FeaturePOST /v1/extract

DOM Extraction

Named fields by selector, or ready-made inspect packs — raw data from the rendered DOM.

What you can do with extract and inspect

Pull specific fields via CSS selector or fetch a ready-made inspect pack (meta, headings, links, …). Ideal for light scraping, content checks, and agent inputs.

Named fields via selector (text, html, attribute, count, list, exists, similar, selector)
Adaptive relocation when classes are renamed (adaptive + match_text)
Inspect packs via include[] without writing your own selectors
XHR/fetch bodies via capture_xhr, optional fetch_mode=http without JS
Wait for selectors / cookie banners before extract
Structured data objects for CI and automation

API Call

editable

Request body

{'{'}

{'}'}

Extract: adjust selector and viewport.

Response

No result yet — run the API call.

Endpoints & parameters

Choose an endpoint — below you'll see an example request and explanations for all relevant parameters.

POST /v1/extract

Extracts named fields via CSS selectors.

Example request

json

{
  "url": "https://example.com",
  "width": 1280,
  "height": 720,
  "fields": [
    {
      "name": "title",
      "selector": "h1",
      "type": "text"
    },
    {
      "name": "cta_href",
      "selector": "a.primary",
      "type": "attribute",
      "attribute": "href"
    },
    {
      "name": "cards",
      "selector": "article.card",
      "type": "similar"
    },
    {
      "name": "cta_path",
      "selector": "a.buy",
      "type": "selector"
    }
  ],
  "wait_for_selector": "h1",
  "dismiss_cookie_banner": true,
  "capture_xhr": "*/api/*"
}

Parameters

urlstring required
Target URL.
fieldsarray (≤50) required
Or template_id — each field: name, selector, type (+ attribute when type=attribute).
fields[].type'text'|'html'|'attribute'|'count'|'list'|'exists'|'similar'|'selector'
How the selector is evaluated. similar = sibling cards, selector = generated CSS path.
fields[].adaptiveboolean
Re-resolve selector when it no longer matches.
fields[].match_textstring
Text hint for adaptive relocation.
template_idstring
Alternative to fields: saved extract template.
width / heightnumber
Viewport.
wait_for / wait_timeout_msvarious
Wait for selector before extract.
dismiss_cookie_bannerboolean
Attempts to dismiss known cookie banners before measuring or checking.
wait_for_selectorstring
Waits for a CSS selector before continuing.
wait_ms / settle_msnumber
Additional wait time after loading (depends on endpoint).
devicestring
Playwright device preset; overrides width/height/UA when set.
color_scheme'light'|'dark'|'no-preference'
Preferred color scheme in the browser.
block_adsboolean
Blocks known ad and tracker requests.
capture_xhrstring
Glob or substring for XHR/fetch URLs whose bodies Scout captures (max. 20).
fetch_mode'browser'|'http'
http loads without JavaScript (static document). Default: browser.
solve_captchaboolean
Best-effort Turnstile/checkbox on your own sites. Cloud: domain verify. Self-host: always allowed.
record_harboolean
Optionally records HAR (where supported by the endpoint).
respect_robotsboolean
Respect robots.txt (default: on).