Screenshots

The screenshot action captures an image of the page. The simplest form is the screenshot: true shorthand, which navigates to the URL and captures a default screenshot.

curl https://request.usestring.ai/v1/fetch \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "url": "https://example.com", "screenshot": true }'

Options

For more control, use a screenshot action inside an actions array:

{
  "url": "https://example.com",
  "actions": [{ "type": "screenshot", "full_page": true, "quality": 80 }]
}

At most one screenshot action is allowed per request.

Response

The screenshot is returned as base64 in the JSON body, alongside the final page in data. data is HTML by default, or converted Markdown when the request sets format: "markdown":

{
  "finalUrl": "https://example.com",
  "data": "<!doctype html>…",
  "screenshot": {
    "format": "png",
    "base64": "iVBORw0KGgoAAAANSUhEUgAA…"
  }
}

Decode screenshot.base64 to get the image bytes. format is png or jpeg.

Markdown output

Set format: "markdown" to convert the final page in data while keeping the response envelope and screenshot fields in JSON. The default markdownMode: "full" retains the full cleaned page body, including navigation, sidebars, and footers. Set markdownMode: "readable" to remove boilerplate and reduce output size. mainContentOnly: true is also supported when page chrome should be removed explicitly.

{
  "url": "https://example.com",
  "format": "markdown",
  "markdownMode": "full",
  "screenshot": true
}

See Response formats for the full preservation behavior.