AI Summarize

An intelligent Obsidian plugin that leverages OpenAI's latest models to generate high-quality summaries of your notes. Featuring advanced chunking for large documents, multiple placement options, and customizable profiles for different summarization needs.

Features

Core Functionality

• AI-Powered Summarization: Utilize OpenAI's latest models including GPT-4o, GPT-5, and reasoning models (o1, o3, o4)
• Smart Chunking: Automatically handles large documents by splitting them into manageable chunks while maintaining context
• Real-time Streaming: See summaries generate in real-time with token-by-token streaming
• Concurrency Control: Built-in safeguards prevent overlapping requests and allow cancellation of in-progress summaries

Flexible Placement Options

• Replace Selection: Replace selected text with the generated summary
• Insert Below: Add summary below the selected text
• Frontmatter: Automatically add summary to note's YAML frontmatter under summary property

Advanced Configuration

• Multiple Profiles: Create and manage different summarization profiles with unique settings
• Model Selection: Choose from GPT-3.5, GPT-4, GPT-4 Turbo, GPT-4o, GPT-5, and reasoning models
• Custom Prompts: Tailor the summarization prompt to your specific needs
• System Instructions: Define tone, style, language constraints, and output format
• Token Management: Set custom token limits with intelligent context window budgeting
• Automatic Model Detection: Plugin fetches and displays available models from your OpenAI account

Inline Summarization from menu:

Frontmatter Summarization from command palette:

Installation

From Obsidian Community Plugins

1. Open Obsidian and navigate to Settings → Community Plugins
2. Click Browse and search for "AI Summarize"
3. Click Install and then Enable
4. Configure your OpenAI API key in the plugin settings

Manual Installation

1. Download the latest release from the GitHub releases page
2. Extract the files to your vault's .obsidian/plugins/ai-summarize/ directory
3. Reload Obsidian
4. Enable the plugin in Settings → Community Plugins

Usage

Initial Setup

1. Get an API Key: Sign up at OpenAI and generate an API key
2. Configure Plugin:
   • Go to Settings → AI Summarize
   • Enter your OpenAI API key
   • The plugin will automatically fetch available models
3. Customize Settings (Optional):
   • Choose your preferred AI model
   • Adjust max tokens (default: 1000)
   • Customize the prompt and system instructions
   • Select summary placement preference

Generating Summaries

Method 1: Context Menu

1. Select text in your note (minimum 30 words)
2. Right-click to open the context menu
3. Click "AI summarize"
4. Watch as the summary generates in real-time

Method 2: Command Palette

1. Select text in your note (minimum 30 words)
2. Open command palette (Ctrl/Cmd + P)
3. Search for "AI Summarize: Summarize selection"
4. Execute the command

Large Document Handling

When summarizing large selections that exceed the model's context window:

• The plugin automatically splits content into chunks
• Each chunk is summarized individually
• A final "meta-summary" combines all partial summaries
• Progress is displayed with part numbers (e.g., "Part 1/3")

Configuration

Profile Management

Create multiple profiles for different summarization scenarios:

• Technical Notes: Use GPT-4 with specific technical language instructions
• Meeting Notes: Use GPT-4o-mini for quick, bullet-point summaries
• Research Papers: Use reasoning models (o1, o3) for deep analysis

Creating a Profile:

1. Go to plugin settings
2. Click "New profile"
3. Configure model, tokens, prompt, and placement
4. Name your profile
5. Switch between profiles using the dropdown

Tips for Best Results

Prompt Engineering: Depending on what you summarize you can use better propmpts and save these to different profiles!

Good: "Summarize the following in 3-4 sentences focusing on key decisions and action items"

Better: "Create a concise summary highlighting: 1) Main decisions made, 2) Action items with owners, 3) Open questions"

System Instructions:

• Specify language: "Always respond in Spanish"
• Define format: "Use bullet points with emoji indicators"
• Set tone: "Use professional, academic language"
• Add constraints: "Keep response under 100 words"

Key Features Implementation

Streaming Response:

• Uses OpenAI Responses API with streaming
• Throttled writing (50ms intervals) to reduce DOM operations
• Graceful fallback to non-streaming if streaming fails

Chunk Processing:

• Estimates tokens using ~4 chars per token heuristic
• Calculates available context window per model
• Splits text on paragraph boundaries
• Generates meta-summary from chunk summaries

Concurrency Control:

• Tracks current operation with run ID
• Uses AbortController for cancellation
• Prevents overlapping summarization requests

Contributing

Contributions are welcome! Here's how you can help:

1. Fork the repository
2. Create a feature branch: git checkout -b feature/amazing-feature
3. Make your changes and test thoroughly
4. Commit your changes: git commit -m 'Adds amazing feature'
5. Push to the branch: git push origin feature/amazing-feature
6. Open a Pull Request

Issues and Support

Encountered a bug or have a feature request?

1. Check existing issues
2. If not found, create a new issue
3. Provide:
   • Obsidian version
   • Plugin version
   • Model used
   • Steps to reproduce
   • Error messages (if any)

License

This project is licensed under the GPL-3.0 License.

Support the Project

If you find this plugin valuable and want to support its continued development:

Other ways to support:

• ⭐ Star the repository
• 🐛 Report bugs and suggest features
• 📖 Improve documentation
• 💻 Contribute code
• 📢 Share with others

Author

Alp Sariyer

• Website: alpsariyer.dev
• GitHub: @ravenwits

Acknowledgments

• Built with the Obsidian API
• Powered by OpenAI
• Inspired by the Obsidian community's need for intelligent note summarization