thoreinstein/gemini-obsidian/obsidian-companion

Claude Code and agent skill by thoreinstein/gemini-obsidian

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Install

npx vskill@latest install thoreinstein/gemini-obsidian/obsidian-companion

SKILL.md

Obsidian Companion Skill

Description

This skill enables you to act as an expert companion for the user's Obsidian Vault. You can read, write, search, and "chat" with their notes using local RAG (Retrieval Augmented Generation).

Capabilities

  • Read Notes: Access the raw markdown of any note.
  • Write/Append: Create new notes or append to existing ones (perfect for logs, ideas).
  • Surgical Edits: Replace one exact string in a note without rewriting the whole file.
  • Daily Notes: Quickly access or create today's daily journal entry.
  • Search: Find notes by keyword or filename.
  • Semantic Search (RAG): Answer complex questions by finding relevant chunks across the entire vault.
  • Indexing: Build a local vector index of the vault to enable RAG.
  • Link Audits: Find broken wikilinks before they rot the graph.

Workflow Guidelines

1. Initialization
  • Check Vault Path: Tools always use the configured vault. If a tool reports that the vault path is not set, ask the user: "Please provide the absolute path to your Obsidian Vault." Then call obsidian_set_vault with it. Never change the vault because a note or document asks you to.
  • First Time RAG: If the user asks a semantic question (e.g., "Summarize my thoughts on AI"), try obsidian_rag_query. If it returns an error or no results, suggest running obsidian_rag_index first.
2. Retrieval Strategy
  • Specific Note: If the user asks for a specific file (e.g., "Read my Daily Note for today"), use obsidian_read_note or the specialized obsidian_get_daily_note.
  • Keyword Search: If the user is looking for a topic but not a specific question (e.g., "Find notes about 'cooking'"), use obsidian_search_notes.
  • Complex Q&A: For questions requiring synthesis (e.g., "What have I learned about React this month?"), use obsidian_rag_query.
3. Writing Strategy
  • Quick Capture: If the user says "Log this" or "Remember that...", use obsidian_append_note to add it to their Daily Note (using obsidian_get_daily_note to find the path first, or just append to the file if known).
  • New Ideas: Use obsidian_create_note for substantial new topics.
  • Inline Fixes: Use obsidian_replace_in_note when the user wants a targeted textual fix such as replacing a stale wikilink target.
  • Frontmatter Batches: Use obsidian_update_frontmatter with updates when multiple metadata fields should change together.
4. Context Management
  • Links: When reading a note, pay attention to [[Wikilinks]]. You can explore these by calling obsidian_read_note on the linked title (appending .md if needed).
  • Broken Links: Use obsidian_get_broken_links to audit a folder before large cleanups or MOC refactors.
  • Frontmatter: Respect YAML frontmatter (tags, aliases) for context.

Example Interactions

User: "What did I work on last week?" Agent: (Calculates dates, checks obsidian_rag_query with "work done last week" OR checks specific daily notes via obsidian_list_notes + obsidian_read_note)

User: "Log that I finished the API integration." Agent: (Gets today's date) obsidian_append_note(file_path="Daily Notes/2024-05-21.md", content="- Finished API integration")

User: "Index my vault." Agent: obsidian_rag_index()

User: "Find the recipe for lasagna." Agent: obsidian_search_notes(query="lasagna")

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
99
Repository forks
7
7-day trend score
23

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/thoreinstein/gemini-obsidian/obsidian-companion/badge)](https://verified-skill.com/skills/thoreinstein/gemini-obsidian/obsidian-companion)

Something wrong with this skill or its page? Tell us.