open-source tool to expose your FastAPI endpoints as Model Context Protocol tools with zero config. Simple. Flexible. Production-ready.
@GithubProjects How would this prevent unauthorized clients from accessing the backend? would it be relegated only to backend calls that would otherwise be exposed to public traffic anyways?
@GithubProjects I've used this, and it was very easy. However, I wanted more control, and more explicit separation between new MCP tools and existing fast API endpoints. I ended up switching to the approach described here: medium.com/@yogeshwaribah…
@GithubProjects That's a beauty for everyone familiar with FastAPI already
@GithubProjects I need to start digging what this MCP is
@GithubProjects Check this with a practical use case: a fast API endpoint with DuckDB.
Very helpful, I already added a similar concept for Crawl4ai docker server, to gv people zero config MCP experience. I believe FastAPI will eventually supports this natively. This is the way if mcp going to be a standard. Meaning web framework, such as FastAPI, expressjs, … they should support this natively in their framework. But again, if MCP really becomes a thing. Thx for sharing.
@GithubProjects @grok What does this mean? How is this different from using FastAPI alone?
@GithubProjects Cool. I use fastapi every day this is huge for me
@GithubProjects 🐾 Oooh, wrapping FastAPI in MCP feels like giving pandas pre-peeled bamboo—delightfully efficient! 🎋
@GithubProjects This is the day. I wake up.
That looks like a useful tool! At jenova ai, we're also focused on simplifying AI development. One of our key features is letting users create custom AI apps using natural language, without needing any coding experience. This allows them to customize how their AI app interacts with search tools like web search, YouTube search, Reddit search, and GitHub search, enabling a broader range of tasks and customizations.