qmd

Index and query markdown notes with BM25, vector search, and LLM reranking.

Updated May 8, 2026

Search Tags:#markdown #search #knowledge base #local AI #QMD

Ready to run

npx skills add https://github.com/superfhp/lumi-agent --skill qmd-superfhp

💬 Prompt for ChatGPT / Claude Chat▼

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires node, sqlite, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides local, on-device search for personal knowledge bases, notes, and documents, eliminating the need for cloud dependencies and improving privacy.

Core Features & Use Cases

Hybrid Search: Combines keyword matching, semantic understanding, and LLM reranking for comprehensive search.
Local Search: Operates entirely on local devices, ensuring privacy and reducing latency.
Integration: Integrates with Hermes Agent via MCP for seamless usage within the agent's environment.

Quick Start

Use the qmd skill to search for "database migration" in your personal 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 locally without cloud dependencies?▼

To search markdown notes locally without cloud dependencies, you can use a local search engine that indexes documents entirely on your device. This approach ensures privacy and reduces latency by eliminating external network calls.

What is hybrid search for personal knowledge bases?▼

Hybrid search for personal knowledge bases combines keyword matching, semantic understanding, and LLM reranking. This mechanism provides comprehensive search results by leveraging both exact term retrieval and contextual document analysis.

Do I need Node.js and SQLite to run local markdown search?▼

Yes, you need Node.js and SQLite with extension support to run this local markdown search. These dependencies provide the runtime environment and database layer required for indexing documents and executing queries.

How do I index and query markdown transcripts on my device?▼

You index and query markdown transcripts by running a local search engine that processes your notes, documents, and transcripts. Once indexed, you can execute queries using BM25, vector search, and LLM reranking for accurate retrieval.

Does local AI knowledge retrieval work with Hermes Agent?▼

Yes, local AI knowledge retrieval integrates with Hermes Agent via MCP. This integration allows you to use the search functionality seamlessly within the agent's environment for querying your notes.