Search local knowledge bases with hybrid BM25 and vector retrieval.
npx skills add https://github.com/Plaidmustache/hermes-nulab --skill qmd-plaidmustache
Please help me install this Agent Skill. Skill: qmd Source: https://github.com/Plaidmustache/hermes-nulab/tree/main/optional-skills/research/qmd Command: npx skills add https://github.com/Plaidmustache/hermes-nulab --skill qmd-plaidmustache
Local, on-device search of your personal knowledge bases, notes, docs, and meeting transcripts to keep data private and quickly retrievable.
Install qmd locally, index your collections with qmd embed, and start querying with qmd search or qmd query.
High-intent search queries and answers about installing and using this skill.
Local note and document search uses a hybrid retrieval pipeline combining BM25 keyword search and vector embeddings with LLM reranking. This on-device indexing keeps data private and quickly retrievable without cloud dependencies.
Hybrid retrieval combines BM25 keyword search with vector search and LLM reranking. This mechanism improves result quality by matching exact terms while understanding semantic context across notes and meeting transcripts.
To index local document collections, install the tool locally and run the embed command on your files. After indexing, query the collection using search or query commands within CLI and MCP-integrated workflows.
Local vector search works for team knowledge bases and project documentation through on-device indexing. It supports MCP-integrated workflows for seamless automation, making it suitable for personal research and team environments.
The best way to search meeting transcripts locally is using a hybrid retrieval approach. It combines BM25 and vector search with LLM reranking to provide high-quality results while keeping transcripts private on your device.
ChimeraFoundationaIndex local markdown notes, transcripts, and docs for hybrid on-device search.
lightningfastslsSearches trusted sources and assembles topic findings into a structured knowledge graph with provenance tracking.
nelohenriqIndex local markdown notes and transcripts for hybrid on-device search via MCP.
jonnabioIndex and retrieve documents using semantic embeddings and metadata.
