skills.rest
hochoa13
hochoa13COMMUNITY
Explore all skills by this author →
GitHub

qmd

Search local knowledge bases with hybrid BM25 and vector retrieval.

Updated Mar 13, 2026
Search Tags:#rag#search#knowledge base#notes#document search#local ai
One-click install
npx skills add https://github.com/hochoa13/Asistente-Harlest --skill qmd-hochoa13
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: qmd
Source: https://github.com/hochoa13/Asistente-Harlest/tree/main/optional-skills/research/qmd
Command: npx skills add https://github.com/hochoa13/Asistente-Harlest --skill qmd-hochoa13

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a powerful, local-first search engine for your personal knowledge base, notes, and documents, enabling fast and intelligent retrieval of information without relying on cloud services.

Core Features & Use Cases

  • Hybrid Search: Combines keyword (BM25), semantic vector search, and LLM reranking for highly accurate results.
  • Local Operation: All models and data are stored and processed on your machine, ensuring privacy and offline capability.
  • Use Case: Quickly find specific notes about a past project, locate meeting transcripts discussing a particular topic, or search through documentation for a relevant code snippet using natural language queries.

Quick Start

Search your notes for information about the latest project decisions.

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 knowledge base and notes without sending data to the cloud?▼

Local search combines keyword matching, semantic vector search, and LLM-powered reranking to find precise information across personal notes and documents. This hybrid approach retrieves highly relevant results for natural language queries without relying on cloud services.

Can I use local AI search with my existing CLI and AI agents?▼

Yes, this local search integrates with CLI and MCP for seamless use with AI agents. This allows your AI assistants to directly query your personal notes and documents through the on-device hybrid retrieval engine.

What is the best way to find specific meeting transcripts or project notes using natural language?▼

Using a local-first search engine that supports natural language queries is the best way to find meeting transcripts and project notes. It combines BM25 keyword matching with semantic vector search to locate specific information across your documents.

Do I need an internet connection to run semantic vector search on my documents?▼

No, an internet connection is not required because all models and data are stored and processed locally on your machine. This local-first operation ensures full offline capability for searching your knowledge base.

How does LLM-powered reranking improve document search results?▼

LLM-powered reranking refines document search results by reordering outputs from initial keyword and semantic vector retrieval. This hybrid retrieval method ensures the final results are highly accurate and comprehensive for your knowledge base queries.

Related Skills

Explore More Software Engineering Skills→
dy9759dy9759

content-digest

Convert URLs, documents, images, and text into structured notes saved to WPS Notes.

Community
Advanced
RayLee1997RayLee1997
6

web-research

Search the web and news via Brave Search, evaluate sources, and synthesize cited answers.

Community
Advanced
InverterNetworkInverterNetwork

pinecone-research

Store and retrieve conversation context using Pinecone vector databases.

Official
Intermediate
adeboldadebold

AgentDB Vector Search

Perform vector-based semantic search against an AgentDB index.

Community
Advanced