Skip to content
Skryp

Read any page as clean Markdown or page data

Scrape returns a page's main content as lean Markdown, plus the data the page already publishes (JSON-LD, product data, metadata) at no model cost. Long pages stay on the server behind a reference your agent reads in sections, so its context holds only what it asked for.

Try it: convert a page to Markdown

  • Formats: Markdown, clean HTML, raw HTML, text, links, images, metadata, page data, screenshot and network calls.
  • Focus: say what you need (“shipping costs”) and get only those sections.
  • Up to 50 URLs per call, fetched in parallel.
  • Browser steps when a page needs them: click, type, press, scroll, wait.

Limits

  • Pages behind a sign-in need a saved login: your own account, connected once in the dashboard.
  • PDFs are converted to text; scanned images inside them are not read.

Price

1 credit for a standard request, 5 through a residential proxy (10 with JavaScript rendering). Advanced unblocking costs 25, charged only on success. Failed pages are free. Full rate card

Start freeRead the quickstart
Scrape a page · REST
curl -X POST https://api.skryp.dev/v1/scrape \
  -H "Authorization: Bearer $SKRYP_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "url": "https://shop.example/kettle",
  "formats": [
    "markdown",
    "structured_data"
  ]
}'
Scrape a page · MCP tool call
{
  "tool": "scrape",
  "arguments": {
    "url": "https://shop.example/kettle",
    "formats": [
      "markdown",
      "structured_data"
    ]
  }
}

Related