Key Features
- Multi-model support: Claude, GPT, Gemini, ChatBrowserUse (optimized), local models
- Real browser control: Click, type, scroll, screenshot, navigate
- Form automation: Job applications, data entry, checkout flows
- Web research: Find, compile, and summarize information
- Custom tools: Extensible architecture for specialized tasks
- CLI interface: Persistent browser sessions from terminal
Supported Models
| Model | Best For |
|---|---|
| ChatBrowserUse | Browser automation (purpose-built, cheapest) |
| Claude Sonnet | Complex reasoning tasks |
| Gemini Flash | Fast, cost-effective tasks |
| Local models | Privacy-sensitive automation |
CLI Usage
browser-use open https://example.com # Navigate
browser-use state # See clickable elements
browser-use click 5 # Click element #5
browser-use type "Hello world" # Type text
browser-use screenshot output.png # Capture pageProject Templates
# Generate a starter project
uvx browser-use init --template default
# Templates: default, advanced, tools