Cloudflare Scraper

Fetch one public web page and return its HTML, cookies, browser user agent, HTTP status, and timestamp. Verification pages and access blocks are reported as failures. Some sites cannot be read; a successful scrape is not guaranteed.

Input

url is required and must be an HTTP or HTTPS URL. Existing proxy and cookies inputs remain supported:

{
  "url": "https://www.capterra.com/",
  "cookies": [{ "name": "session", "value": "example" }],
  "proxy": { "url": "http://proxy.example.com:8080", "username": "user", "password": "pass" }
}

Proxies are supplied by default. A custom proxy overrides the default. Cookie domain, path, and secure values are filled in when omitted. Only GET is supported. Earlier versions ignored method, body, and headers; unsupported values now produce an explicit error.

Results

A successful run writes one item to the default dataset and the same item to the default key-value store as OUTPUT:

{
  "url": "https://www.capterra.com/",
  "content": "<html>...</html>",
  "cookies": [],
  "userAgent": "Mozilla/5.0...",
  "status": 200,
  "success": true,
  "timestamp": "2026-09-30T12:00:00.000Z"
}

An unsuccessful run has status FAILED, an empty default dataset, and an OUTPUT record containing url, error, success: false, and timestamp. Unlike earlier versions, failed pages are not dataset items, so they are not billed as successful pages.

Billing

When Pay-Per-Event pricing is enabled, each successfully returned page is one automatic apify-default-dataset-item event. The minimum is one successful page per run; no extra minimum event is needed. Failed runs produce no paid page event. The actor stops before requesting a page when the maximum charge cannot cover another item. See the actor's Pricing tab for the active price; this code update does not change pricing.

Operator setup

Set FLARESOLVERR_URLS to a comma-separated list of solver URLs. Set PROXY_SERVER, PROXY_USERNAME, and PROXY_PASSWORD for the included proxy. Store credentials as secret actor environment variables. Requests try the configured instances in rotating order, with a 50-second attempt limit and a 150-second overall limit.

Run bun install, bun run build, and bun test. Start locally with apify run or bun run start and an input in the default key-value store.