Scraping CAPTCHA-Protected Websites

Most high-value websites use CAPTCHAs as part of their anti-bot defense. This guide covers strategies for scraping these sites reliably using CaptchaAI, including how to identify CAPTCHA types, solve them automatically, and build resilient scrapers.

Common CAPTCHA Implementations

Strategy 1: Detect and Solve on Demand

The most reliable approach — scrape normally and solve CAPTCHAs only when they appear:

import requests
import time
from bs4 import BeautifulSoup

API_KEY = "YOUR_API_KEY"

class ProtectedScraper:
    def __init__(self):
        self.session = requests.Session()
        self.session.headers.update({
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"
        })

    def scrape(self, url):
        resp = self.session.get(url)

        # Check for CAPTCHA
        if self._has_captcha(resp.text):
            resp = self._handle_captcha(resp.text, url)

        return resp.text

    def _has_captcha(self, html):
        indicators = ["g-recaptcha", "cf-turnstile", "h-captcha", "captcha"]
        return any(ind in html.lower() for ind in indicators)

    def _handle_captcha(self, html, url):
        soup = BeautifulSoup(html, "html.parser")

        # reCAPTCHA v2
        rc = soup.find("div", class_="g-recaptcha")
        if rc:
            token = self._solve_recaptcha(rc["data-sitekey"], url)
            return self.session.post(url, data={"g-recaptcha-response": token})

        # Cloudflare Turnstile
        ts = soup.find("div", class_="cf-turnstile")
        if ts:
            token = self._solve_turnstile(ts["data-sitekey"], url)
            return self.session.post(url, data={"cf-turnstile-response": token})

        raise Exception("Unknown CAPTCHA type")

    def _solve_recaptcha(self, site_key, page_url):
        resp = requests.get("https://ocr.captchaai.com/in.php", params={
            "key": API_KEY, "method": "userrecaptcha",
            "googlekey": site_key, "pageurl": page_url
        })
        return self._poll(resp.text.split("|")[1])

    def _solve_turnstile(self, site_key, page_url):
        resp = requests.get("https://ocr.captchaai.com/in.php", params={
            "key": API_KEY, "method": "turnstile",
            "sitekey": site_key, "pageurl": page_url
        })
        return self._poll(resp.text.split("|")[1])

    def _poll(self, task_id):
        for _ in range(60):
            time.sleep(5)
            result = requests.get("https://ocr.captchaai.com/res.php", params={
                "key": API_KEY, "action": "get", "id": task_id
            })
            if result.text == "CAPCHA_NOT_READY": continue
            if result.text.startswith("OK|"): return result.text.split("|")[1]
            raise Exception(result.text)
        raise TimeoutError()

# Usage
scraper = ProtectedScraper()
html = scraper.scrape("https://example.com/data")

Strategy 2: Pre-Solve for Known CAPTCHA Pages

If you know which pages always have CAPTCHAs, solve preemptively:

def scrape_known_captcha_page(url, site_key):
    # Solve before even loading the page
    token = solve_recaptcha(site_key, url)

    # Submit directly with token
    resp = requests.post(url, data={
        "g-recaptcha-response": token,
        "query": "search term"
    })
    return resp.text

Strategy 3: Cloudflare-Protected Sites

Sites behind Cloudflare often require a cf_clearance cookie:

def get_cloudflare_clearance(url, proxy):
    resp = requests.get("https://ocr.captchaai.com/in.php", params={
        "key": API_KEY,
        "method": "cloudflare_challenge",
        "pageurl": url,
        "proxy": proxy,
        "proxytype": "HTTP"
    })
    task_id = resp.text.split("|")[1]

    for _ in range(60):
        time.sleep(5)
        result = requests.get("https://ocr.captchaai.com/res.php", params={
            "key": API_KEY, "action": "get", "id": task_id
        })
        if result.text == "CAPCHA_NOT_READY": continue
        if "cf_clearance" in result.text:
            # Parse cf_clearance and user_agent from response
            return result.text
    raise TimeoutError()

Multi-Page Scraping Pattern

def scrape_multiple_pages(base_url, site_key, pages):
    scraper = ProtectedScraper()
    results = []

    for page in pages:
        url = f"{base_url}?page={page}"
        try:
            html = scraper.scrape(url)
            soup = BeautifulSoup(html, "html.parser")
            items = soup.find_all("div", class_="item")
            results.extend([item.text.strip() for item in items])
            print(f"Page {page}: {len(items)} items")
        except Exception as e:
            print(f"Page {page} failed: {e}")

        time.sleep(random.uniform(2, 5))

    return results

Troubleshooting

FAQ

Which sites are hardest to scrape?

Sites using Cloudflare Enterprise, PerimeterX, or Akamai Bot Manager are the most challenging. CaptchaAI handles their CAPTCHA components; combine with stealth browsers and proxies for best results.

Can I scrape sites that require login?

Yes. Log in first (solving any login CAPTCHA), maintain the session cookies, then scrape authenticated pages. CaptchaAI handles CAPTCHAs at any stage.

How do I handle JavaScript-rendered pages?

Use Selenium, Puppeteer, or Playwright to render JavaScript, then extract CAPTCHA parameters and solve via CaptchaAI. See Selenium CAPTCHA Handling.

Related Guides

• How to Handle CAPTCHA Challenges in Web Scraping Workflows
• Headless Browser CAPTCHA Issues
• CAPTCHA Detection in Scraping Explained