nicholasglazer/gnosis-mcp

v0.6.0 Breaking

This release includes breaking changes for platform teams planning a safe upgrade.

Published 3mo MCP Data & Storage

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

ai claude developer-tools documentation knowledge-base llm

+10 more

mcp mcp-server model-context-protocol pgvector postgresql python search self-hosted sqlite vector-db

Summary

AI summary

SQLite backend is now the default, enabling zero‑config search out of the box.

Full changelog

Zero-config SQLite default

Gnosis MCP now works out of the box with SQLite — no database server needed.

pip install gnosis-mcp
gnosis-mcp ingest ./docs/
gnosis-mcp serve

Two commands from install to working search.

What's new

SQLite backend with FTS5 full-text search (porter tokenizer, BM25 scoring)
Auto-detection: no DATABASE_URL → SQLite, postgresql://... → PostgreSQL
Auto-init: ingest creates the database automatically, no init-db step needed
Backend Protocol abstraction — clean separation between SQLite and PostgreSQL
Claude Code plugin — marketplace.json + .mcp.json for one-command install
176 tests passing without any database

Packaging

Default: pip install gnosis-mcp (SQLite, zero config)
PostgreSQL: pip install gnosis-mcp[postgres] (adds asyncpg)
XDG-compliant default path: ~/.local/share/gnosis-mcp/docs.db

Breaking changes

None. Existing PostgreSQL users with GNOSIS_MCP_DATABASE_URL set will see no change.

Full changelog: https://github.com/nicholasglazer/gnosis-mcp/compare/v0.5.0...v0.6.0

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track nicholasglazer/gnosis-mcp

Get notified when new releases ship.

About nicholasglazer/gnosis-mcp

Zero-config MCP server for searchable documentation. Loads markdown into SQLite (default) or PostgreSQL with FTS5/tsvector keyword search and optional pgvector hybrid semantic search.

All releases →