Search local knowledge bases with BM25, vector search, and LLM reranking.
npx skills add https://github.com/DaddyElonMusk69/motis-agent --skill qmd-daddyelonmusk69
Please help me install this Agent Skill. Skill: qmd Source: https://github.com/DaddyElonMusk69/motis-agent/tree/main/optional-skills/research/qmd Command: npx skills add https://github.com/DaddyElonMusk69/motis-agent --skill qmd-daddyelonmusk69
Search across personal knowledge bases locally using a hybrid retrieval engine that combines keyword matching, vector semantics, and LLM-based reranking to deliver high-quality results without cloud dependencies.
Install qmd, add a collection, then run a search to retrieve relevant documents.
High-intent search queries and answers about installing and using this skill.
Local knowledge search allows you to query personal notes and documents offline. You can index markdown and text collections on-device, retrieving relevant information without sending data to the cloud.
Hybrid retrieval improves document relevance by combining BM25 keyword matching with vector semantic search. This dual approach captures both exact term matches and contextual meaning, which is then refined by LLM-based reranking.
Yes, you can use local vector search for offline indexing of markdown collections. The system indexes your notes, documents, and transcripts locally, supporting fast offline searches directly on your device.
The best way to integrate local knowledge search with an MCP workflow is using the built-in MCP integration. This allows your MCP-connected clients to query your locally indexed knowledge base directly through the hybrid retrieval engine.
No internet connection is needed for core offline indexing and searching. The system operates locally for privacy-minded workflows, applying BM25 and vector search on-device to deliver high-quality results without cloud dependencies.
A limitation of this CLI-based hybrid retrieval engine is that it requires manual terminal commands for management. It is designed for local text and markdown collections, meaning it may not parse proprietary file formats without prior conversion.
Rawgrowth-ConsultingSearch local Markdown knowledge bases with hybrid BM25 and vector retrieval.
cpmappstudioSearch past conversations to surface relevant decisions and patterns.
eddieluengSearch local markdown and text collections with hybrid retrieval and reranking.
attentiondotnetPerform semantic search over large document collections using AgentDB.
