Built by Metorial, the integration platform for agentic AI.
Retrieve the fully rendered HTML of a web page after JavaScript execution. Provide a URL to navigate to, or supply raw HTML to render in the browser. Returns the complete DOM including dynamically generated content from single-page applications.
Run a Google Lighthouse audit on a web page to analyze performance, accessibility, SEO, best practices, and PWA metrics. Filter by categories or specific audits to reduce response size. Returns audit scores and detailed metrics.
Extract structured data from a fully rendered web page using CSS selectors. The page is loaded in a real browser with JavaScript execution, then elements matching your selectors are extracted with their text content, inner HTML, attributes, and bounding box positions.
Generate a PDF from a web page or raw HTML. Navigate to a URL or render provided HTML in a headless browser and export it as a PDF. Supports page formatting options including paper size, margins, headers/footers, background printing, and landscape orientation. Returns the PDF as a base64-encoded string.
Capture a screenshot of a web page or rendered HTML. Supports full-page captures, custom viewports, clipping regions, and multiple image formats (PNG, JPEG, WebP). Returns the screenshot as a base64-encoded string.
Perform a web search and optionally scrape each result page. Returns structured search results with titles, URLs, and descriptions. When scraping is enabled, each result URL is fetched and processed into clean markdown, raw HTML, or other formats. Supports multiple sources (web, news, images), language filtering, time-based filtering, and category filters.
Access a protected web page by bypassing basic bot detection mechanisms. Returns the page content, cookies, and/or a screenshot after unblocking. Useful for sites that block standard headless browser requests. For sites with advanced protections or interactive CAPTCHAs, BrowserQL may be needed instead.
Execute custom JavaScript/Puppeteer code in a headless browser context. Browserless sets up a browser and page, then runs your code with access to the Puppeteer `page` object. Use this for multi-step browser interactions like navigating, filling forms, clicking buttons, and extracting data within a single request.