An MCP server for the docs of any repo that uses Sphinx

Name: An MCP server for the docs of any repo that uses Sphinx
Availability: InStock
Author: btcalex

by btcalex·Mar 4, 2026·3 points·0 comments

Visit Project View on HN

AI Analysis

●●SolidNiche GemSolve My ProblemShip It

Faster doc search for AI agents than vanilla Sphinx, with RRF hybrid retrieval.

Strengths

•Tested on real large repos (pandas, celery, cpython) with measurable search quality gains.
•Hybrid mode uses RRF ranking to combine FTS5 + vector search intelligently, not just concatenate results.
•Zero external dependencies in FTS5 mode — pure SQLite, works offline.

Weaknesses

•Sphinx-only; doesn't handle other doc formats (mdBook, Hugo, Quarto) or non-text-based docs.
•Limited MCP ecosystem adoption — depends on Claude Desktop, Cursor, or manual agent integration uptake.

Post Description

It's a fairly simple stdio MCP server that provides AI agents a faster way to search through docs for any Sphinx-powered documentation. It builds Sphinx text docs and indexes them in SQLite (FTS5). There is also an optional hybrid search mode which creates embeddings and a vector db (sqlite-vec) and uses both approaches via RRF to get the best answer to your agent.

I've run this on several repos of varying size and complexity (pandas, celery, cpython) and have been impressed with the resulting answers.