Scrape

Fetches the given URL and returns comprehensive data in a single response.

Response includes:

• markdown: Cleaned, readable markdown content
• html: Cleaned HTML (main content only)
• rawHtml: Original unprocessed HTML
• metadata: Page metadata (title, description, og:*, author, dates, etc.)
• links: Internal and external links found on the page
• stats: Processing time statistics

Options:

• onlyMainContent: Extract only main content (default: true)
• contentSelector: Custom CSS selector for content extraction

Browser Rendering: Use waitFor (selector or ms) or actions to force browser rendering.

Actions (Browser Automation):

• wait: Wait for milliseconds { type: "wait", milliseconds: 2000 }
• click: Click element { type: "click", selector: "#btn" }
• type: Type text { type: "type", selector: "#input", text: "hello" }
• press: Press key { type: "press", key: "Enter" }
• scroll: Scroll { type: "scroll", direction: "down" }
• evaluate: Execute JS { type: "evaluate", script: "..." }

Features:

• Smart SPA/Cloudflare detection with automatic browser fallback
• Extracts embedded data from Next.js, Nuxt.js, Notion pages
• Full pipeline: fetch → clean → extract → convert → postprocess

Costs 1 credit.