---
name: QMD Memory Search
slug: qmd-memory-search
category: AI Engineering
description: QMD Memory Search upgrades ClaudeClaw’s grep-based memory search with local hybrid BM25, vector semantic search, and LLM reranking. Use it when you want better recall across large memory collections without API keys.
github: "https://github.com/sbusso/claudeclaw/tree/main/skills/add-qmd"
language: TypeScript
stars: 193
forks: 60
install: "npx degit https://github.com/sbusso/claudeclaw/tree/main/skills/add-qmd ~/.claude/skills/add-qmd"
installs_to: ~/.claude/skills/add-qmd
source_path: skills/add-qmd/SKILL.md
collection_size: 25
category_size: 3278
collection_url: "https://dirskills.com/collections/sbusso/claudeclaw"
added: 2026-09-06T05:19:00.794Z
last_synced: 2026-09-06T05:19:00.794Z
canonical_url: "https://dirskills.com/skills/qmd-memory-search"
---

# QMD Memory Search

QMD Memory Search upgrades ClaudeClaw’s grep-based memory search with local hybrid BM25, vector semantic search, and LLM reranking. Use it when you want better recall across large memory collections without API keys.

**Install:**

```bash
npx degit https://github.com/sbusso/claudeclaw/tree/main/skills/add-qmd ~/.claude/skills/add-qmd
```

## README

# Add QMD Memory Backend

QMD (https://github.com/tobi/qmd) is a local search engine for markdown files. It combines BM25 keyword search, vector semantic search, and LLM re-ranking — all running on-device via node-llama-cpp with GGUF models.

This skill upgrades ClaudeClaw's built-in grep-based `memory_search` MCP tool with QMD's hybrid search, giving agents much better recall across large memory collections.

## Prerequisites

- ClaudeClaw with memory tools already working (memory_search, memory_save, memory_get)
- ~2GB disk space for GGUF embedding + reranking models (downloaded on first run)
- Node.js 20+

## What This Skill Does

1. Installs QMD as an MCP server dependency
2. Configures QMD to index all group memory directories (`groups/*/memory/`, `groups/*/CLAUDE.md`, `groups/*/conversations/`)
3. Replaces the grep-based `memory_search` in `agent/runner/src/ipc-mcp-stdio.ts` with QMD's API
4. Adds QMD indexing to the agent startup flow (incremental re-index on each run)

## Implementation Notes

### QMD MCP Server

QMD exposes its search as an MCP server. Add to the agent runner's `mcpServers` config:

```typescript
qmd: {
  command: 'npx',
  args: ['qmd', 'mcp', '--collection', collectionPath],
  env: {},
}
```

### Indexing

QMD indexes markdown files into a local SQLite database. The collection should be configured per-group:

```bash
# Index a group's memory
qmd index --collection groups/{folder}/.qmd groups/{folder}/memory/ groups/{folder}/CLAUDE.md groups/{folder}/conversations/
```

### Search Integration

Replace the grep-based `memory_search` tool body with a call to QMD's MCP:

```typescript
// Before (grep-based):
const results = grepFiles(args.query, allFiles);

// After (QMD):
// Use the qmd MCP server's search tool
// Returns semantically ranked results with relevance scores
```

### Fallback

If QMD is not installed or indexing fails, fall back to the built-in grep-based search. This ensures memory tools always work even without QMD.

## Not Implemented Yet

This skill is a specification for future implementation. The built-in grep-based memory tools work without QMD. Run this skill when you want to upgrade to semantic search.