{"name":"Agent Ready API","description":"Score any website against the Vercel Agent Readability Spec, the llmstxt.org specification, and agent-protocol specs (MCP, A2A, agents.json, agent-permissions.json, UCP, x402, NLWeb). 68 checks across discovery, structure, and context. Returns per-check pass/fail with actionable fix guidance.","version":"1.0.0","when_to_use":"Use Agent Ready when you need to evaluate or improve how well a website supports AI agents and LLMs — for example before citing a source, integrating a third-party API, or auditing your own site's agent-readiness. It scores any public URL against the Vercel Agent Readability Spec, llmstxt.org, and agent-protocol specs (MCP, A2A, agents.json, NLWeb) and returns per-check pass/fail with fix guidance.","instructions":"To run a scan: POST a URL to the start-scan capability, then poll get-scan with the returned id until `status` is `complete` and read the `checks` array for pass/fail and remediation. For quick interactive use, call the no-auth mcp-apps endpoint (no API key required); for automated workflows, authenticate the REST API or MCP server with a Bearer key issued at /dashboard/api-keys. All operations are read-only and safe to call — respect the advertised rate limits.","auth":{"type":"bearer","description":"Issue API keys at https://agent-ready.dev/dashboard/api-keys. Rate limits: 10 requests/minute, 200/day on the Pro tier."},"payment":{"protocol":"x402","description":"Agent-facing pay-per-scan over HTTP 402 for autonomous agents that carry a wallet and call /api/scan directly (anonymous, over the free quota). POST https://agent-ready.dev/api/scan with an X-PAYMENT header (USDC on Base mainnet): $0.02 for a 25-page scan, $0.25 for 250 pages. The authenticated REST API (/api/v1/scans) uses Bearer API keys, not x402.","discovery":"https://agent-ready.dev/.well-known/x402"},"openapi":"https://agent-ready.dev/api/v1/openapi.json","capabilities":[{"name":"start-scan","description":"Start a new agent-readability scan. Returns a scan id; poll the get-scan operation until status is `complete`.","endpoint":"https://agent-ready.dev/api/v1/scans","method":"POST"},{"name":"get-scan","description":"Retrieve scan results by id, including per-check pass/fail, scores, and remediation guidance.","endpoint":"https://agent-ready.dev/api/v1/scans/{id}","method":"GET"},{"name":"mcp","description":"MCP server endpoint. Use the canonical MCP server card at /.well-known/mcp.json for client auto-discovery.","endpoint":"https://agent-ready.dev/api/v1/mcp","method":"POST"},{"name":"mcp-apps","description":"No-auth MCP Apps (OpenAI Apps SDK) endpoint. Exposes a ui:// resource and _meta.ui tools that render an interactive agent-readability score-card widget in-conversation (Agent-to-UI / generative UI). Three tools — scan_site, get_scan, and ask (NL search over Agent Ready's methodology, checks, and specs) — no API key required.","endpoint":"https://agent-ready.dev/api/apps/mcp","method":"POST"}],"contact":{"documentation":"https://agent-ready.dev/docs/api","homepage":"https://agent-ready.dev"}}