We've been deep in the world of API tooling for years at liblab, primarily focusing on generating SDKs and documentation from OpenAPI specs. Recently, we kept running into a recurring frustration: connecting AI tools to existing APIs is way more complex than it should be.
You start with a simple goal – let an LLM talk to your API – and suddenly you're neck-deep in spinning up infrastructure, wrestling with authentication, and writing a ton of custom glue code just to translate natural language into structured API calls. Then you have to maintain it all.
To scratch our own itch, we built the MCP Generator.
It takes your API specification and generates a fully functional MCP (Model Context Protocol) server that's deployed to the cloud and ready to use in about 30 seconds. You can also download the server for free local usage. The idea is to cut out all the boilerplate and manual integration work. You just drop the generated URL into your AI tool, and you can start interacting with your API through natural language right away.
This builds on our existing SDK code generation engine, but extends it to support the MCP protocol. This means that as your API evolves, your MCP server can be regenerated to stay in sync without you needing to touch a thing.
We've seen some interesting early uses, and thought the HN community might have some cool ideas too:
- Building devtools that allow LLMs to perform actions against your internal or external APIs. - Querying internal metrics or services using natural language questions. - Surfacing structured documentation and content through conversational AI interfaces. - Speaking with your API service instead of coding functions.
We just launched and would genuinely love to get your feedback, thoughts, and even your most challenging edge cases. We're here to dive into the technical details and answer any questions.
You can try it out here: https://mcp.liblab.com/
Looking forward to hearing what you think!
