
MCP-OpenAPI
An MCP server that exposes HTTP methods defined in an OpenAPI specification as tools, enabling interaction with APIs via the Model Context Protocol.
README
MCP-OpenAPI
A Model Context Protocol server that exposes HTTP methods from an OpenAPI spec as tools.
[!WARNING] This is a proof of concept, yet to be fully tested and documented.
Usage
Each MCP-OpenAPI server exposes operations from a single OpenAPI spec. You could run multiple instances to expose multiple specs.
This package is available on PyPI @ [mcp-openapi
][pypi]. You can start a server with uvx mcp-openapi --openapi-url https://httpbin.org/spec.json (sse | stdio)
.
An example Claude config (while running fastapi dev tests/todos.py
with port 8000):
{
"mcpServers": {
"todos": {
"command": "uvx",
"args": [
"mcp-openapi",
"--openapi-url=http://localhost:8000/openapi.json",
"stdio"
]
}
}
}
The OpenAPI url can also be passed as an environment variable, MCP_OPENAPI_URL
.
When running as SSE, you can configure the server with:
--fastmcp-sse-host
- the host to serve the MCP server on--fastmcp-sse-port
- the port to serve the MCP server on
There are additional global options:
--fastmcp-debug
- enable debug mode for the MCP server--fastmcp-log-level
- the log level for the MCP server
These can also be configured via environment variables using the FASTMCP_
prefix, e.g. FASTMCP_LOG_LEVEL=DEBUG
.
How it works
- The MCP-OpenAPI server is initialised with an OpenAPI spec URL.
- The server fetches and parses the OpenAPI spec, and registers each path-operation as a tool.
- A FastMCP server is started with the registered tools.
- When a client requests a tool, the MCP server makes a HTTP request to the API and returns the response.
Supported OpenAPI/Swagger versions
Swagger 2.0 | OpenAPI 3.0 | OpenAPI 3.1 |
---|---|---|
:x: | :heavy_check_mark: | :heavy_check_mark: |
This package supports OpenAPI 3.0 and 3.1 specs, in JSON and YAML formats.
We're using openapi-parser under the hood, which seems to be pretty comprehensive, and supports resolving references to external specs.
Future configuration options
[!INFO] These are still to be implemented.
Restricting endpoints
By default, all endpoints are exposed as tools. You can restrict which endpoints are exposed:
- By path patterns,
- By HTTP method,
- Explicitly listing the individual operations to expose,
- Selecting routes using OpenAPI tags.
Handling API authentication
MCP-OpenAPI makes API requests to the target API - it can use a global auth token:
--auth-token
- a bearer token, or basic credentials in base64 format, used depending on the OpenAPI spec.
It would be nice to be able to make requests using varying authentication tokens (e.g. per client user), but without having the LLM see the tokens. TBD.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.