usecomputer - Desktop automation CLI for AI agents
Takes screenshots, clicks, types, scrolls, drags on macOS, Linux, Windows
Tags
Updated: 2026-05-11Capabilities
Typical Inputs
Typical Outputs
What this skill does
- take screenshot
- click
- type text
- press key
- scroll
- drag
- list windows
- list displays
- validate coordinates
Inputs
- Accessibility permission
- X11 session
- interactive desktop
- window ID
- coordinates
- text input
Outputs
- screenshot image
- JSON metadata
- debug point image
Requirements
- macOS
- Linux with X11
- Windows desktop session