Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Output audio file example changed from .caf to .m4a in documentation. - No changes to code or functionality; documentation update only.
What This Skill Does
record is a macOS CLI tool for capturing audio from the microphone, video or screenshots from the screen, and video or photos from the camera. Output file paths print to stdout and status messages to stderr, making it usable in shell pipelines and AI agent workflows. Supports JSON output and a duration flag for non-interactive use.
The stdout/stderr separation and --duration flag let scripts and AI agents capture media without manual interaction or GUI tools.
When to use it
- Recording a voice memo from the terminal without opening an app
- Taking a screenshot of a specific window by name from a script
- Capturing webcam footage during a timed session
- Recording screen with microphone audio for a bug report
- Automating media capture steps inside an AI agent pipeline
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Record 10 seconds of audio and save it to /tmp/meeting.m4a
- 1Confirm with the user that microphone audio will be captured for 10 seconds
- 2Run: record audio --duration 10 --output /tmp/meeting.m4a
- 3Read the output file path from stdout
- 4Report the saved path back to the user
Audio saved to /tmp/meeting.m4a
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.