Back to directory

Web Quality Skills

A stack-agnostic collection of Agent Skills that measure and improve a website's performance, accessibility, and SEO using Lighthouse, Core Web Vitals, and Chrome DevTools.

Web Quality Skills

Web Quality Skills is an unofficial, measurement-first collection of Agent Skills for improving web projects against Google Lighthouse, Core Web Vitals, WCAG accessibility standards, and search-engine guidance, working with any framework including React, Vue, Angular, Svelte, Next.js, Nuxt, and Astro. Where interface guidelines describe what to build, these skills focus on how to verify that it performs, is accessible, and is discoverable.

Individual skills run live Lighthouse audits and evidence-led reviews when Chrome DevTools MCP is available, diagnose performance traces for issues affecting core loading and responsiveness metrics, apply WCAG 2.2 accessibility checks, cover modern SEO requirements, and apply general security and code-quality practices. The performance skills keep four kinds of evidence distinct: real-user field data, first-party analytics, a single controlled trace or Lighthouse run, and static source inspection, so a recommendation is not based on the wrong kind of measurement.

It is distributed as Shell scripts under the MIT license.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.