HeadlessX icon

HeadlessX

Anti-detection web scraping with HeadlessX v2 API

Website → Extract HTML (JS Rendered)

AI-generated

Summary

Fetches fully rendered HTML from a webpage using JavaScript execution.

Inputs

  • url (required) — The URL of the web page to process.
  • htmlJsOptions — A collection of options to control the render process: Custom Headers (HTTP headers), Enable Preview (generates a downloadable binary file), Extra Wait Time (additional ms to wait for dynamic content), Timeout (maximum time to wait for page load in ms), User Agent (custom user agent string), and Wait Until (condition to consider navigation complete).

Output shape

A single item containing the rendered HTML string in JSON and optionally a binary file if Preview is enabled.

Includes metadata such as URL, content length, timestamp, and preview mode status.

Examples

Example 1: Render a dynamic page

Enter a URL for a Single Page Application and set Wait Until to 'Network Idle'.

Example 2: Download the rendered HTML

Enable 'Enable Preview' to generate a downloadable binary file alongside the JSON output.

Links

Discussion