agent-ready-dev-mcp.yml
Agent Ready operates TWO hosted Streamable HTTP MCP endpoints on its apex host plus a stdio package on npm. (1) https://agent-ready.dev/api/v1/mcp — the "agent-ready" server (protocol version 2025-06-18, serverInfo agent-ready 1.0.0): initialize, tools/list, resources/list, resources/read and prompts are public so a client can introspect without a credential; tools/call for scan_site and get_scan requires a Bearer ar_live_ Pro API key, and the endpoint advertises RFC 9728 protected-resource metadata at /.well-known/oauth-protected-resource. (2) https://agent-ready.dev/api/apps/mcp — the "agent-ready-apps" MCP Apps server (OpenAI Apps SDK / MCP-UI): no auth at all, same three tools bound to a ui://widget/ agent-ready-score.html HTML resource, the surface behind the Agent Ready app in the ChatGPT directory; it is also served on the dedicated mcp.agent-ready.dev host with its own SEP-1649/SEP-2127 server card. (3) npm agent-ready-mcp — a stdio wrapper over the REST API (npx -y agent-ready-mcp@latest) that runs keyless on the anonymous free tier. Anonymous tools/list succeeded on both HTTP endpoints, so the tool set is observed, not inferred; the names, descriptions and outputSchema are Agent Ready-specific (vercelScore, llmstxtScore, checkGroups) and match no shared platform fingerprint — this is a provider-built server. The provider's /.well-known/mcp.json server card (schema 2025-10-17 server.schema.json) lists both hosted servers.
One-click install for Cursor, VS Code, Claude, and 20+ other MCP clients, powered by API Commons MCP Install — visit install.apicommons.org for more information.
Documentation
Documentation link · transport
Tools
scan_site— Scan a public URL for AI agent readability; returns scores and per-check remediation.get_scan— Fetch a scan by id.ask— Natural-language NLWeb search over Agent Ready's methodology, checks and specs.
About MCP
The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.
Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.
Work with this as data
Every MCP server here is available over the APIs.io API and to AI agents over MCP.