qmd

Index and search local markdown notes with hybrid retrieval.

2|Updated Apr 25, 2026

Search Tags:#markdown #rag #search #mcp #retrieval #local #knowledge-base

Ready to run

npx skills add https://github.com/AlexiosBluffMara/mercury --skill qmd-alexiosbluffmara

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: qmd
Source: https://github.com/AlexiosBluffMara/mercury/tree/main/optional-skills/research/qmd
Command: npx skills add https://github.com/AlexiosBluffMara/mercury --skill qmd-alexiosbluffmara

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Local knowledge is dispersed across notes, docs, and transcripts, making retrieval slow and manual.

Core Features & Use Cases

Hybrid search combines BM25, vector embeddings, and LLM reranking to surface relevant results from local collections.
MCP integration enables Hermes Agent to use qmd tools without repeated skill loading.
Local-first indexing supports personal or team knowledge bases with offline operation.

Quick Start

Install qmd globally and add a collection to start searching.

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 with hybrid retrieval?▼

You can search local markdown notes and transcripts using hybrid retrieval by indexing them locally with qmd, which combines BM25, vector search, and LLM reranking to surface relevant results from your collections.

How does hybrid retrieval combine BM25 and vector search for local knowledge bases?▼

Hybrid retrieval for local knowledge bases combines BM25 keyword matching, vector embeddings, and LLM reranking to process and surface relevant documents from local collections.

Do I need Node.js 22 and SQLite to run local knowledge base search?▼

Yes, local knowledge base search requires Node.js 22+ and SQLite with extensions on macOS when needed, alongside local model downloads to function properly.

Can I use MCP integration to search local documents with an agent?▼

Yes, MCP integration enables agents like Hermes to use local document search tools directly without repeated skill loading, streamlining retrieval across collections.

What is the best way to index personal documents for offline search?▼

The best way to index personal documents for offline search is using a local-first indexing tool that supports offline operation across personal or team knowledge bases.

What are the limitations of offline search for local markdown collections?▼

Limitations of offline search for local markdown collections include requiring local model downloads and specific environment setups like SQLite extensions on macOS to function.