This thesis provides a concise overview of obfuscation techniques and methods for evading EDR (Endpoint Detection and Response) or XDR (Extended Detection and Response) protection. It also explains the concepts of static and dynamic file analysis, as well as basic mechanisms for detecting malicious processes within the Windows environment. At the end of the thesis, a case study was conducted comparing four Command & Control frameworks (Hoaxshell, Metasploit, Powershell Empire, Havoc C2) using the VirusTotal tool to easily assess the effectiveness of static and dynamic evasion of EDR solutions. The conclusion of the paper is that, although EDR systems are relatively good at detecting malicious code based on static analysis, the ability to bypass such systems largely depends on the operator's capability to create custom programs and does not rely on ready-made solutions for conducting attacks, whether in the context of cybercrime or ethical hacking. Finally, the paper poses the hypothesis that the problem of detecting malicious code or bypassing defensive systems will never be a fully "solved" issue and raises the question of using artificial intelligence for assisted code obfuscation generation.