cloudscraper

Cloudflare v3 challenges run in a sandboxed JavaScript environment and require specific handling. While cloudscraper supports v3 by default, it is recommended to use the js2py interpreter and increase the delay to allow for complex challenge execution.

To optimize for v3, use the following configuration:

• interpreter='js2py': The most compatible interpreter for v3.
• delay: Increase this value (e.g., 5) to allow more time for the challenge to solve.
• enable_stealth=True: Provides additional protection against v3 detection.

import cloudscraper

# Optimized configuration for v3 challenges
scraper = cloudscraper.create_scraper(
    interpreter='js2py',  # Recommended for v3 challenges
    delay=5,              # Allow more time for complex challenges
    debug=True            # Enable debug output to see v3 detection
)

response = scraper.get("https://example.com")
print(response.text)

For maximum compatibility with challenging websites, you can enable enable_stealth and provide stealth_options to mimic human behavior.

Available stealth_options keys:

• min_delay: Minimum delay between actions.
• max_delay: Maximum delay between actions.
• human_like_delays: Boolean to enable human-like timing.
• randomize_headers: Boolean to randomize request headers.
• browser_quirks: Boolean to enable browser-specific quirks.

import cloudscraper

# Advanced configuration for challenging websites
scraper = cloudscraper.create_scraper(
    # Challenge handling
    interpreter='js2py',        # Best compatibility for v3 challenges
    delay=5,                    # Extra time for complex challenges

    # Stealth mode
    enable_stealth=True,
    stealth_options={
        'min_delay': 2.0,
        'max_delay': 6.0,
        'human_like_delays': True,
        'randomize_headers': True,
        'browser_quirks': True
    },

    # Browser emulation
    browser='chrome',

    # Debug mode
    debug=True
)

response = scraper.get("https://example.com")

To prevent 403 errors caused by TLS fingerprinting, request throttling, or concurrent request conflicts, use the following configuration when calling cloudscraper.create_scraper().

Critical Settings for Stability:

• min_request_interval: Set to 2.0 or higher to prevent TLS blocking.
• max_concurrent_requests: Set to 1 to prevent concurrent TLS conflicts.
• rotate_tls_ciphers: Set to True to avoid detection via cipher suite patterns.
• auto_refresh_on_403: Set to True to enable automatic recovery from 403 errors.

import cloudscraper

# OPTIMAL CONFIGURATION for preventing your specific 403 issues
scraper = cloudscraper.create_scraper(
    debug=True,  # Enable for monitoring (disable in production)
    
    # 🔑 KEY SETTINGS to prevent 403 errors
    min_request_interval=2.0,      # CRITICAL: Prevents TLS blocking
    max_concurrent_requests=1,     # CRITICAL: Prevents concurrent conflicts
    rotate_tls_ciphers=True,       # CRITICAL: Avoids cipher detection
    
    # 🛡️ Enhanced protection
    auto_refresh_on_403=True,      # Auto-recover from 403 errors
    max_403_retries=3,             # Max retry attempts
    session_refresh_interval=1800, # Refresh session every 30 minutes
    
    # 🥷 Optimized stealth mode
    enable_stealth=True,
    stealth_options={
        'min_delay': 1.0,          # Reasonable delays
        'max_delay': 3.0,
        'human_like_delays': True,
        'randomize_headers': True,
        'browser_quirks': True
    }
)

response = scraper.get('https://your-target-site.com')

You can provide a list of proxies to rotating_proxies to rotate the IP address for each request. Use proxy_options to define the rotation strategy and ban time.

proxy_options keys:

• rotation_strategy: The method of rotation (e.g., 'smart').
• ban_time: Time in seconds to wait before reusing a proxy.

import cloudscraper

proxies = [
    'http://user:pass@proxy1.example.com:8080',
    'http://user:pass@proxy2.example.com:8080',
    'http://user:pass@proxy3.example.com:8080'
]

scraper = cloudscraper.create_scraper(
    # Proxy rotation
    rotating_proxies=proxies,
    proxy_options={
        'rotation_strategy': 'smart',
        'ban_time': 300
    },

    # v3 challenge support
    interpreter='js2py',
    delay=5,

    # Stealth mode
    enable_stealth=True
)

# Each request may use a different proxy
for i in range(5):
    response = scraper.get("https://example.com")
    print(f"Request {i+1}: {response.status_code}")