Skip to main content

Website Audit

POST 
/audit

Performs a comprehensive SEO and AI-readiness audit of a specified URL. The audit analyzes:

  • Site files (robots.txt, llms.txt, SKILL.md, .well-known directory)
  • SEO fundamentals (title, description, headings)
  • Structured data (Schema.org, JSON-LD, Microdata)
  • Content structure, semantic HTML, and token budget
  • Image accessibility
  • Automation readiness for AI agents
  • JavaScript rendering and bot accessibility
  • MCP / WebMCP discovery surfaces:
    • <link rel="mcp"> HTML discovery tag
    • /.well-known/mcp.json (legacy MCP manifest)
    • /.well-known/mcp/server-card.json (MCP server card, SEP-1649 draft)
    • /.well-known/webmcp/tools.json (Chrome Labs WebMCP tools manifest)
  • Agent Skills Discovery (/.well-known/agent-skills/index.json, Cloudflare RFC v0.2.0) and top-level SKILL.md
  • Deterministic token budget (Readability + Turndown, chars/4 heuristic)
  • Content freshness (legacy field, status Unknown)
  • Internal linking (legacy field, status Unknown)
  • HTML semantics (legacy field, status Unknown)

Every file the audit fetched or detected is reported in resources, which tells a missing file apart from one that could not be fetched.

Returns an overall score (0-100) and detailed recommendations for improvement. The overall score is the clamped sum of the per-criterion scores emitted by the LLM plus post-hoc bonuses: +1 for <link rel="mcp">, +1 for WebMCP tools.json, +1 for legacy mcp.json or mcp/server-card.json, +2 for Agent Skills Discovery or top-level SKILL.md, and +2 when the estimated token budget is ≤ 30 000.

Request​

Responses​

Audit completed successfully