Obsidian LLM Token Counter

A lightweight, performance-focused Obsidian plugin that displays the estimated token count for LLMs (like GPT-4, Claude, Gemini) in your status bar.

Designed to be unobtrusive and efficient, it helps you manage context windows before sending your notes to AI tools.

✨ Features

• 🚀 Zero Lag: Uses a debounced update strategy (updates 500ms after you stop typing) to ensure your typing experience remains buttery smooth, even in large files.
• 📊 Real-time Status Bar: Integrates seamlessly into the Obsidian footer alongside word count and backlinks.
• ⚙️ Configurable Models: Switch between different tokenization strategies (GPT-4, GPT-3.5, Legacy).
• 🔋 Battery Friendly: Only calculates when necessary.

📥 Installation

Manual Installation

1. Download the main.js, manifest.json, and styles.css from the latest Release.
2. Create a folder named obsidian-llm-token-counter in your vault's .obsidian/plugins/ directory.
3. Move the downloaded files into that folder.
4. Reload Obsidian and enable the plugin in Community Plugins settings.

Development Build

1. Clone this repository.
2. Run npm install to install dependencies.
3. Run npm run build to compile the plugin.
4. Copy the files to your plugin directory.

🛠️ Usage

Once enabled, look at the bottom right of your Obsidian window. You will see a counter like: 1.2k tokens

• Hover over the counter to see which model encoding is currently active.
• Click (future feature) to copy the count or open settings.

🧩 Roadmap

• Basic Structure & Status Bar Integration
• Debounce Logic for Performance
• Integration with js-tiktoken for 100% accurate GPT-4 counting
• Support for Claude (Anthropic) and Gemini (Google) tokenizers
• "Warning Threshold" setting (turn red when approaching 8k/32k tokens)

🤝 Contributing

Contributions are welcome! Please see DOCUMENTATION.md for technical details on how the plugin is architected.

📄 License

MIT License. See LICENSE for more information.