# tobi/qmd

mini cli search engine for your docs, knowledge bases, meeting notes, whatever. Tracking current sota approaches while being all local

Repository: https://github.com/tobi/qmd
Canonical: https://ross.abutalabs.com/products/qmd
Language: TypeScript
License: MIT
License Family: permissive
Last push: 2026-08-18T14:26:31+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 98, longevity 19
- inputs: {"age_days": 268, "days_push": 15, "days_rel": 17, "gap_med": 2, "n_releases_24m": 14}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 29259, forks 1823 (observed 2026-08-28T04:11:53.202985+00:00)

## What it is
QMD is a local, on-device CLI search engine that indexes markdown notes, docs, meeting transcripts, and knowledge bases. It combines BM25 full-text search, vector semantic search, and LLM re-ranking (via node-llama-cpp with GGUF models) with reciprocal rank fusion, and exposes an MCP server for AI agent workflows.

## Use cases
- search my local markdown notes from the terminal
- build a local search engine for my docs and meeting notes
- semantic search over my knowledge base without the cloud
- give my AI agent a tool to search my files
- hybrid keyword and vector search with reranking
- index meeting transcripts and query them in natural language

## When to choose
- you want fully local, private search over personal docs and notes
- you need JSON/file-list output designed for agentic and MCP workflows
- you want SOTA hybrid retrieval (BM25 + vectors + HyDE + LLM reranking) on-device

## When to avoid
- you need to index large shared team wikis with multi-user access
- you want a hosted or web-based search service with a GUI
- you lack hardware for local GGUF model inference

## Facets
- artifact type: cli-tool
- maturity: active
- function: search-engine, rag, llm-inference, cli, mcp
- domain: developer-tools, large-language-models
- platform: cli, bun, cross-platform, windows
- tags: local-search, bm25, vector-search, embeddings, hybrid-search, reranking, hyde, gguf, on-device, knowledge-base, agentic-workflows, markdown, search, retrieval-augmented-generation, command-line, nodejs, macos, linux

## Member repositories
- tobi/qmd (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:53.202985+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:53:04.270809+00:00, confidence not recorded.
  - readme: https://github.com/tobi/qmd (fetched 2026-08-28T04:11:53.202985+00:00, sha efb10e394406)
- Data as of 2026-08-30T08:39:29.467469+00:00.