A high-performance local search engine designed for indexing and retrieving Markdown notes and documents using keyword, vector, and hybrid search methods.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install qmd-skill-4
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install qmd-skill-4 using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is qmd?
qmd is a specialized tool developed to bring advanced search capabilities to your local Markdown-based knowledge bases and documentation sets. It provides a robust bridge between static text files and AI-powered retrieval, allowing users to index collections and perform rapid queries. Unlike cloud-based solutions, qmd runs entirely locally, ensuring data privacy and low-latency performance for your personal documentation.
By integrating Openclaw Skills like qmd into your environment, you enable developers and researchers to treat their local file system as a high-speed searchable database. It supports various search modes including traditional BM25 keyword matching and vector-based semantic similarity, making it an essential addition to any digital garden or technical documentation setup. The skill is optimized for speed, offering instant results for keyword searches while providing deeper semantic understanding when needed.
qmd Use Cases
- Searching through extensive personal notes or local knowledge bases for specific information.
- Finding related content across multiple Markdown documents based on semantic similarity.
- Retrieving specific technical documentation sections during intensive coding sessions.
- Automating local file indexing to provide context for AI agent interactions using Openclaw Skills.
- Managing and querying large collections of "messy" Markdown files without requiring strict structure.
How qmd Works
- Initialize a collection by pointing the tool to a local directory containing Markdown files using a glob mask.
- Run the indexer to build a BM25 keyword index which provides instant text retrieval for specific terms.
- Optionally generate embeddings for semantic search capabilities using local GGUF models processed on your machine.
- Execute search commands to find files by keyword, vector similarity, or hybrid ranking depending on the precision required.
- Retrieve full document content or specific file IDs to integrate the found data into your active workflow.
qmd Setup
To get started with this skill, ensure you have Bun >= 1.0.0 and SQLite extensions installed on your system. You can then install and configure your first collection.
bun install -g https://github.com/tobi/qmd
qmd collection add /path/to/notes --name notes --mask "**/*.md"
qmd embed
Ensure your PATH includes $HOME/.bun/bin to execute the commands from any directory.
qmd Data Schema & Taxonomy
The skill organizes information into named collections with specific metadata for efficient retrieval. The structure of the indexed data is managed as follows:
| Component | Description |
|---|---|
| Collection | A named logical group of indexed files (e.g., "notes" or "docs"). |
| Mask | The glob pattern used to include specific file types (e.g., "**/*.md"). |
| Index | A local SQLite-based store containing both BM25 text indices and vector embeddings. |
| Cache | Local directory (~/.cache/qmd/models/) used for storing GGUF models for local inference. |
qmd Advanced Features
- Hybrid search mode that combines BM25 keyword matching with LLM-based reranking for the highest quality results.
- Native JSON and file-only output formats designed for seamless integration with other Openclaw Skills and automation scripts.
- Incremental update capabilities that allow you to refresh the index for changed files without a full re-index.
- Multi-get functionality to retrieve content from several files simultaneously using a comma-separated list or wildcard patterns.
- Automated maintenance support via cron schedules to keep your local knowledge base search-ready at all times.
SKILL.md
Loading
Related Openclaw Skills
Clawdbot (Snibet Edition)
A high-performance tweet engine for builders that generates sharp, human-sounding X content without the AI fluff.
Weave
A powerful CLI tool for creating crypto invoices, generating cross-chain payment quotes, and monitoring settlement status.
TikTok Slideshow Creator
Generate high-converting TikTok image carousels with automated text overlays and seamless draft integration via the ViralBaby API.
Salesforce CLI Skill
A powerful integration that enables AI agents to query, manage, and automate Salesforce CRM data and metadata using the official Salesforce CLI.
QMD Memory Skill
A local, high-performance hybrid search and memory engine designed to eliminate API embedding costs for AI agents.
Hardcover
A GraphQL-based skill for AI agents to query and manage personal reading libraries and book metadata from Hardcover.app.
Featured*
