generated: '2026-08-27' method: probed status: published source: >- https://developers.openai.com/mcp — probed live on 2026-08-27 with an MCP `initialize` handshake followed by `tools/list` over Streamable HTTP. Both calls succeeded ANONYMOUSLY (HTTP 200, content-type text/event-stream, no OAuth challenge). serverInfo returned {"name":"openai-docs-mcp","version":"1.0.0"}, protocolVersion 2025-06-18. The five tools below are the verbatim response, saved with their real inputSchemas to mcp/openai-docs-mcp-tools.json. Documented at https://developers.openai.com/learn/docs-mcp. description: >- 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. deployment: mode: remote endpoint: https://developers.openai.com/mcp auth: none verified: probed server: name: openai-docs-mcp version: 1.0.0 transport: streamable-http protocol_version: '2025-06-18' capabilities: tools: listChanged: true auth: none scope: documentation and OpenAPI retrieval only — no API control plane tools_file: mcp/openai-docs-mcp-tools.json tools: - name: search_openai_docs title: Search OpenAI Docs description: >- Search across platform.openai.com, developers.openai.com and learn.chatgpt.com docs. Results include URLs. input: {query: string (required), limit: integer 1-50, cursor: string} read_only: true - name: list_openai_docs title: List OpenAI Docs description: List or browse the crawled pages when the right query is not yet known. input: {limit: integer 1-50, cursor: string} read_only: true - name: fetch_openai_doc title: Fetch OpenAI Doc description: >- Fetch the markdown for a specific doc page so it can be quoted exactly; an `anchor` fetches just one section. input: {url: string (required), anchor: string} read_only: true - name: list_api_endpoints title: List API Endpoints description: List all OpenAI API endpoint URLs available in the OpenAPI spec. input: {} read_only: true - name: get_openapi_spec title: Get OpenAPI Spec description: >- Return the OpenAPI spec for a specific API endpoint URL; optionally filter code samples by language or return only code samples. input: {url: string (required), languages: array, codeExamplesOnly: boolean} read_only: true related_but_not_this_server: - name: Remote MCP servers as Responses API tools url: https://developers.openai.com/api/docs/guides/tools-connectors-mcp relationship: consumer note: OpenAI models CALL third-party MCP servers; this is not a server OpenAI publishes. - name: Apps SDK url: https://developers.openai.com/apps-sdk relationship: platform note: >- A framework for third parties to build MCP servers that render UI inside ChatGPT. Component library catalogued in components/openai-components.yml. - name: codex mcp url: https://learn.chatgpt.com/docs/codex/cli relationship: client note: The Codex CLI connects to external MCP servers. Catalogued in cli/openai-cli.yml. probe: - url: https://developers.openai.com/mcp method: POST body: '{"jsonrpc":"2.0","id":1,"method":"initialize",...}' status: 200 content_type: text/event-stream result: 'serverInfo openai-docs-mcp 1.0.0; capabilities.tools.listChanged true' - url: https://developers.openai.com/mcp method: POST body: '{"jsonrpc":"2.0","id":2,"method":"tools/list"}' status: 200 content_type: text/event-stream result: 5 tools with full inputSchema