fastapi_mcp trends on GitHub: exposes FastAPI endpoints directly as MCP tools
Decision Brief
Rather than a generic OpenAPI-to-MCP converter, fastapi_mcp works as a native FastAPI extension: initialize with FastApiMCP(app) and call mount() to generate an MCP server at the /mcp path, preserving request/response model schemas and Swagger docs. Communication goes over FastAPI's ASGI interface, so no extra HTTP calls are needed between MCP and endpoints; the MCP service can be mounted on the original app or deployed separately. Auth reuses FastAPI's Depends(). Install via uv add fastapi-mcp or pip install fastapi-mcp; requires Python 3.10+ (3.12 recommended), MIT licensed. For backend developers wiring internal services into LLM toolchains, this means no separate MCP description and auth logic per endpoint—your existing FastAPI dependency system and docs carry over. For teams on ASGI deployments wanting fewer internal HTTP round trips, direct ASGI communication also cuts latency.
Sources
- Skill Radar(GitHub 趋势)
Trending hands-on MCP servers, agent skills, and AI-coding tools discovered from GitHub search momentum.
- GitHub:tadata-org/fastapi_mcp
留言
登入后即可留言,和其他 builder 交换实测心得。
还没有留言,抢头香。