skills.rest
jleechanorg
jleechanorgOFFICIAL
Explore all skills by this author →
GitHub

qmd

Index and search local knowledge bases with hybrid retrieval.

Updated Jun 28, 2026
Search Tags:#markdown#search#mcp#local#knowledge-base#vector-search#hybrid-search
One-click install
npx skills add https://github.com/jleechanorg/hermes-agent --skill qmd-jleechanorg
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: qmd
Source: https://github.com/jleechanorg/hermes-agent/tree/main/optional-skills/research/qmd
Command: npx skills add https://github.com/jleechanorg/hermes-agent --skill qmd-jleechanorg

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Local, on-device search for personal knowledge bases that keeps data private while enabling fast, accurate retrieval across notes, transcripts, and documentation.

Core Features & Use Cases

  • Hybrid search that combines keyword matching, semantic understanding, and LLM-powered reranking, all running locally.
  • Indexes Markdown notes, meeting transcripts, and documents for cross-collection queries.
  • MCP integration to expose Hermes Agent-friendly tools without needing cloud services.

Quick Start

Add a directory as a collection and run a hybrid search to query your notes.

Frequently Asked Questions about qmd

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search markdown notes and transcripts locally without sending data to the cloud?▼

You can index and search personal knowledge bases locally using a hybrid retrieval pipeline. This on-device search supports markdown notes, transcripts, and documentation while keeping your data completely private without needing cloud services.

What is hybrid retrieval for knowledge bases and how does it improve search results?▼

Hybrid retrieval combines keyword matching, semantic understanding, and LLM-powered reranking into a single search pipeline. This approach improves accuracy by merging exact text matching with contextual vector search across your indexed collections.

Do I need Node.js to run local vector search on my markdown collections?▼

Yes, local vector search requires Node.js version 22 or higher. You also need to download local models and install the qmd CLI to index collections and execute hybrid retrieval queries on your device.

Can I integrate local knowledge search with an MCP server for agent access?▼

Yes, the tool supports optional MCP integration to expose Hermes Agent-friendly tools. This allows direct agent access to your local collections without relying on external cloud services for retrieval.

What's the best way to query multiple document collections entirely on my device?▼

Add directories as collections and run a hybrid search to query across them. The local pipeline indexes your documents and supports cross-collection searches combining keyword, semantic, and reranked results.

Can I use this local search tool without downloading large language models?▼

No, downloading local models is a prerequisite for the hybrid retrieval pipeline. These models are required to perform the semantic understanding and LLM-powered reranking for your search queries entirely on-device.

Related Skills

Explore More Content & Communication Skills→
afel6afel6

qmd

Search local markdown and meeting transcripts with hybrid BM25, vector, and reranking retrieval.

Community
Intermediate
anilcan-karaanilcan-kara

qmd

Index and search local knowledge bases with hybrid retrieval via CLI and MCP.

Community
Intermediate
eastreamseastreams
641

native-mcp

Connect to external MCP servers and expose their tools as native Hermes Agent capabilities.

Official
Advanced
Chris-Chai-MinjaeChris-Chai-Minjae

qmd

Index local text files for hybrid keyword and semantic search.

Community
Advanced