Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Captures screenshots of any URL from the command line using capture-website-cli. Supports full-page capture, custom viewport dimensions, dark mode, device emulation, and waiting for specific elements before shooting.
Handles dynamic and slow-loading pages with configurable delays and element-wait options that simpler screenshot utilities lack.
When to use it
- Archive a tweet before it gets deleted
- Capture a breaking news article at a specific moment
- Take a mobile-view screenshot to check responsive layouts
- Screenshot a web app in dark mode
- Grab a dynamically loaded page after content finishes rendering
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Screenshot this tweet for me: https://x.com/elonmusk/status/2026052687423562228
- 1Run capture-website with the URL and --output set to the workspace path
- 2Set --width=1280 and --height=800 for a standard viewport
- 3Add --delay=2 if the page loads slowly
- 4Send the saved PNG file to the user
A PNG screenshot of the tweet saved to the workspace and delivered to the user
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.