Deep Vault Research
SkillSearchLets your agent answer questions using a research skill that searches and links together your Obsidian notes.
Use Deep Vault Research in Claude, ChatGPT or Ahel Desktop
Free. Sign in, add Deep Vault Research and connect your AI. About a minute.
Also: Claude Code · Cursor · Codex
Then ask your AI: use the Deep Vault Research skill
Details
Instructions available. Your AI can read the instructions. Execution depends on the setup they require.
Account requirements not reviewed. Check the skill instructions before use; Ahel provides instructions and does not run this skill.
No other account needed.
Add Ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
About this skill
Use when the user asks a question about their vault, wants to explore a topic across notes, or says "research", "find out", "what do I know about", or "summarize my notes on". Chains semantic search, note reading, and link traversal for deep vault research.
What this skill tells your AI
The instructions your AI receives, as published by thoreinstein/gemini-obsidian in skills/research/SKILL.md and read by Ahel’s review.
Research a topic across the user's Obsidian vault using a multi-pass approach.
Workflow
-
Semantic Discovery — Call
obsidian_rag_querywith the user's question (limit 5-10). This surfaces the most relevant chunks across the vault. -
Targeted Reading — For each high-relevance result, call
obsidian_read_noteto get the full context. Skim for the most pertinent sections. -
Link Traversal — Check
obsidian_get_linkson the most relevant notes to find connected knowledge. Read promising linked notes. -
Synthesis — Combine findings into a clear answer with citations:
- Reference specific notes by name:
[[Note Name]] - Quote relevant passages when helpful
- Note any gaps or contradictions in the vault's knowledge
- Reference specific notes by name:
Arguments
The user's query follows the skill invocation. Example: /research what do I know about authentication patterns
Tips
- If RAG returns no results, suggest running
/indexfirst - For broad topics, do multiple RAG queries with different phrasings
- Always cite which notes your answer draws from
Signals
- GitHub stars
- 102
- Forks
- 9
- Last commit
- Jul 2026
- Hacker News mentions
- 20
Advanced
- Item type
- skill
- Key
research-thoreinstein- Source
- github.com/thoreinstein/gemini-obsidian