Browser Screenshot

Browser Screenshot

When invoked, capture a screenshot of the given URL via a Browserless Chromium service running locally or in your environment.

Inputs

• url -- target page (required)
• viewport -- {width, height}, default {1920, 1080}
• output_path -- local file path for the resulting PNG, default /tmp/screenshot.png
• full_page -- boolean, default false

Steps

1. POST to http://:/screenshot with the URL and viewport in the request body.
2. Save the returned PNG bytes to output_path.
3. Reference the saved path in your reply to the user.

Output

• File path to the PNG screenshot.

Example invocation

curl -s -X POST http://localhost:30001/screenshot \
  -H 'Content-Type: application/json' \
  -d '{
        "url": "https://example.com",
        "options": {"type": "png", "fullPage": false},
        "viewport": {"width": 1920, "height": 1080}
      }' \
  -o /tmp/screenshot.png

Notes

• Browserless /screenshot endpoint requires no authentication by default.
• For the full setup guide (cross-host usage, troubleshooting, generating an SOP), see sops/SOP_Headless_Browser_Screenshot.md.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: MrTrenchTrucker
• Source: MrTrenchTrucker/contak-browser-skills
• License: MIT

Install and usage instructions live in the source repository linked above.