MCP endpoint
Also known as Model Context Protocol endpoint · MCP server
A discoverable Model Context Protocol server URL that lets AI agents call your tools, resources, and data through a standardized interface.
What it is
The Model Context Protocol (MCP) is an open standard that defines how AI applications connect to external tools, resources, and prompts. An MCP endpoint is the network address of an MCP server, typically served over HTTP with Streamable HTTP transport, that exposes those capabilities to agents. Advertising it via a well-known path or response header lets clients discover and connect to it.
Why it matters
As AI shifts from reading pages to taking actions, an MCP endpoint turns your site from a document into a callable service that agents like Claude or ChatGPT can use directly. Exposing safe, well-scoped capabilities lets agents check availability, fetch live data, or transact on your behalf, making your business present in agentic workflows rather than only in search results.
How to verify
Confirm the server URL responds to MCP initialization, for example by attempting a connection with an MCP client or Inspector and checking that tools/list and resources/list return. If you advertise it via a well-known path or header, curl that path and verify it points to a reachable HTTPS endpoint with valid auth and CORS configuration.
How to fix
Build an MCP server exposing a minimal, safe set of tools and resources, deploy it over HTTPS with the Streamable HTTP transport, and add authentication and rate limiting. Then advertise its URL where agents and developers can find it, such as documentation, a well-known path, or your brand-facts file, and keep the tool schemas accurate.
Related terms
- brand-facts.json / ai-agent.jsonA machine-readable JSON file of authoritative brand facts published at a stable, well-known path for AI agents to consume.
- AI-crawler policyExplicit robots.txt rules for AI user agents (GPTBot, ClaudeBot, PerplexityBot, Google-Extended) that deliberately allow or block them.
- Agent-readiness score (isitagentready)An optional reference number from Cloudflare's isitagentready.com that rates, roughly 0-100, how easily AI agents can access and use a site.
- llms.txtA Markdown file at /llms.txt that gives LLMs a curated, plain-language map of your site and links to its most important content.
Official references
External, opens in a new tab.
Put this into practice.
Work through every check by hand and turn it into a shareable GEO Score report — or scan your site automatically in seconds.