AI LLM

A fast, lightweight, zero-dependency local LLM plugin for Obsidian (Desktop & Mobile). Connect seamlessly to your local or self-hosted Ollama instance and interact with models directly inside your notes and via a modern sidebar chat panel.

✨ Features

• 📱 100% Cross-Platform (Desktop & Mobile): Built with native fetch APIs and standard Obsidian storage APIs. No node fs dependencies, fixing all mobile data.json loading errors.
• 💬 Modern & Compact Sidebar Chat View: Sleek, responsive panel with live streaming, real-time status indicators, auto-scroll, stop generation button, and message history.
• ⚡ High Performance & Zero Heavy Dependencies: Lightweight bundle that loads instantly without third-party node packages.
• 🛠️ Editor Shortcuts & Commands:
  • Ask LLM with selected text
  • Summarize selected text
  • Continue my story and enhance text
  • Make a story from my text
  • Fix grammar & improve writing
  • Explain selected text
• 🚀 Quick Prompt Chips: Instant quick actions (✨ Summarize, 📝 Story, 💡 Brainstorm, 🔧 Fix Grammar, ❓ Explain) built into the sidebar chat interface.
• ⚙️ Configurable System Prompts & Endpoints: Easily switch models, customize default system roles, and test server connection right from plugin settings.

🚀 Getting Started

1. Prerequisites (Install Ollama)

Ensure you have Ollama installed and running.

# Pull a model (e.g. tinyllama, llama3:8b, mistral, qwen, etc.)
ollama run tinyllama

Mobile Usage Tip: To use this plugin on Obsidian Mobile (iOS/Android), host Ollama on your local network (e.g. http://192.168.1.X:11434) or set OLLAMA_HOST=0.0.0.0 so your mobile device can reach your PC/server.

2. Plugin Installation

Option A: Manual Installation

1. Clone or download this repository into your vault's plugin directory:

   cd /path/to/your/vault/.obsidian/plugins/
   git clone https://github.com/Sparky4567/obsidian_ai_plugin.git ai_llm
   cd ai_llm
   npm install
   npm run build
   

2. Open Obsidian Settings -> Community Plugins.
3. Reload plugins and enable AI LLM.

Option B: Release Zip

1. Download main.js, manifest.json, and styles.css from the latest GitHub Release.
2. Place them inside .obsidian/plugins/ai_llm/ in your vault.
3. Enable the plugin in Obsidian Community Plugins.

🎯 Usage & Workflows

💬 Sidebar Chat View

1. Click the bot ribbon icon on the left sidebar (or run command Open AI Chat Panel).
2. Type your question or prompt in the compact input field at the bottom.
3. Use the quick prompt action chips above the input box for fast formatting.
4. Click Send (or press Enter). To stop streaming anytime, click Stop.

📝 Editor Commands (In-Note Generation)

1. Highlight any text inside your active note.
2. Open the Command Palette (Ctrl+P or Cmd+P).
3. Search for any AI LLM command:
   • Ask LLM with selected text
   • Summarize selected text
   • Fix grammar & improve writing
   • Explain selected text
4. The generated response will stream directly under or in place of your selection.

⚙️ Configuration

Go to Obsidian Settings -> AI LLM Settings:

• Ollama Endpoint: Set your Ollama server URL (Default: http://127.0.0.1:11434). Click Test Connection to verify availability.
• LLM Model: Select from automatically discovered models on your Ollama server.
• Default System Role: Customize the system persona (e.g., "You are a helpful assistant providing concise answers").

🛠️ Building & Development

# Install development dependencies
npm install

# Build production bundle
npm run build

# Watch mode for live dev edits
npm run dev

📄 License & Support

Distributed under the MIT License. See LICENSE for details.

Developed with ❤️ by Sparky4567.

• Website: artefaktas.eu
• Support on Ko-fi: ko-fi.com/artefaktas