obsidian-gemini  by allenhutchison

Gemini AI agent for Obsidian knowledge management

Created 1 year ago
273 stars

Top 94.5% on SourcePulse

GitHubView on GitHub
1 Expert Loves This Project
Project Summary

This Obsidian plugin integrates Google's Gemini AI models to provide advanced AI-powered assistance for note-taking and knowledge management. It targets Obsidian users seeking personalized, context-aware AI interactions, enabling tasks like writing, research, and vault management directly within their notes.

How It Works

The plugin leverages the Google Gemini API, offering an "Agent Mode" with sophisticated tool-calling capabilities. This allows the AI to actively interact with the user's Obsidian vault by searching, reading, creating, editing, and organizing files. It employs context-aware processing, using notes as context for AI interactions, and features experimental semantic vault search for meaning-based queries. The architecture supports extensible agent skills and external tool integration via MCP servers.

Quick Start & Requirements

  • Primary install: Install via Obsidian Community Plugins (search for "Gemini Scribe") or manually download release assets to .obsidian/plugins/obsidian-gemini/.
  • Prerequisites: A Google Gemini API key is required, obtainable from Google AI Studio. The API key is securely stored using Obsidian's SecretStorage.
  • Links: Google AI Studio.

Highlighted Details

  • Agent Mode with Tool Calling: Enables AI to perform vault operations (file management, research, content generation) with persistent sessions and granular permissions.
  • Semantic Vault Search (Experimental): Indexes notes for keyword-independent, meaning-based search using Google's File Search API.
  • IDE-Style Completions: Provides real-time, context-aware inline text suggestions as you type.
  • MCP Server Support (Experimental): Connects to external tool servers via stdio or HTTP with OAuth authentication.
  • Agent Skills: Supports extensible AI capabilities defined by the agentskills.io specification.
  • Image Paste Support: Allows direct pasting of images into the chat for multimodal AI analysis.
  • Secure API Key Storage: API keys are migrated to Obsidian SecretStorage (OS keychain).

Maintenance & Community

Developed by Allen Hutchison. Contributions are welcome, with development guidelines provided in CLAUDE.md. No specific community channels (e.g., Discord, Slack) are listed in the README.

Licensing & Compatibility

Licensed under the MIT License. This permissive license generally allows for commercial use and integration with closed-source projects.

Limitations & Caveats

Features marked as "Experimental" (MCP Server Support, Semantic Vault Search) may be subject to change or instability. A Google Gemini API key is a mandatory requirement for the plugin's functionality.

Health Check
Last Commit

14 hours ago

Responsiveness

Inactive

Pull Requests (30d)
60
Issues (30d)
58
Star History
46 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.