openai-docs-mcp
OpenAI runs one remote MCP server of its own and it is a DOCUMENTATION server, not an API-control server. `openai-docs-mcp` lets an agent search, browse and fetch the markdown behind platform.openai.com, developers.openai.com and learn.chatgpt.com, and — the part that matters for agent readiness — pull the OpenAPI fragment for any named OpenAI API endpoint via `get_openapi_spec`. No tool on it calls the OpenAI API: an agent cannot create a completion, upload a file, or manage a project through this server. Every tool is annotated readOnlyHint:true / destructiveHint:false, and `execution.taskSupport` is "forbidden" on all five, so the server itself declares it does no long-running work. It needs no key, which makes it one of the very few MCP surfaces in the catalog an agent can reach with zero configuration. DELIBERATELY NOT COUNTED HERE: OpenAI's much larger MCP presence is as a CONSUMER and a PLATFORM — the Responses API accepts remote MCP servers as tools, the Apps SDK is a framework for third parties to BUILD MCP servers, and the Codex CLI connects to external MCP servers via `codex mcp`. None of those is OpenAI running a server for its own API, and folding them in would credit the sales surface instead of the published one.
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.
Endpoint
Hosted endpoint · transport streamable-http
Connect
This is a remote MCP server — point an MCP client at the endpoint URL. Clients with native remote support (Claude, Cursor, VS Code, …):
For stdio-only clients (older Claude Desktop, etc.), bridge with mcp-remote:
If the server requires authentication, add the provider's token/header (e.g. an X-API-Key) per its docs. A quick reachability check:
Tools
search_openai_docs— Search across platform.openai.com, developers.openai.com and learn.chatgpt.com docs. Results include URLs.list_openai_docs— List or browse the crawled pages when the right query is not yet known.fetch_openai_doc— Fetch the markdown for a specific doc page so it can be quoted exactly; an `anchor` fetches just one section.list_api_endpoints— List all OpenAI API endpoint URLs available in the OpenAPI spec.get_openapi_spec— Return the OpenAPI spec for a specific API endpoint URL; optionally filter code samples by language or return only code samples.
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.