webpage-screenshot

What This Skill Does

The webpage-screenshot skill provides OpenClaw agents with the capability to capture visual representations of any public URL. By leveraging Playwright-driven automation, the skill renders a web page in a headless Chromium browser and saves the complete document as a high-quality PNG image. Whether you need a simple visual record or a full-page scroll capture, this tool ensures precise rendering of the requested web content.

Installation

To integrate this skill into your OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/malenconiaprincep/webpage-screenshot

Ensure that you have performed npm install within your project root. Additionally, because this skill relies on browser automation, you must install the necessary browser binaries by running npx playwright install chromium before attempting the first screenshot.

Use Cases

This skill is ideal for developers, researchers, and automated QA engineers. Common scenarios include:

• Generating visual documentation for web applications or API portals.
• Archiving snapshots of landing pages for design auditing.
• Converting dynamic web content into static images for reporting.
• Visual tracking of website changes over time during development cycles.

Example Prompts

1. "Take a screenshot of https://github.com/openclaw/openclaw and save it to ./docs/homepage.png."
2. "Could you please visit https://example-app.dev and capture the full page as an image?"
3. "Open https://status.aws.amazon.com and provide a screenshot of the current page status."

Tips & Limitations

The screenshot functionality captures the entire page layout as it appears in a 1280px wide viewport. It is important to note that sites requiring authentication or interactive login sessions may not be accessible via this automated script. For sites protected by CAPTCHAs, automated rendering might be blocked. For lightweight interaction or parsing structure without saving files, consider using the browser MCP integration to navigate and snapshot the page document object model.