CloudBrowser icon

CloudBrowser

Interact with websites using a cloud-based browser instance

Overview

This node, named "CloudBrowser," enables interaction with websites through a cloud-based browser instance. It supports navigating to URLs, capturing screenshots, retrieving HTML content, generating PDFs, and controlling browser instances remotely. This is particularly useful for web scraping, automated testing, visual monitoring of web pages, or generating reports from dynamic web content.

For the Content resource with the Get Screenshot From Website operation, the node navigates to a specified URL using a cloud browser, then takes a screenshot of the page according to user-defined options such as full page capture, image format (PNG or JPEG), quality, and clipping area.

Practical examples:

  • Automatically capture screenshots of product pages for price comparison.
  • Generate visual snapshots of dashboards or reports at scheduled intervals.
  • Capture website previews for link sharing or archiving purposes.

Properties

Name Meaning
URL to Navigate The URL of the webpage to open and capture a screenshot from.
Navigation Options Options controlling how navigation behaves:
- Wait Until: When to consider navigation finished (load, domcontentloaded, networkidle0, networkidle2).
- Timeout (Ms): Max time to wait for navigation.
Browser Configuration Settings for the cloud browser instance:
- Browser Type: Choose between Chrome, Chromium, or ChromeHeadlessShell.
- Headless Mode: Run browser without UI.
- Stealth Mode: Enable stealth to avoid detection.
- Keep Open (Seconds): How long to keep browser open before auto-closing (0 = never).
- Label: Name for the browser instance.
- Save Session: Save session for reuse.
- Recover Session: Recover previously saved session.
Custom Arguments Additional command-line arguments to pass to the browser on startup.
Ignored Default Arguments Default browser arguments to ignore when launching.
Proxy Configuration Proxy server settings:
- Host, Port, Username, Password.
Screenshot Options Controls screenshot specifics:
- Full Page: Capture entire scrollable page or just viewport.
- Quality: Image quality for JPEG (0-100).
- Type: Image format (jpeg or png).
- Clip: Defines rectangular area to capture (x, y, width, height).

Output

The node outputs JSON data containing details about the screenshot and the webpage:

  • url: The final URL of the loaded page.
  • title: The page title.
  • screenshot: A base64-encoded data URI string of the screenshot image.
  • screenshotBinary: The raw binary data of the screenshot file.
  • filename: Generated filename for the screenshot (e.g., screenshot_1672531200000.png).
  • fileExtension: File extension based on image type (png or jpg).
  • mimeType: MIME type of the image (image/png or image/jpeg).

The binary output can be used directly in workflows that support binary data handling, such as saving files or sending images.

Dependencies

  • Requires an active connection to the external CloudBrowser API service via an API token credential.
  • Uses Puppeteer library internally to control the browser instance over WebSocket.
  • Needs network access to the target URLs and optionally proxy configuration if required.
  • The node expects the CloudBrowser API endpoint https://production.cloudbrowser.ai/api/v1/Browser/Open to be reachable.

Troubleshooting

  • No WebSocket address received: Indicates failure to open a browser instance. Check API token validity, network connectivity, and CloudBrowser service status.
  • Timeout errors during navigation: Increase the timeout value in Navigation Options or verify the target URL's availability.
  • Invalid clip dimensions: Ensure clip coordinates and sizes are within the page bounds.
  • Authentication or permission errors: Verify the API token credential is correctly configured and has necessary permissions.
  • Browser connection failures: Confirm the WebSocket address is valid and the browser instance is running.
  • Unsupported operation/resource: Make sure the selected Resource and Operation combination is supported by the node.

Links and References

Discussion