Obsidian search vault

Skill kogakure/skills/obsidian/obsidian-search-vault

Collection of AI Agent Code skills

Install

npx -y skills add kogakure/skills --skill obsidian-search-vault

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
8 stars8 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Search the Obsidian Zettelkasten vault using qmd hybrid search (BM25 + vector + LLM re-ranking). Use when the user wants to find notes, quotes, people, books, literature notes, permanent notes, or any content in their personal knowledge base.

SKILL.md

3.9 KB, as published. Nobody here has run it

Search Vault

Search the user's Obsidian Zettelkasten using qmd. The user's query is passed as arguments to /obsidian-search-vault.

Argument parsing

The argument string may contain:

A query (required): any free-form text
-c <collection>: restrict search to a specific collection
--full: fetch and display the full content of the top result
-n <num>: number of results (default: 5)
--bm25: use fast keyword-only search instead of hybrid
--vector: use pure vector search instead of hybrid

If no arguments are provided, ask the user what they want to search for.

Collections

Name	Contents
`quotes`	Short quotes and aphorisms from philosophers, scientists, politicians
`people`	Biographical notes on real and fictional persons
`books`	Book notes with ratings and summaries
`literature-notes`	Notes from articles, videos, and podcasts
`moc`	Maps of Content — thematic navigation hubs
`permanent-notes`	Evergreen atomic Zettelkasten notes
`unsorted-notes`	Large mixed collection (5000+ notes)
`series`	TV series notes
`movies`	Movie notes
`lyrics`	Lyrics and poems
`clippings`	Web clippings and saved articles
`podcasts`	Podcast notes
`companies`	Company and business notes
`prompts`	AI prompt notes
`countries`	Country notes
`cities`	City notes
`places`	Place notes (restaurants, cafes, locations)
`trips`	Trip and travel notes

Steps

Parse the query and flags from the user's arguments
Choose the right command (default: qmd query for hybrid search)
Run the search and display results clearly
If --full is specified, or the user asks to open/read a note, fetch it with qmd get <docid>
If results are poor, suggest alternative collections or rephrase

Commands

# Hybrid search — query expansion + BM25 + vector + LLM re-ranking (best quality)
qmd query "<query>"
qmd query "<query>" -c <collection>
qmd query "<query>" -n 10

# Fast BM25 keyword search (instant, no models)
qmd search "<query>"
qmd search "<query>" -c <collection>

# Pure vector similarity search
qmd vsearch "<query>"

# Fetch full note content
qmd get qmd://collection/filename.md

# List files in a collection
qmd ls <collection>

Output format

Present results as a clear numbered list:

Title with score badge
Collection name
Relevant excerpt from the note
qmd get path for reference

After showing results, offer to:

Fetch the full content of any result
Search a specific collection
Try a different query