QMD
QMD is a MCP server for web research and discovery, published by tobi.
Hybrid search engine combining BM25, vector embeddings, and LLM re-ranking for local document retrieval in agentic workflows.
Is it any good?
Effectiveness · benchmark pending
We haven't benchmarked QMD yet, so its score is provisional — computed from safety, maintenance, and documentation only, with popularity given no weight. When we benchmark it, we measure whether it produces materially better output than a frontier model alone, and that result leads the score. How we measure.
About QMD
Description · AI-summarised from the tobi README
QMD is an on-device search engine that indexes markdown notes, meeting transcripts, documentation, and knowledge bases. It supports keyword search (BM25), semantic search (via embeddings), and hybrid search with LLM re-ranking—all running locally via node-llama-cpp with GGUF models. Ships with CLI tools, an MCP (Model Context Protocol) server for Claude and other agents, and a Node.js/Bun SDK for embedding in custom applications. Results include context metadata to help LLMs make better document selection decisions.
Where it fits
Browse related skills
Install QMD
Install
Install via MCP server
{
"mcpServers": {
"tobi-qmd": {
"command": "npx",
"args": ["-y", "@tobi/tobi-qmd"]
}
}
}Install with your AI
Zero install · paste into ChatGPT, Claude, Cursor…
Install the "QMD" skill into my project by following all instructions at https://skillsdirectory.co/s/tobi-qmd
Your assistant reads /s/tobi-qmd and follows it — fetching the files from the source, placing them, and confirming before it runs anything. No account needed. How this works.
Frequently asked
FAQ
What does QMD do?
Hybrid search engine combining BM25, vector embeddings, and LLM re-ranking for local document retrieval in agentic workflows.
How do I install QMD?
QMD is a MCP server. Primary install path: { "mcpServers": { "tobi-qmd": { "command": "npx", "args": ["-y", "@tobi/tobi-qmd"] } } }
What is the Skill Score for QMD?
QMD has a Skill Score of 7.2/10 (Solid). It is currently a provisional score — QMD hasn't been benchmarked yet, so it weights safety 75/100 (40%), maintenance 55/100 (30%), and documentation 85/100 (30%). Popularity carries zero weight until effectiveness is measured. See methodology.
Is QMD free?
QMD is published under the MIT license. See the vendor's source repository for any usage limits or paid tiers.
Related skills in Web research and discovery
Related
LlamaIndex
Leading document agent and OCR platform for building LLM applications over private data with retrieval, agents, and workflows.
Exa MCP Server
Connects AI assistants to Exa's search capabilities — web search, code search, and company research — through one MCP interface.
Local Deep Researcher
Fully local web research and report writing assistant powered by LangGraph and any local LLM via Ollama or LMStudio.
Embed this score
For maintainers · always-current, links back
Put it in your README — it updates as the score changes.[](https://skillsdirectory.co/skills/tobi-qmd)
<a href="https://skillsdirectory.co/skills/tobi-qmd"><img src="https://skillsdirectory.co/badge/tobi-qmd.svg" alt="Scored by skillsdirectory" height="28" /></a>