qmd

Index and search local text collections with hybrid retrieval.

Updated Apr 9, 2026
One-click install
npx skills add https://github.com/nadicodeai/argo-agent --skill qmd-nadicodeai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: qmd
Source: https://github.com/nadicodeai/argo-agent/tree/main/optional-skills/research/qmd
Command: npx skills add https://github.com/nadicodeai/argo-agent --skill qmd-nadicodeai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Search across notes, documents, and transcripts is slow, fragmented, and relies on cloud-based services.

Core Features & Use Cases

  • Hybrid retrieval combines BM25 keyword search, vector semantic search, and LLM reranking to deliver relevant results across markdown and text files.
  • Local-first indexing enables offline use and privacy, with optional MCP integration for automation.
  • Use cases include quickly finding meeting decisions, researching topics across large note collections, and locating code or documentation snippets.

Quick Start

Install qmd and index your first collection to begin fast, local search across your 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 notes and transcripts locally without cloud services?

You can search notes and transcripts locally by using a local-first indexing tool that operates offline. This Skill indexes your text collections on your machine, enabling fast retrieval without relying on cloud-based services.

How does hybrid retrieval combine keyword and semantic search for documents?

Hybrid retrieval improves document search by combining BM25 keyword matching with vector-based semantic search. It then applies LLM reranking to these combined results, ensuring higher relevance across your markdown and text files.

Can I index markdown files for offline vector search?

Yes, you can index markdown and text files for offline vector search. The Skill performs local-first indexing, allowing you to run semantic searches on your personal knowledge base completely offline.

What is the best way to find meeting decisions across large note collections?

The best way to find meeting decisions is by indexing your note collection for hybrid retrieval. By applying BM25 keyword search, vector semantic search, and LLM reranking, you can quickly locate specific decisions across large document sets.

Does local knowledge base search support MCP integration for automation?

Yes, local knowledge base search supports modular MCP integration. This allows you to automate retrieval tasks and access your indexed notes, transcripts, and documents directly through CLI commands.

Why use LLM reranking with BM25 and vector search?

Using LLM reranking with BM25 and vector search significantly improves result relevance. While keyword and semantic search retrieve initial matches, the reranking step reorders them to better align with your actual query intent.