Capabilities
| Feature | Description |
|---|---|
| Vision-based | Sees pixels, not DOM. Works on any UI. |
| Cross-platform | Desktop, web, Android, iOS, HMI |
| Background mode | Runs without taking over your screen (Windows) |
| Natural language | Describe tasks in plain English |
Use Cases
- Agentic development: Give Claude Code/Cursor UI control
- Test automation: CSV-based test suites with auto-reporting
- Process automation: Automate workflows across any app