Agent skill · vasilyu1983
docs-notes-retrieval
Builds local-first note vault retrieval for Obsidian, markdown notebooks, and NotebookLM-adjacent exports. Use when packaging notes into LLM-ready context.
What it needs
About 8k tokens when loaded.
What this skill does
Notes Retrieval Quick Start — Concrete Commands Three common scenarios using the scripts in scripts/. Scenario 1: Vault inventory Before indexing or packaging, understand what is in the vault. Scenario 2: Packaged context for an LLM session Build a markdown bundle ready to paste into a Claude or GPT session. Scenario 3: Orphan note cleanup Find and review notes with no inbound or outbound wikilinks — usually stale or misplaced. Orphan cleanup workflow: 1. Run orphans to get the list. 2. Review word count: orphans < 50 words are likely stubs — archive or delete. 3. Orphans > 200 words with no links may be valuable but unconnected — add wikilinks or a backlink from a map-of-content note. 4. Re-run orphans after cleanup to verify the count dropped. Judgment call before acting on the list: scanvault.py orphans only detects [[wikilink]] topology. In vaults that organize by folder hierarchy or tags instead of wikilinks (common in Logseq-style or PARA-folder vaults), a genuinely well-connected note can show up as a false-positive orphan. Cross-check folder placement and tags before archiving anything the tool flags — treat the orphan list as a triage queue, not a deletion list. --- Use this skill to build or operate local-first retrieval pipelines for note vaults, markdown knowledge bases, and notebook-style exports. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills vasilyu1983/docs-notes-retrieval