Index and search local knowledge bases with hybrid BM25, vector, and LLM reranking.
npx skills add https://github.com/helix4u/hermes-agent --skill qmd-helix4u
Please help me install this Agent Skill. Skill: qmd Source: https://github.com/helix4u/hermes-agent/tree/main/optional-skills/research/qmd Command: npx skills add https://github.com/helix4u/hermes-agent --skill qmd-helix4u
Locates and retrieves information from personal knowledge bases, notes, and transcripts stored locally, combining multiple retrieval strategies to deliver relevant results.
Install and initialize qmd, then add your collections and run your first search.
High-intent search queries and answers about installing and using this skill.
You can search local notes and docs offline using a hybrid retrieval system that combines BM25, vector search, and LLM reranking to deliver relevant results while keeping your data private.
Hybrid retrieval for local knowledge bases combines BM25 keyword matching, vector search, and LLM reranking to index and locate relevant information across markdown notes, documents, and meeting transcripts.
Yes, you can search meeting transcripts locally by indexing them in configured collections and optionally exposing the search tools to Hermes Agent through an MCP server setup.
Yes, a local Node.js environment is required to initialize the skill, configure your document collections, and execute offline vector search and BM25 queries.
You can set up multiple collections by installing and initializing the search tool, then configuring your local document groups in the settings to run targeted queries across them.
Offline BM25 and vector search operates entirely locally and requires a Node.js environment, meaning processing scale is limited by your hardware and you must manually configure collections for queries.
ChimeraFoundationaIndex local markdown notes, transcripts, and docs for hybrid on-device search.
jacardlIndex and search local knowledge bases with hybrid retrieval.
Yangel-hideSearch local knowledge stores with hybrid BM25, vector, and LLM reranking.
joeleung0122Index local files and search them with BM25-based ranking.
