The financially motivated threat actor UNC4990 has been identified leveraging weaponized USB devices as the initial infection vector in a targeted campaign against various sectors in Italy, including health, transportation, construction, and logistics. According to a report by Mandiant, a subsidiary of Google, UNC4990’s operations involve widespread USB infections, followed by the deployment of the EMPTYSPACE downloader. The threat actor employs a modular and adaptable approach, utilizing third-party websites like GitHub, Vimeo, and Ars Technica to host encoded additional stages, which are downloaded and decoded via PowerShell early in the execution chain.
Operating since late 2020, UNC4990 is assessed to be based in Italy, given its extensive use of Italian infrastructure for command-and-control (C2) purposes. The ultimate goal of UNC4990 remains unclear, although one instance involved the deployment of an open-source cryptocurrency miner after months of beaconing activity. Mandiant’s report sheds light on the multifaceted nature of the attack, with details previously documented by Fortgale and Yoroi in December 2023, tracking the adversary under the name Nebula Broker. The infection process begins with the victim double-clicking on a malicious LNK shortcut file on a USB device, initiating the execution of a PowerShell script responsible for downloading the EMPTYSPACE downloader from a remote server via an intermediate PowerShell script hosted on Vimeo.
Further analysis reveals the use of popular platforms like Ars Technica, GitHub, GitLab, and Vimeo for hosting malicious payloads during the campaign. While Mandiant assures that the content hosted on these services posed no direct risk to everyday users, UNC4990’s backdoor, QUIETBOARD, proves to be a Python-based threat with extensive capabilities. QUIETBOARD can execute arbitrary commands, alter crypto wallet addresses, propagate malware to removable drives, take screenshots, and gather system information. The backdoor also exhibits modular expansion, allowing it to run independent Python modules such as coin miners and dynamically fetch and execute Python code from the C2 server, highlighting the threat actor’s experimental and adaptable approach in developing their toolset.
