Setup & Installation
openclaw skills install @emptyopen/macos-desktop-control
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
npx clawhub install macos-desktop-control
What This Skill Does
Gives agents the ability to see and interact with a macOS desktop. Uses cliclick for mouse and keyboard input, and the native screencapture utility to capture the current screen state as a PNG.
Lets agents control any macOS app regardless of whether it exposes an API or CLI interface.
When to use it
- Clicking UI buttons in apps that lack an API
- Typing into desktop application fields
- Automating repetitive GUI workflows on Mac
- Taking screenshots to verify visual state
- Moving the mouse to trigger hover-based UI elements
Example Workflow
Here's how your AI assistant might use this skill in practice.
INPUT
User asks: open the Settings app and take a screenshot of the Wi-Fi panel
AGENT
- 1Calls see to capture current screen state
- 2Identifies the Settings icon position from the screenshot
- 3Calls click with coordinates to open Settings
- 4Navigates to Wi-Fi using click and type commands
- 5Calls see again to capture the Wi-Fi panel
OUTPUT
PNG screenshot saved to /tmp/claw_view.png showing the Wi-Fi settings panel
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.
cliclick installed via Homebrew (brew install cliclick)macOS screen recording permission granted to the terminal or agent process