGENTS.md

ui-design-agent-kit

AGENTS.md23 starsChanged 27 days agoApache-2.0View on GitHub

This repository maintains a UI-specialist agent, not an application scaffold. Improve its instructions, integrations, and evaluations when asked. Do not create a demo app unless the user requests an interface or a runnable example. For UI requirements planning, reference/material research, design, or implementation, read .agents/skills/ui-design-agent/SKILL.md and use its tool routing. It is the entrypoint for product reasoning, visual direction, implementation, motion, and evidence-based verification. Communicate in the user's language. The installed supporting skills are: - uak-design-system: local design-system and UX…

The file

We split this AGENTS.md into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Keep the main behavior in ui-design-agent/SKILL.md.

    line 38
  2. Very specific:

    Keep installed upstream skills pinned in tooling/sources.lock.json and preserve license notices.

    line 41
  3. Distinctive:

    Do not enable or authenticate optional services as a side effect.

    line 45

≈ “Common” is an estimate: at least 1 in 5 of the 42,283 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.

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

UI Design Agent Kit

This repository maintains a UI-specialist agent, not an application scaffold. Improve its instructions, integrations, and evaluations when asked. Do not create a demo app unless the user requests an interface or a runnable example.

UI work

For UI requirements planning, reference/material research, design, or implementation, read .agents/skills/ui-design-agent/SKILL.md and use its tool routing. It is the entrypoint for product reasoning, visual direction, implementation, motion, and evidence-based verification. Communicate in the user's language.

The installed supporting skills are:

  • uak-design-system: local design-system and UX lookup.
  • uak-visual-critique: visual critique, targeted refinement, and design craft (adapted from the upstream impeccable skill).
  • anatomy: crafted isometric SVG figures that explain a technical idea as a physical machine — for docs/landing-page figures and premium technical drawings (whole-package vendored, MIT, pinned 2026-10-16).
  • motion: official Motion animation guidance and MCP workflows.
  • remotion-video-agent: local orchestrator for React video composition, frame math, transitions, Studio preview, and render verification.
  • remotion-best-practices, remotion-create, remotion-docs, remotion-markup, remotion-render, and remotion-studio: official Remotion Agent Skills loaded on demand.
  • video-edit-agent family (video-edit-assembly, video-recap-workflows, video-speech-workflows, env-setup): real-footage editing — material discovery, assembly, narrated recaps, speech condensing, and captions, enforced through the video-agent-kit plugin's own Python MCP. Host-provided and not vendored here: this kit routes to it but does not ship or pin it.

Load the minimum relevant supporting instructions. Explicit user requirements, the target project's established design system, and task scope outrank generic style recommendations from these skills. Never infer permission to redesign, deploy, buy services, upload private code, or alter global configuration.

Maintain this kit

  • Keep the main behavior in ui-design-agent/SKILL.md; avoid parallel copies of a long system prompt. remotion-video-agent/SKILL.md owns the video-specific behavior; docs/usage.md contains invocation examples, not policy.
  • Keep installed upstream skills pinned in tooling/sources.lock.json and preserve license notices. Record adaptations in THIRD_PARTY_NOTICES.md.
  • MCP configuration lives in .codex/config.toml. motion_plus and figma are optional and disabled. The deprecated Remotion MCP is intentionally absent. Do not enable or authenticate optional services as a side effect.
  • npm ci --ignore-scripts, npm run verify, and npm test verify the kit. npm run doctor:mcp contacts the configured public services and starts the configured local servers for read-only smoke checks; it does not log in.
  • Tests must distinguish installed instructions, valid configuration, a live MCP connection, a successful tool call, and a visually verified application.
  • Prompt evaluation cases live in evals/scenarios.json. Static validation is not an agent-behavior evaluation or proof of design quality.

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.