<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>UI Audit on Top AI Skills</title><link>https://topaiskills.com/tags/ui-audit/</link><description>Recent content in UI Audit on Top AI Skills</description><generator>Hugo</generator><language>en</language><lastBuildDate>Mon, 24 Aug 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://topaiskills.com/tags/ui-audit/index.xml" rel="self" type="application/rss+xml"/><item><title>Web Design Guidelines: Vercel's Live UI Review Skill</title><link>https://topaiskills.com/skills/coding/web-design-guidelines/</link><pubDate>Mon, 24 Aug 2026 00:00:00 +0000</pubDate><guid>https://topaiskills.com/skills/coding/web-design-guidelines/</guid><description>&lt;p&gt;Vercel&amp;rsquo;s web-design-guidelines skill doesn&amp;rsquo;t write your UI. It audits it — and the rulebook isn&amp;rsquo;t even stored in the skill. The SKILL.md file is a 39-line launcher; the actual review standard ships from a separate repository and gets fetched fresh on every run. At skills.sh #37 with 570,998 installs and 21,343 installs last week (checked 2026-08-24), it&amp;rsquo;s the most-used UI audit skill on the leaderboard, published by &lt;a href="https://github.com/vercel-labs/agent-skills" rel="nofollow" target="_blank"&gt;Vercel Labs&lt;/a&gt; (30.4K stars). I read the launcher and pulled the live rulebook it points at — 190 lines, 16 rule groups — to see what a real audit actually flags.&lt;/p&gt;</description></item></channel></rss>