This is dogfooding as code. It spins up a browser, walks your app like an actual user (typing, clicking, watching console), and won't let you skip interactions or fake timestamps. Hard gates on console errors, 5xx responses, layout collapse, and accessibility violations mean no wiggle room on the stuff that breaks production. The interaction manifest tracks every click with timestamps, and there's an audit-the-audit meta-check that rejects rushed reports where timings look implausible. Runs eleven stress scenarios, tests responsive breakpoints, and produces a ranked findings list with reproduction steps. Use it before deploy when you need confidence that real usage patterns won't explode, not just that the DOM looks okay in a static pass.

skills.sh installs
3K
GitHub stars
1.1K
last updated
Aug 21, 2026

Install

$ npx -y skills add jezweb/claude-skills --skill ux-audit --agent claude-code

Installs into .claude/skills of the current project.

Files

UX Audit

Walk through a live web app AS a real user. The audit is interaction-first — typing, clicking, sending, watching, screenshotting. A static DOM sweep cannot produce a verdict.

Verdict states

The audit ends in exactly one of:

  • Pass — Critical = 0, High = 0, all hard gates green, Interaction Manifest complete.
  • Conditional Pass — Critical = 0, High = 0, all hard gates green, but Medium/Low present.
  • Fail — at least one Critical or High finding, OR a hard gate red.
  • Incomplete — Interaction Manifest missing required entries, a phase wasn't run, OR the audit-the-audit meta-check fires (manifest timestamps clustered < 0.5s apart, screenshots fewer than 2 × routes, console reads fewer than 1 × routes, Phase 3 took < 1m for an exhaustive audit). Not legal to upgrade to Pass even if everything observed looked fine.

If the work doesn't include a complete Interaction Manifest, the only legal verdict is Incomplete. "It looked OK" is not Pass. A clean Pass with implausible timings is rejected — the agent must redo the audit with real interaction.

Hard gates

These auto-fail the audit. They cannot be downgraded.

Sponsors

Details

Repository
github.com/jezweb/claude-skills
Categories
Frontend Development, Testing & QA, Debugging, Automation & Workflows, Design & UI/UX

More from jezweb/claude-skills

All 166
1nexu-io avatarweb-artifacts-buildernexu-io/open-designBuild complex claude.ai HTML artifacts with React and Tailwind. Anthropic's reference workflow for shipping rich, embeddable artifacts.3Kskills.sh installs
2triggerdotdev avatartrigger-realtimetriggerdotdev/skillsSubscribe to Trigger.dev task runs in real-time from frontend and backend. Use when building progress indicators, live dashboards, streaming AI/LLM responses, o2.9Kskills.sh installs
3nexu-io avatarartifacts-buildernexu-io/open-designSuite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui).2.9Kskills.sh installs
4nexu-io avatardigits-fintech-swiss-templatenexu-io/open-designSwiss-grid fintech deck template in black / warm paper / neon-lime contrast. Use when users ask for premium data-story slides with strict modular layout, bold n2.9Kskills.sh installs
5nexu-io avatarswiss-user-research-video-templatenexu-io/open-designSwiss-style user-research narrative template in warm-paper editorial aesthetics. Use when users ask for a premium research deck or story-first live artifact wit2.9Kskills.sh installs
6affaan-m avatarmotion-uiaffaan-m/eccProduction-ready UI motion system for React/Next.js. Use when implementing animations, transitions, or motion patterns.2.8Kskills.sh installs