メニュー
Search and retrieve markdown documents from local knowledge bases using qmd. Supports BM25 keyword search, vector semantic search, and hybrid search with LLM re-ranking. Use for querying indexed notes, documentation, meeting transcripts, and any markdown-based knowledge.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | qmd |
| description | Search and retrieve markdown documents from local knowledge bases using qmd. Supports BM25 keyword search, vector semantic search, and hybrid search with LLM re-ranking. Use for querying indexed notes, documentation, meeting transcripts, and any markdown-based knowledge. |
Search and retrieve documents from locally indexed markdown knowledge bases.
git clone https://github.com/tobi/qmd && cd qmd && bun install -g .
Add collection first (required before embed):
qmd collection add ~/notes --name notes
Create embeddings (enables vsearch/query):
qmd embed
First run downloads ~2.5GB of models to ~/.cache/qmd/models/
Verify setup:
qmd status
Always use --json flag for structured output when invoking qmd commands programmatically.
qmd search "authentication flow" --json
qmd search "error handling" -n 10 --json
qmd search "config" -c notes --json
qmd vsearch "how does login work" --json
qmd vsearch "authentication best practices" -n 20 --json
qmd query "implementing user auth" --json
qmd query "deployment process" --min-score 0.5 --json
| Option | Description |
|---|---|
-n NUM | Number of results (default: 5, or 20 with --json/--files) |
-c, --collection NAME | Restrict to specific collection |
--min-score NUM | Minimum score threshold |
--full | Return complete document content |
--all | Return all matches |
| Format | Flag | Use Case |
|---|---|---|
| Snippets | (default) | Human reading |
| JSON | --json | Parsing in code |
| Files only | --files | Paths + scores only |
| Markdown | --md | Documentation |
| CSV | --csv | Spreadsheet export |
qmd get docs/guide.md --json
qmd get notes/meeting.md:50 -l 100 --json
qmd get "#a1b2c3" --json # Hash from search results
qmd multi-get "docs/*.md" --json
qmd multi-get "api.md, guide.md" --json
qmd multi-get "notes/**/*.md" --max-bytes 20480 --json
Hash references (e.g., #a1b2c3) are chunk identifiers shown in search output.
qmd status # Check index health and collection stats
qmd update # Re-index changed files
qmd update --pull # Git pull first, then re-index
qmd collection list # List all collections
qmd collection remove NAME # Remove a collection
qmd cleanup # Remove orphaned data, vacuum DB
| Mode | Speed | Quality | Best For |
|---|---|---|---|
| search | Fast | Good | Exact keywords, known terms |
| vsearch | Medium | Better | Conceptual queries, synonyms |
| query | Slow | Best | Complex questions, uncertain terms |
~/.cache/qmd/models/)qmd can run as an MCP server:
qmd mcp
Exposes tools: qmd_search, qmd_vsearch, qmd_query, qmd_get, qmd_multi_get, qmd_status
Google Workspace CLI for Gmail, Calendar, Drive, Docs, Sheets, Tasks, and more
Render Mermaid diagrams as SVG and PNG using the Beautiful Mermaid library. Use when the user asks to render a Mermaid diagram.
Generate, edit, and remix images using the Reve AI API. Use when creating images from text prompts, editing existing images with instructions, or combining/remixing multiple reference images. Requires REVE_API_KEY or REVE_AI_API_KEY environment variable.
Fastmail email, masked emails, calendar, contacts, and vacation management
Search the web with AI-powered answers and citations via Perplexity API. Use for research queries needing current info with sources. Env: PERPLEXITY_API_KEY.