agent-audit

Agent Audit MCP Server

Agent Audit turns Lighthouse results into structured fix packs for developers working with coding agents. It covers performance, accessibility, technical SEO, AI crawler signals, and link and image issues, with evidence and criteria for checking fixes.

Server facts

agent-audit MCP Server

Project

License
Not specified

Tech

Runtime / language
Not specified
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
4
Last commit
Data as of

Tags

  • Accessibility
  • Claude
  • Cursor
  • Lighthouse
  • Seo
  • ai agent
  • core web vitals
  • mcp

Overview

Agent Audit connects website checks with code changes by turning Lighthouse output for a supplied URL into fix packs for coding agents. Its analyze_website_performance tool audits mobile and desktop pages and delivers structured results alongside a Markdown version. Performance reporting covers FCP, LCP, CLS, and TBT using medians from 3 runs. The audit scope also includes accessibility findings with selectors for affected elements, plus technical SEO checks involving canonical tags, robots, metadata, JSON-LD, and Open Graph.

The fix packs pair evidence from unsuccessful checks with repository search guidance and implementation instructions. Selectors and URLs help identify the reported problems, while acceptance criteria describe how to check a correction by repeating the audit. Beyond performance and accessibility, Agent Audit examines AI crawler signals and readiness for llms.txt, as well as broken links, absent image alt text, and image optimization. The workflow serves developers using coding agents to move from website audit findings toward specific repository changes and subsequent verification.

agent-audit setup notes

  • agent-audit records stdio only; confirm a network transport upstream before planning remote access.
  • agent-audit has no recorded license; confirm reuse terms in its repository.

Related MCP servers