qmd-obsidian

To index and search an Obsidian vault, you add the vault as a collection using a markdown file mask, then execute a query to retrieve matching files. The skill requires qmd tooling installed and uses commands like collection add and search to surface notes.

Yes, you can perform semantic search across Obsidian notes and meetings. The skill supports multi-mode search including BM25 keyword searches, semantic searches, and hybrid ranking to deliver precise and contextually relevant results from your vault.

Yes, you need qmd tooling installed locally. The skill relies on qmd command-line operations such as status, collection add, context add, search, vsearch, and query to index the vault and retrieve structured outputs.

Keyword search uses BM25 ranking to find exact term matches, while semantic search interprets meaning to find conceptually related notes. The skill supports both modes plus hybrid ranking to surface precise results across notes, docs, and meetings.

You get structured JSON or Markdown outputs by running qmd search or query commands against your indexed Obsidian vault. The skill emits these formats natively, allowing you to integrate the retrieved notes and meeting context into downstream applications.