About Mdkit
mdkit converts documents — PDF, Office files, HTML — into clean, LLM-ready Markdown and JSON over a REST API and an MCP server, so you can feed real-world documents straight into a model or a RAG pipeline without hand-cleaning them.
Larger jobs run through an async job queue: submit a document, poll for the result, or get a webhook when it's done.
It's a small, self-hosted service run by a solo operator, with a free tier to try it and metered credits for volume.