agentsclimarketplace

Audit

Skill educlopez/ui-craft/cli/assets/gemini/skills/audit

Design engineering system for AI coding agents — ship UI with craft-level quality. Install as an agent skill.

Install
npx -y skills add educlopez/ui-craft --skill audit

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Technical UI audit — a11y, performance, responsive. Produces a prioritized findings table. Invoke when the user asks for audit on their UI, or mentions 'audit' alongside design / UI / frontend work.

SKILL.md

3.8 KB, 867 tokens by cl100k_base, as published. Nobody here has run it

<!-- HARNESS MIRROR — do not edit here. Canonical source: skills/ or commands/. After editing source, copy into cli/assets/<harness>/ and repo-root harness mirrors. -->

Context: this sub-skill is one lens of the broader ui-craft skill. If the ui-craft skill is also installed, read its SKILL.md first for Discovery + Anti-Slop + Craft Test, then apply the specific lens below.

Run a technical audit of the UI at $ARGUMENTS. Load the ui-craft skill and apply the audit lens.

Step 0: Visual capture (mandatory)

Code-only review is insufficient. Every audit/critique starts with the surface as the user sees it. Try the following in order; use the first one available:

  1. Playwright MCP — if playwright MCP server is available, use it. Capture full-page screenshots at three viewports: desktop (1280×800), tablet (768×1024), mobile (375×812). Capture dark mode if the app supports it.
  2. Browser DevTools / Chrome MCP — second choice; same viewport set.
  3. Other browser automation (agent-browser, cursor-ide-browser) — third choice.
  4. Ask the user — last resort. If no automation is available, request screenshots from the user before proceeding. Be specific:
    • "Visual review needs screenshots. Please provide:
      • Full-page at 1280px (desktop)
      • Full-page at 768px (tablet)
      • Full-page at 375px (mobile)
      • Dark mode of each, if supported."

Do not begin the review until visuals are captured or provided. State this explicitly to the user when no automation succeeds — don't silently fall back to code-only review.

If the user declines to provide screenshots, run a code-only pass and clearly mark the report [CODE-ONLY REVIEW — visual issues not assessed] at the top so the limitation is explicit.

Note: audit is knob-agnostic — accessibility and performance are not tunable.

Scope (non-negotiable checks):

  1. Accessibility — read references/accessibility.md:
    • Visible :focus-visible on every interactive element
    • Keyboard reachable, no focus traps
    • Touch targets ≥ 44px (mobile)
    • Color not the only signal for state
    • Form labels, error association, required indication
    • prefers-reduced-motion honored for all animations
  2. Performance — read references/motion.md Rendering Performance section:
    • Only transform / opacity animated (no width/top/height)
    • No transition: all
    • will-change scoped to active interaction, removed after
    • Images have width/height or aspect-ratio (CLS)
    • No layout thrash in scroll/resize handlers
  3. Responsive — read references/responsive.md:
    • Mobile-first breakpoints, no fixed-width components
    • env(safe-area-inset-*) respected on fixed elements
    • Touch zones don't overlap
    • No horizontal scroll at 320px

Output format — the Review Format table from SKILL.md:

BeforeAfterWhy

Group findings by priority: Critical (blocks usability/a11y) → High-impact (immediately noticeable) → Quick wins (polish).

Do NOT rewrite code unless asked. Report findings first; wait for approval before editing.

Close with a Craft Report (references/review.md → Craft Report), wrapping the findings table above — Checked states scope (a11y/perf/responsive, at which viewports), Passed carries anything inspected and found sound, Changed stays empty pre-approval, Verdict names the top priority. Produce it even on a clean audit — a11y/perf/responsive coming back clean is the finding.

Next step: /harden — turn the findings into real production coverage (rung 1).

Gives 0 of the 12 instructions most design frontend skills give in 867 tokens

Counted across 1,170 of the 1,878 authors here whose files we hold, read 2026-08-06

  • use css variables for color consistencyin 73 of 1170, across 24 files
  • match implementation complexity to the aesthetic visionin 70 of 1170, across 20 files
  • commit to one bold aesthetic direction before codingin 70 of 1170, across 25 files
  • add atmospheric background effects and texturesin 58 of 1170, across 10 files
  • use unexpected spatial compositions and layoutsin 55 of 1170, across 7 files
  • implement real working codein 55 of 1170, across 7 files
  • vary themes and aesthetics across different designsin 48 of 1170, across 7 files
  • launch chromium in headless modein 47 of 1170, across 4 files
  • close the browser when donein 47 of 1170, across 4 files
  • run provided scripts with help flag firstin 47 of 1170, across 4 files
  • use descriptive selectors for elementsin 47 of 1170, across 4 files
  • wait for network idle statein 46 of 1170, across 3 files

Said here and by no other author read

  • eliminate horizontal scroll at 320px width
  • ask the user for screenshots if browser automation is unavailable
  • do not begin review until visuals are captured or provided
  • mark the report as code-only if no screenshots exist
  • close with a craft report wrapper
  • verify keyboard reachability without focus traps

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.