Russian Threat Group Leverages Trojan Apps
The UAT-11795 threat actor is deploying the Starland RAT via trojanized installers to harvest credentials and crypto assets.
A financially motivated Russian operation, identified as UAT-11795, has launched a widespread campaign targeting users of common business and utility software. By distributing trojanized installers, the threat actor aims to compromise systems to exfiltrate sensitive data and cryptocurrency holdings through the use of a novel backdoor known as Starland RAT.
Tactical Distribution and Initial Access
The campaign has been active since at least June 2025. Researchers have observed the threat actor distributing malicious payloads hidden within installers for applications such as WebEx, Zoom, MobaXterm, DBeaver, and FaceIT. While the exact initial infection vector remains unconfirmed, security analysts suspect the ClickFix method is used to trick users into executing the malicious code.
Once the initial HTA file is executed, it fetches a trojanized NSIS installer. This package includes a Python loader masquerading as a text file named LICENSE.txt. This loader initiates the compromise by modifying the Windows Registry to ensure persistence before decrypting the primary RAT payload.
Capabilities of the Starland Implant
Upon infection, the malware performs several defensive evasion checks to ensure it is not running within a sandbox environment. It establishes long-term access by creating scheduled tasks and modifying the Windows Startup folder. The Starland RAT is designed for high-impact data theft and system control, gathering comprehensive information about the host.
- System environment data, including processor details, RAM, OS version, and installed antivirus software.
- Active Directory information, such as domain controller structure and specific victim privileges.
- Cryptocurrency wallet assets covering over 40 distinct desktop and browser-extension wallets.
The malware also supports advanced features, including the ability to execute shell commands, inject both 32-bit and 64-bit shellcode, and capture desktop screenshots. Depending on the architecture of the infected machine, the implant deploys secondary payloads like CastleStealer or Remcos RAT to further expand its operational reach.
C2 Framework and Redundancy
The threat actor utilizes a sophisticated command-and-control infrastructure. When the primary hardcoded C2 server is unreachable, the malware attempts to contact a fallback domain retrieved by querying a Polygon smart contract. Furthermore, the group employs an undocumented PowerShell framework dubbed WLDR.
This framework operates entirely in system memory and utilizes encrypted PBKDF2-SHA256 beaconing. To ensure the exclusivity of their control, the attackers bind payload delivery directly to the hardware identifier (HWID) of the specific victim machine.
Implications for System Security
The use of trojanized legitimate software highlights the ongoing risk posed by unofficial download channels. Because the malware is designed to target both enterprise domain privileges and personal financial assets, organizations must prioritize software verification. Security teams are advised to review the indicators of compromise detailed by Cisco Talos to detect potential infections within their environments. Relying exclusively on official vendor portals for software acquisition remains the primary defense against this type of supply-chain-style masquerading.
Sources
- BleepingComputer Original source
- Cisco Talos says Also reporting
- Remcos Also reporting
Continue Reading
LXD, MicroCloud flaw lets clients be impersonated
Canonical patched a maximum-severity LXD and MicroCloud bug that let attackers assume a trusted client's identity without its private key.
A hands-on way to gauge AI ransomware risk
OWASP volunteers released RAID-AI, a free manual self-assessment that maps AI controls to NIST and CISA frameworks.
AWS AgentCore Flaw Let One Prompt Hijack All Agents
Researchers say a single chat prompt could extract cloud credentials and let an attacker take over every AgentCore agent in an AWS account and region.