qmd

Index and query local knowledge bases with hybrid BM25, vector, and LLM reranking.

1|Updated Jan 31, 2026
One-click install
npx skills add https://github.com/Monjyu1101/AiDiy2026 --skill qmd-monjyu1101
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: qmd
Source: https://github.com/Monjyu1101/AiDiy2026/tree/main/backend_hermes/optional-skills/research/qmd
Command: npx skills add https://github.com/Monjyu1101/AiDiy2026 --skill qmd-monjyu1101

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Index and query local knowledge bases across notes, transcripts, and documentation using a hybrid retrieval engine that blends BM25, vector search, and LLM reranking for high-quality results.

Core Features & Use Cases

  • Hybrid retrieval that combines BM25, vector search, and LLM reranking for high-quality results.
  • Local, on-device search across markdown/text collections including meeting transcripts and documentation.
  • MCP integration and CLI access for Hermes agents to query personal knowledge bases.

Quick Start

Query your local knowledge base with qmd to retrieve relevant notes and transcripts.

Frequently Asked Questions about qmd

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

FAQPage Schema
How do I search local markdown notes and transcripts offline?

You can search local markdown notes and transcripts offline using a hybrid retrieval engine that blends BM25 and vector search. This approach operates entirely on-device without cloud dependencies, ensuring private and offline knowledge retrieval.

What is hybrid retrieval and how does it improve local knowledge search?

Hybrid retrieval improves local knowledge search by combining BM25 keyword matching, vector search, and LLM reranking. This blended approach enhances result quality by leveraging both exact term frequency and semantic context from your personal knowledge base.

Can I use local knowledge search with MCP integration on macOS and Linux?

Yes, local knowledge search supports MCP integration and CLI access on both macOS and Linux environments. This allows Hermes agents to query your personal documentation and meeting transcripts directly through MCP exposure.

What's the best way to query personal documentation without cloud dependencies?

The best way to query personal documentation without cloud dependencies is using an edge-first local search tool. It indexes your text collections on-device, utilizing LLM reranking and vector search to deliver high-quality retrieval results securely.

Does hybrid retrieval support indexing meeting transcripts and text collections?

Yes, hybrid retrieval supports indexing meeting transcripts and text collections locally. It processes markdown and text files across your local knowledge base, applying BM25 and vector search to retrieve relevant transcript sections accurately.

Why use an offline knowledge base instead of cloud search for personal notes?

You use an offline knowledge base to ensure data privacy and eliminate cloud dependencies when searching personal notes. Local indexing provides edge-first retrieval, allowing you to query transcripts and documentation securely without external network calls.