Skip to main content

How to solve reCAPTCHA v2 Enterprise with CaptchaAI

Solve Google's enterprise-grade reCAPTCHA v2 challenges with CaptchaAI

Irina Dobreva avatar
Written by Irina Dobreva
Updated this week

reCAPTCHA v2 Enterprise is Google’s advanced implementation of the “I’m not a robot” challenge, designed with enhanced security analytics and adaptive risk evaluation. CaptchaAI fully supports reCAPTCHA v2 Enterprise, handling its challenge flows with accurate configuration to ensure reliable solving across protected environments.

Step 1: Identify & Classify the reCAPTCHA

  1. Open browser DevTools (Network tab) and reload the page

  2. Locate the reCAPTCHA v2 anchor request:
    ​https://www.google.com/recaptcha/enterprise/anchor?...

    • enterprise/ indicates that this is a reCAPTCHA v2 Enterprise.

  3. Check for sa parameter:

    • Found: URL contains sa=VALUE (e.g., sa=LOGIN) → Submit Action. → Use the exact sa value (case-sensitive)

    • Missing: No sa parameter → Use Proxy. → Requires residential proxy + user agent handling

  4. Extract googlekey:

    • From anchor URL: k=6LdxxXXxAAAAAAcXxx...

    • OR from HTML: data-sitekey="6LdxxXXxAAAAAAcXxx..."

Locate the 'sa' (Action) and 'k' (googlekey) parameters using browser Developer Tools


Step 2: Submit to CaptchaAI API (in.php)

Complete request example

curl https://ocr.captchaai.com/in.php \   
-F "key=YOUR_API_KEY" \   
-F "method=userrecaptcha" \   
-F "googlekey=6LdxxXXxAAAAAAcXxxXxxX91xxxxxxxx8xxOx7A" \   
-F "pageurl=https://example.com/login" \   
-F "proxy=login:[email protected]:8080" \   
-F "proxytype=HTTP" \   
-F "enterprise=1" \   
-F "action=LOGIN" \   
-F "json=1"

Recommended Parameters:

  • json=1: Recommended for structured response & User Agent retrieval.
    ​

Conditional Parameters:

  • action: MUST match sa parameter value from the anchor request.

  • proxy: Residential proxy is recommended if no action found.

  • proxytype: Proxy type (HTTP, HTTPS, SOCKS4, SOCKS5.).

Success Response

OK|1234567890

Success Response (JSON)

{"status":1,"request":"1234567890"}

Step 3: Retrieve Solution (res.php)

Wait 15-20 seconds, then every 5 second poll:

Complete request example

curl https://ocr.captchaai.com/res.php \   
-F "key=YOUR_API_KEY" \   
-F "action=get" \   
-F "id=1234567890" \   
-F "json=1"

Success Response

OK|0cAFcWeA7UXUoUHu2tcAi...79z67O4M1OloFPQ

Success Response (JSON)

{     
"status": 1,     
"result": "0cAFcWeA7UXUoUHu2tcAi...79z67O4M1OloFPQ",     
"user_agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) 144.0.7559.96 Safari/537.36" 
}

Note: Always use json=1 with reCAPTCHA v2 Enterprise to get the UserAgent used at solving process

Step 4: Inject Solution

Note: The actual captcha implementation on each website may vary. Examples below shows a very common case.

HTTP Clients (Python Example):

import requests
# Your proxy
RESIDENTIAL_PROXY = "http://user:[email protected]:8080"  # Same residential proxy used during captcha solving

# Values obtained from CaptchaAI AFTER solving
CAPTCHA_TOKEN = "0cAFcWeA7UXUoUHu2tcAi...79z67O4M1OloFPQ"  # From CaptchaAI res.php response
USER_AGENT = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) 144.0.7559.96 Safari/537.36"  # MUST be exact value from API response

# Configure session with critical headers and proxy
session = requests.Session()
session.headers.update({
    "User-Agent": USER_AGENT,
    "Referer": "https://example.com/login",
    "Origin": "https://example.com"
})
session.proxies = {
    "http": RESIDENTIAL_PROXY,
    "https": RESIDENTIAL_PROXY
}

# Submit form WITH reCAPTCHA token
response = session.post(
    url="https://example.com/login",
    data={
        "email": "[email protected]",
        "password": "secure_password",
        "g-recaptcha-response": CAPTCHA_TOKEN  # Injected token
    },
    timeout=30,
    allow_redirects=True
)

Step 5: Verify Success

✔️ Page loaded / Form Submitted → Success!
❌ Access denied → Re-solve (cookie may be expired or proxy/UA mismatched)

HTTP Clients (Python Example):

# Verify success (adjust checks per target site)
if response.status_code == 200 and "dashboard" in response.url:
    print("Login successful! Session established.")
else:
    print(f"Submission returned status {response.status_code}")
    print(f"Response preview: {response.text[:200]}")

And voilà, we’ve successfully passed reCAPTCHA v2 Enterprise, handled verification, and confirmed a valid session, completing the full automation cycle, with CaptchaAI doing all the heavy lifting behind the scenes.

Tips

  • Always use json=1 to receive User Agent value.

  • Proxy consistency is non-negotiable.

  • Using the returned User Agent value is highly recommend.

Did this answer your question?