A local search engine that provides instant keyword and semantic search capabilities for Markdown-based knowledge bases.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install qmd-markdown-search
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-markdown-search 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: Local Markdown Search?
qmd is a high-performance CLI tool designed to index and search local Markdown notes, documentation, and personal knowledge bases. It acts as a bridge between your static files and your AI-driven workflows, making it a vital part of the Openclaw Skills ecosystem. By utilizing a combination of BM25 keyword matching and vector-based semantic search, it allows for both instant retrieval and deep contextual discovery.
Unlike traditional search tools, qmd is optimized for the way developers and knowledge workers organize information. It handles "messy" Markdown gracefully by using content-based chunking rather than strict structural rules. This ensures that when you use Openclaw Skills to interact with your data, the results are relevant, fast, and processed entirely on your local machine for maximum privacy.
qmd: Local Markdown Search Use Cases
- Searching personal knowledge bases or "second brains" for specific insights.
- Retrieving technical documentation from local repositories during coding sessions.
- Finding semantically related notes when keyword matches are insufficient.
- Providing AI agents with a high-speed interface to access disk-based documentation through Openclaw Skills.
How qmd: Local Markdown Search Works
- The user defines a collection by pointing qmd to a directory of Markdown files and applying a file mask.
- qmd performs an initial indexing pass using BM25, enabling instant keyword search functionality.
- An optional embedding step generates vector representations of the content using local LLM models (e.g., Qwen).
- Users or AI agents query the index using simple search, vector search, or hybrid reranking modes.
- The tool returns document IDs, snippets, or full file contents based on the query parameters.
qmd: Local Markdown Search Setup
To get started, ensure you have Bun and SQLite installed on your machine. Then, install the skill globally:
bun install -g https://github.com/tobi/qmd
Initialize your local environment and index your first collection to enable Openclaw Skills functionality:
qmd collection add /path/to/notes --name notes --mask "**/*.md"
qmd embed
qmd: Local Markdown Search Data Schema & Taxonomy
qmd organizes data into logical collections with flexible metadata tagging to optimize retrieval performance for Openclaw Skills.
| Component | Description |
|---|---|
| Collections | Named groups of files defined by local paths and glob patterns. |
| BM25 Index | A keyword-based index used for the fastest retrieval results. |
| Vector Store | Local embeddings generated via GGUF models for semantic similarity. |
| Cache | Local model storage typically found in ~/.cache/qmd/models/. |
qmd: Local Markdown Search Advanced Features
- Scheduled indexing via cron to keep search results synchronized with file changes.
- Hybrid search mode which combines keyword matching with LLM-based reranking for maximum precision.
- Agent-friendly output formats including JSON and file-path-only lists for programmatic consumption.
- Local LLM integration for query expansion, ensuring high-quality semantic matches without cloud dependencies.
- Multi-document retrieval (multi-get) for gathering context across several files simultaneously.
SKILL.md
Loading
Related Openclaw Skills
Vagus: Phone MCP Connection
A bridge connecting AI agents to Android devices via the Model Context Protocol for real-time sensor data and physical device interactions.
Jasper ConfigGuard
A robust CLI tool for safely updating OpenClaw configurations with automated rollbacks and health monitoring.
hopeIDS Security Skill
An inference-based intrusion detection system for AI agents that provides real-time scanning, automated quarantine, and human-in-the-loop security management.
exo-installer
The centralized command-line manager for installing, updating, and health-checking the entire E.x.O. suite of AI agent tools.
CardPointers CLI
A command-line interface and AI agent skill to query credit card recommendations, wallet cards, and rewards offers directly from CardPointers.
OpenKrill: x402 Micropayments
A powerful skill for AI agents to autonomously handle x402 protocol micropayments and wallet management.
Featured*
