Evilnum APT Group Employs New Python RAT
The PyVil remote access Trojan enables attackers to exfiltrate data, perform keylogging, take screenshots, and deploy tools for credential theft.
The Evilnum advanced persistent threat (APT) group has adopted a new Python remote access Trojan (RAT) to target financial tech organizations with well-crafted spear-phishing attacks. This is one of several changes detected in the group's tactics over recent weeks, researchers report.
Evilnum first appeared in 2018. Since then, it has employed a variety of attack strategies against fintech companies. Most of its victims are located across the UK and EU, though some attacks have landed in Australia and Canada. Evilnum has been seen using attack components scripted in JavaScript and C#, as well as tools from malware-as-a-service provider Golden Chickens.
Cybereason researchers have noticed Evilnum using a new Python-scripted RAT they call PyVil, as well as changes in its infection chain, persistence, and infrastructure, to evade detection.
The APT group opts for highly targeted attacks over broad phishing campaigns. It often exploits Know Your Customer (KYC) regulations, which mandate businesses keep records with essential data belonging to each customer, as well as identify people who can act on a customer's behalf.
Evilnum often begins infections with emails delivering ZIP archives that contain LNK files disguised as personal documents -- for example, credit cards, driver's licenses, and utility bills. These are likely stolen and belong to real people, researchers note in a report on their findings.
When the LNK file is opened, it deploys the JavaScript Trojan, which replaces the LNK file with a real file and renders the process invisible to the victim. Evilnum has used six iterations of the JavaScript Trojan, all of which have minor differences but the same core functionalities: They upload and download files, steal cookies, collect antivirus information, and execute commands, among other tricks. Sometimes a C# Trojan with similar functionality is used instead.
Now, researchers say, Evilnum appears to be changing its process to better conceal its activity. During the infection stage, it instead uses modified versions of legitimate executables to try and remain undetected.
Instead of delivering four LNK files in a ZIP archive, it only archives one. This LNK file is disguised as a PDF with content like utility bills, credit card photos, and driver's license photos. When the LNK file is executed, a JavaScript file is written to disk and executed, replacing the LNK file with a PDF. This time, the JavaScript acts as a dropper and lacks command-and-control (C2) communication capabilities.
This marks the first stage in a new infection chain that ends with the delivery of the PyVil RAT, compiled with py2exe.
"This Python RAT has various capabilities," says Tom Fakterman, cybersecurity analyst with Cybereason. "[It's] able to download additional Python scripts … it has the ability to steal credentials, the ability to deliver more malware and executables, and run commands on the system."
It can also perform keylogging, take screenshots, and collect information such as the antivirus products on the machine, USB devices connected, and version of Chrome running.
This is a new Python RAT, and the code inside the executable is obfuscated with extra layers in order to prevent decompilation of the payload, Fakterman explains in a write-up of his findings. Because the RAT is hidden under all this code, it can perform functions while remaining hidden.
While analyzing PyVil, researchers noticed several times that the malware received new Python modules from the C2 server. This is a custom version of the LaZagne Project, which Evilnum has used in the past, and the script is written to dump passwords and collect cookie data to send to the C2. They note this can be used to grow attack functionality.
"This innovation in tactics and tools is what allowed the group to stay under the radar, and we expect to see more in the future as the Evilnum group's arsenal continues to grow," Fakterman states in his write-up.
About the Author
You May Also Like
The Dirt on ROT Data
Dec 18, 2024Securing Your Cloud Data Across the Attack Timeline
Jan 15, 2025The Artificial Future Trend Micro Security Predictions for 2025
Jan 16, 2025