jDocmunch MCP

Our take

Section-level indexing for documentation — Markdown, RST, AsciiDoc, notebooks, OpenAPI — so agents pull the one section they need instead of eating whole files; part of the author's sprawling 'munch' family of token-efficiency tools. The self-description ('the leading, most token-efficient') is marketing-forward, but the repo has real architecture docs and benchmarks, and no recommender-steering anywhere we looked. Odd metrics worth noting: 184 stars with zero watchers, and the registry pins 1.4.2 while PyPI has churned to 1.92.0 across 126 releases — install fresh from PyPI.

reviewed by hand · 2026-07-06

Something wrong or dead here? Report it

Verification record

last verified

5 days ago

github stars

202

last commit

6 days ago

archived

no

license

NOASSERTION

in registry since

2026-03-21

github.com/jgravelle/jdocmunch-mcp