Search and extract content from documentation files at the section level across multiple formats (.md, .rst, .adoc, .ipynb, .html, .yaml, .json, OpenAPI specs). Built for developers who need to query structured docs programmatically without parsing boilerplate.
io.github.jgravelle/jdocmunch-mcp
Local install
STDIO
No auth required
How models use it and what it is built for.
Search and extract content from documentation files at the section level across multiple formats (.md, .rst, .adoc, .ipynb, .html, .yaml, .json, OpenAPI specs). Built for developers who need to query structured docs programmatically without parsing boilerplate.
Local install — runs as a subprocess.
Where to find authoritative docs and source for jDocmunch.
Paste any of these into Agent Studio after connecting jDocmunch.
Common questions about connecting and running jDocmunch.
What file formats does jdocmunch-mcp support?
It indexes and searches .md, .rst, .adoc, .ipynb, .html, .yaml, .json, and OpenAPI specification files. Queries return results at the section level, making it easy to locate specific content without parsing entire documents.
How do I install and run jdocmunch-mcp?
Install via `uvx jdocmunch-mcp@1.4.2`. The server runs over stdio, so it integrates directly with MCP-compatible clients. No additional configuration is required beyond pointing it at your documentation directory.
Can I search across multiple documentation formats at once?
Yes. jdocmunch-mcp treats all supported formats uniformly, so you can run a single query across mixed .md, .rst, .ipynb, and other files in the same search.
Is there a cost to use jdocmunch-mcp?
No. jdocmunch-mcp is open-source and free to use. It runs locally on your machine, so there are no API fees or usage limits.
What's the difference between section-level search and full-text search?
Section-level search returns results grouped by document sections (headings, cells, or logical blocks), making it easier to understand context. Full-text search would return individual lines or paragraphs, which can be noisy for large docs.
MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 30+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.
Open Agent Studio