qmd

Index and search local markdown files with hybrid BM25 and vector retrieval.

2|1|Updated Jul 14, 2026

Search Tags:#rag #search #mcp #notes #knowledge-base #local-ai

Ready to run

npx skills add https://github.com/heysuhas/hermes_cli --skill qmd-heysuhas

💬 Prompt for ChatGPT / Claude Chat▼

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the challenge of fragmented personal knowledge by providing a high-performance, local-first search engine that indexes your notes, transcripts, and documentation without sending data to the cloud.

Core Features & Use Cases

Hybrid Search: Combines BM25 keyword matching, semantic vector search, and LLM-powered reranking for maximum accuracy.
MCP Integration: Seamlessly connects to the SR Agent as an MCP server to provide instant, context-aware retrieval.
Use Case: Quickly find specific decisions made in past meeting transcripts or locate technical documentation across large, local markdown repositories using natural language queries.

Quick Start

Use the qmd skill to perform a hybrid search for information regarding the database migration plan across all indexed collections.

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 files and transcripts without sending data to the cloud?▼

This skill enables local markdown search without sending data to the cloud by indexing files locally and executing search queries entirely on your machine. It ensures personal knowledge management remains private and offline.

What is hybrid search and how does it combine keyword matching with vector embeddings?▼

Hybrid search combines BM25 keyword matching with semantic vector embeddings and LLM reranking to maximize retrieval accuracy. It ensures both exact term matches and contextual meaning are used to locate relevant documents.

Do I need Node.js 22 and local GGUF models to run semantic search on my notes?▼

Yes, Node.js 22 and local GGUF model execution are required to perform semantic and keyword-based document discovery. These prerequisites enable the local retrieval engine to process your personal notes.

Can I integrate local knowledge base retrieval with an MCP server for automated agentic search?▼

Yes, this skill supports MCP server integration to provide automated agentic retrieval. It connects to the SR Agent as an MCP server to deliver instant, context-aware search results from your knowledge base.

What is the best way to find specific decisions in past meeting transcripts stored locally?▼

The best way to find decisions in past meeting transcripts is using a local hybrid search engine that indexes transcripts and supports natural language queries. It quickly locates specific information across large local repositories.