For the complete documentation index, see llms.txt. This page is also available as Markdown.

Scroll to element

This action instructs the browser to intelligently calculate and scroll the page until the selected element appears within the viewport.

  • Effect: Ensures that the target element is always in view before the tool performs a click or retrieves data. Accurate scrolling helps reduce the "element not in viewport" error commonly encountered on long or lazy-loading web pages.

🎥 Watch the tutorial video: Here.

Configuration parameters:

  • XPath: The XPath identifier path of the target element that you want to bring into view (e.g., //button[@id="btn-load-more"] or //div[@class="footer-section"]).

Practical example: Bring a hidden button into view before clicking

Many modern websites implement anti-interaction technologies (bot trap security). If a button (e.g., "Load more" or "Confirm payment" button) is deeply hidden beneath the interface and has not been scrolled to, issuing a Mouse click command directly on the XPath of that button may result in a system error, or the website may refuse to acknowledge the click command due to the absence of user scroll behavior.

  • Configuration steps:

    1. Drag the Scroll to element action into the flow ➔ Enter the XPath of the target button.

    2. Then drag the Mouse click (or Wait element) action immediately after to perform the interaction.

  • Result: The browser will automatically and smoothly scroll down to the correct position containing that element, making it clearly visible on the interface, ensuring that the subsequent click command is executed successfully 100% like a real user.

Last updated

Was this helpful?