[VDTRIESTE24] Cracking the Code: Decoding Anti-Bot Systems! - Conference by Fabien Vauchelles

Code Obfuscation and Anti-Bot Systems

• Obfuscation is the art of rendering code incomprehensible
• Code flow confusions: using complex conditions and nested loops to hide code logic
• Obfuscation techniques: encryption, encoding, string manipulation, and control flow flattenings
• Anti-bot systems use machine learning and behavior analysis to detect and block bots
• Key signals: browser stack, protocol headers, HTTP requests, and user behavior
• Browser stack analysis: IP address, device architecture, browser version, and GPU information
• Protocol headers: TCP/IP, TLS, and HTTP/2
• HTTP requests: formatting and encryption methods
• User behavior: navigation patterns, sequence of webpage visits, and time delay between requests
• Anti-bot systems use reinforcement learning to adjust and improve their detection methods
• Commercial solutions: available for website owners to detect and block bots
• Proxy networks: used to mask IP addresses and evade detection
• Data center proxies: virtual machines running on cloud providers
• Residential proxies: users sharing their real mobile devices or computers
• Obfuscation is used to prevent code from being deobfuscated or analyzed
• Code analysis: converting by-code to source code, using disassemblers and debuggers
• Reverse engineering: decompiling binary files to extract source code
• Start-of-art techniques: asymmetric encryptions, RSA, and DSA
• Next-generation proxy solutions: using machine learning and AI to detect and block bots