These repositories have been stored on GitHub, a platform that allows developers to manage and share their code, for several years. The attackers strived to make the repositories on GitHub appear legitimate to potential targets by using attractive project descriptions that have likely been generated with AI. If the code from these repositories was launched, the victim’s device would become infected with malware and could be remotely controlled by the attackers.
While the projects were written in multiple programming languages – Python, JavaScript, C, C++ and C# – the malicious payloads stored inside the infected projects had the same goal: to download other malicious components from an attacker-controlled GitHub repository and execute them. These components include a stealer that collects passwords, bank account information, saved credentials, cryptocurrency wallet data and browsing history, packs it into a .7z archive and uploads it to attackers via Telegram.
![]() |
Structure of the archive which the stealer sends to attackers |
“As code sharing platforms such as GitHub are used by millions of developers worldwide, threat actors will certainly continue using fake software as an infection lure in the future. For that reason, it is crucial to handle processing of third-party code very carefully. Before attempting to run such code or integrate it into an existing project, it is paramount to thoroughly check what actions are performed by it. This way, it will be very easy to spot fake projects and prevent malicious code placed in them from being used to compromise the development environment,” comments Georgy Kucherin, Security Researcher at Kaspersky GReAT.