Search local knowledge bases with hybrid BM25, vector, and LLM reranking.
npx skills add https://github.com/AVOI-CEO/avoi-agent --skill qmd-avoi-ceo
Please help me install this Agent Skill. Skill: qmd Source: https://github.com/AVOI-CEO/avoi-agent/tree/main/optional-skills/research/qmd Command: npx skills add https://github.com/AVOI-CEO/avoi-agent --skill qmd-avoi-ceo
Local on-device search across personal knowledge bases, notes, docs, and transcripts is slow and siloed when relying on scattered tools; qmd provides a hybrid retrieval engine that runs locally, delivering fast, relevant results without cloud dependencies.
Install qmd and run a local search using qmd query after adding a collection.
High-intent search queries and answers about installing and using this skill.
You can search personal notes and documents locally by running a hybrid retrieval engine that combines BM25 keyword matching, vector search, and LLM reranking entirely offline on your device.
Hybrid retrieval for local knowledge bases works by combining BM25 keyword matching for exact terms, vector search for semantic similarity, and LLM-based reranking to deliver high-quality document results.
Yes, local semantic search operates offline via CLI on macOS and Linux, and optionally exposes MCP integration to support automation and tooling without requiring external cloud services.
The best way to locate meeting transcripts in local document collections is using a configurable pipeline that applies vector-based semantic search and LLM reranking to quickly return relevant transcript sections.
Yes, you can use this local search approach for large personal knowledge bases because it supports local document collections, embedding management, and a configurable pipeline designed for fast retrieval.
No, you do not need an internet connection because the local document search operates entirely offline, meaning the LLM-based reranking and embedding management run directly on your local machine.
zerwizSearch local knowledge bases with hybrid BM25 and vector retrieval.
s-hiraokuSearch the web and local businesses via Brave Search MCP.
OpenLAIRStore and search embeddings with a four-function API for a local vector database.
Monjyu1101Index and query local knowledge bases with hybrid BM25, vector, and LLM reranking.
