Skip to content
This repository was archived by the owner on Oct 27, 2026. It is now read-only.

v0.1.0 — Initial public release

Choose a tag to compare

@m3data m3data released this 28 Feb 08:28
· 67 commits to main since this release

Semantic retrieval MCP server for markdown-heavy project ecosystems. Indexes files, chunks them, embeds with OpenAI, and serves search over MCP.

Highlights

  • Config-driven: all project-specific values in sense.toml — server code is deployment-agnostic
  • Classification rules: ordered TOML rules map file paths to source types (trace, documentation, research, code, etc.)
  • Temporal decay: recent content scores higher, with configurable half-lives per source type
  • Mode-aware retrieval: integrates with Vibe Harness for context-sensitive search shaping (optional)
  • Claude Code skills: /sense and /sense-sync slash commands included
  • Auto-query hook: companion hook for ambient context injection on every prompt

Quick start

git clone https://github.com/m3data/sense-mcp.git
cd sense-mcp
python -m venv .venv && source .venv/bin/activate
pip install -e .
cp sense.example.toml sense.toml  # edit for your project

Requires Python 3.11+ and an OpenAI API key.

See README for full setup instructions.