Actions7
- Content Actions
- Navigation Actions
Overview
The node provides an interface to interact with websites using a cloud-based browser instance. It supports opening and managing browser sessions remotely, navigating to URLs, clicking on page elements, and retrieving content such as HTML, screenshots, or PDFs from web pages.
For the Navigation > Open Browser operation specifically, the node opens a new browser instance in the cloud with customizable settings like browser type, headless mode, stealth mode, session management, proxy configuration, and custom launch arguments. This is useful when you want to programmatically start a browser session for further automated interactions or scraping tasks without managing local browser installations.
Practical examples:
- Opening a browser session to later navigate to multiple URLs or perform complex interactions.
- Starting a browser with stealth mode enabled to avoid detection by anti-bot systems.
- Using proxy settings to route traffic through specific servers for geo-location testing or anonymity.
- Saving and recovering browser sessions to maintain login states across workflow executions.
Properties
| Name | Meaning |
|---|---|
| Browser Type | Choose the browser engine to use: Chrome, Chromium, or ChromeHeadlessShell. |
| Headless Mode | Whether to run the browser in headless mode (no visible UI). |
| Stealth Mode | Enable stealth mode to reduce detection by websites that block automated browsers. |
| Keep Open (Seconds) | Duration in seconds before the browser auto-closes. Set to 0 to keep the browser open indefinitely. |
| Label | A custom name label for the browser instance to help recognize it. |
| Save Session | Whether to save the browser session data for reuse in future executions. |
| Recover Session | Whether to recover a previously saved browser session instead of starting fresh. |
| Custom Arguments | Additional command-line arguments to pass to the browser on startup. |
| Ignored Default Arguments | List of default browser arguments to ignore/remove when launching the browser. |
| Proxy Configuration | Settings for routing browser traffic through a proxy server, including host, port, username, and password. |
Output
The output JSON object includes:
webSocketAddress: The WebSocket endpoint address of the opened browser instance, used for connecting and controlling the browser.sessionId: An identifier for the browser session.status: A message indicating success, e.g., "Browser opened successfully".
This information is essential for subsequent operations like navigation or interaction within the same browser session.
Dependencies
- Requires an API token credential for authenticating with the external cloud browser service.
- Uses Puppeteer library internally to connect and control the browser via the WebSocket address.
- Network access to the cloud browser API endpoint (
https://production.cloudbrowser.ai/api/v1/Browser/Open).
Troubleshooting
- No WebSocket address received: If the response from the browser service does not include a WebSocket address, the node throws an error. Ensure your API token is valid and the cloud browser service is operational.
- Session recovery issues: If recovering a session fails, verify that a session was previously saved and that the correct session data is accessible.
- Proxy connection failures: Incorrect proxy settings (host, port, credentials) can prevent the browser from launching properly.
- Timeouts or slow responses: Network latency or service downtime may cause delays; consider increasing timeout settings or retrying later.
Links and References
- Puppeteer Documentation – For understanding browser automation concepts used internally.
- Cloud browser service API documentation (not publicly linked here) – Consult your service provider for detailed API usage and limitations.