web-quality-audit

Audit websites for performance, accessibility, SEO, and best practices using live browser evidence.

Updated May 24, 2026
One-click install
npx skills add https://github.com/MWest2020/skill-forge --skill web-quality-audit-mwest2020
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: web-quality-audit
Source: https://github.com/MWest2020/skill-forge/tree/main/skills/web-quality-audit
Command: npx skills add https://github.com/MWest2020/skill-forge --skill web-quality-audit-mwest2020

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires jq, and includes scripts (resource) components.

What problem does it solve? Website quality issues are scattered across performance, accessibility, SEO, and security, and a single aggregate score rarely tells you what to fix. This Skill runs an evidence-led audit that combines live browser measurements with source inspection, then categorizes findings by severity and user impact. ## Core Features & Use Cases - Five audit categories: Covers Performance (Core Web Vitals: LCP, INP, CLS), Accessibility (WCAG), SEO, Best Practices, and Agentic Browsing in one structured review. - Tool routing with fallbacks: Uses Chrome DevTools MCP traces and Lighthouse audits when available, falling back to Lighthouse CLI, PageSpeed Insights, or manual checks. - Static source smoke test: Ships a read-only scripts/analyze.sh analyzer that scans HTML files for missing doctype, viewport, lang, alt attributes, and non-HTTPS URLs, emitting structured JSON. - Use Case: Ask for an audit of your staging site before launch; receive a severity-ranked report with measured evidence, file-level fixes, and a verification plan. ## Quick Start Ask the agent to audit your website's performance, accessibility, and SEO, providing the target URL or a path to local HTML files.

Frequently Asked Questions about web-quality-audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit my website for performance and accessibility?

Provide the target URLs and scope, and the audit collects a live baseline using browser performance traces and Lighthouse, then inspects source code to localize issues. Findings are categorized by severity with measured evidence separated from code-only hypotheses.

What are good Core Web Vitals thresholds for LCP, INP, and CLS?

LCP should be under 2.5 seconds, INP under 200 milliseconds, and CLS under 0.1. The audit measures these via browser traces or CrUX field data and reports whether each metric passes.

Does the audit work without Chrome DevTools MCP installed?

Yes. Every capability has a fallback: Lighthouse CLI or PageSpeed Insights for lab data, manual checks for accessibility and interaction, and the bundled analyze.sh script for static HTML source inspection.

What does the analyze.sh script check in HTML files?

It scans HTML files for missing doctype, charset, viewport meta, lang attribute, title tag, img elements without alt text, and non-HTTPS URLs. It is read-only and outputs structured JSON with issue counts and truncation flags.

Why does the audit not treat a Lighthouse score as proof of quality?

Aggregate scores hide category-specific failures and some retired audits were noisy or inactionable. The audit keeps measured findings separate from hypotheses and requires re-running checks to verify fixes.