qmd_skill
1
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill michaelvessia/nixos-config --skill qmd- SKILL.md2.9 KB
Overview
This skill provides local semantic and full-text search across your markdown docs, notes, and knowledge bases. It combines BM25 full-text search, vector semantic search, and LLM reranking to return relevant snippets or full documents from your local vaults. Use it to quickly find, retrieve, and contextually summarize content stored in Obsidian or any markdown collection.
How this skill works
qmd indexes local markdown files into collections and supports three search modes: BM25 full-text, vector semantic search, and a hybrid LLM-reranked query for highest-quality results. You can restrict searches to specific collections, adjust result limits, request full document content, and export results in JSON, CSV, Markdown, or XML. Retrieval commands let you fetch documents by path, ID, or glob patterns.
When to use it
- You want to find notes or passages in your Obsidian vault or local markdown directory.
- You ask, "what did I write about X" or "do I have notes on X."
- You need context for an LLM prompt from personal docs or meeting notes.
- You need to search journals, meeting notes, or a specific collection.
- You want structured exports (JSON/CSV/MD) of search results for tooling.
Best practices
- Use qmd query for the best results; it applies LLM reranking on candidate matches.
- Limit results with -n and set --min-score to filter low-relevance hits.
- Restrict searches to a collection (-c or --collection) to reduce noise.
- Use --full when you need complete document context for follow-up analysis.
- Export with --json or --files when integrating results into other tools.
Example use cases
- Quickly locate the API design notes you wrote last quarter: qmd query -n 10 "API design"
- Find meeting action items in your work collection: qmd search "action items" -c work
- Fetch a full journal entry for a date or topic: qmd query --full "reflection on Q4"
- List indexed collections and check index health before a mass search: qmd collection list; qmd status
- Export search matches to CSV for analysis: qmd search "bugs" --files
FAQ
Use qmd query for hybrid search with LLM reranking; it yields the most relevant results.
Can I limit searches to a single vault or folder?
Yes. Use -c or --collection to restrict results to a named collection you added to qmd.