qmd

Index and search local markdown notes using BM25, vector search, and LLM reranking.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/THTProtocol/lastochka --skill qmd-thtprotocol
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: qmd
Source: https://github.com/THTProtocol/lastochka/tree/main/optional-skills/research/qmd
Command: npx skills add https://github.com/THTProtocol/lastochka --skill qmd-thtprotocol

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides seamless local search capabilities for personal knowledge bases, notes, and documentation, eliminating the need for cloud-based search services.

Core Features & Use Cases

  • Hybrid Search: Combines BM25, vector search, and LLM reranking for precise results.
  • Local Search Engine: Indexes and searches markdown notes, meeting transcripts, documentation, and text-based files.
  • CLI and MCP Integration: Offers command-line interface and Lastochka MCP integration for convenience.
  • Use Case: Easily find information across a large collection of markdown files without leaving your current workflow.

Quick Start

Add your note collections to qmd using the 'qmd collection add' command and start searching your notes with 'qmd search "keyword"'

Frequently Asked Questions about qmd

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

FAQPage Schema
How do I search my local markdown notes without a cloud service?

Local markdown notes search without a cloud service is achieved by indexing files on-device and querying them directly. This Skill uses BM25, vector search, and LLM reranking to find information across text-based collections without cloud dependencies.

What is the best way to find specific text across a large collection of markdown files?

The best way to find specific text across markdown files is using a hybrid search engine. It combines BM25 keyword matching, vector search, and LLM reranking to deliver precise results from your local knowledge base.

Does this local search engine support integration with command-line interfaces?

Yes, local search engine integration with command-line interfaces is fully supported. It provides a CLI for adding collections and searching notes, alongside Lastochka MCP integration to fit within your current workflow.

Can I use vector search and BM25 together for personal knowledge base retrieval?

Yes, you can use vector search and BM25 together for personal knowledge base retrieval. This Skill combines both methods with LLM reranking to provide highly precise hybrid search results for your local documents.

What file types are supported by on-device knowledge base search?

On-device knowledge base search supports text-based files including markdown notes, meeting transcripts, and documentation. It indexes these local files to provide efficient access without requiring cloud connectivity.

How do I add my note collections to start searching?

To add note collections and start searching, use the 'qmd collection add' command to index your files. Once indexed, use the 'qmd search "keyword"' command to query your local knowledge base directly from your workflow.