Evidence-Based UX Audit — paulunemoon — AI skill

When to use it

Use for a checkout flow, onboarding review or UI/microcopy diagnosis using a live app, Figma, screenshots or code. Distinct from creating new UI: outputs a shareable report and evidence bundle, not an automatic redesign. Description-only inputs yield hypotheses rather than observed defects; code or Figma cannot prove runtime behavior. Accessibility mode samples WCAG 2.1 AA, not formal conformance. Optional web3 checks are scope-gated. Needs user research and analytics to validate behavioral assumptions. Not installed, executed or runtime-tested during discovery. GitHub stars are repository-wide, not a skill-specific rating.

Compatible agents

Claude Code

Installation

Clone the complete upstream repository into ~/.claude/skills/ux-audit (git clone https://github.com/paulunemoon/ux-audit-skill.git ~/.claude/skills/ux-audit), retaining references/ and scripts/. Restart the agent. Ask for a scoped audit in plain language with URL, screenshots or code. Optional .claude/commands wrappers are documented in README; they are not required. Its optional contrast helper needs Python 3.

Review the source and instructions before installing in your agent.

License

MIT

Last checked Oct 29, 2026.

Pair this skill with a visual reference

A skill guides how your agent works; a design reference shows the direction you want. Choose an example for a specific job, such as the headline hierarchy or form layout, and keep your project's existing design rules explicit.

Start with design references you can inspect, then use our guide to briefing an AI coding agent to turn observations into requirements. If you need implementation components, compare UI libraries for your stack.

These pages are public. Browsing a skill does not install it, and TastefulKit's design-reference MCP tools do not search the AI-skill directory.