Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Captures macOS screenshots by recording a 1-second clip via OpenClaw's screen.record node, then extracting a PNG frame with ffmpeg. Works in headless environments where standard screencapture commands often fail due to permission restrictions.
Routes through the screen.record node instead of screencapture, avoiding the permission errors that block standard screenshot tools in headless or automated contexts.
When to use it
- Answering 'what's on my screen?' questions
- Capturing UI state during manual testing
- Attaching a screenshot to a bug report
- Documenting current screen contents in a remote session
- Debugging visual layout without leaving the terminal
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Take a screenshot of my screen
- 1Creates a tmp directory inside the skill folder
- 2Records a 1-second clip using openclaw nodes screen record with no audio
- 3Extracts the first video frame as a PNG using ffmpeg
- 4Reads the PNG file and attaches it to the reply
A PNG screenshot of the current screen attached to the agent's response
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.