18925617_605

Banking Trojan Spreading Through WhatsApp in Brazil

A new and highly advanced banking Trojan, dubbed Maverick, has surfaced in Brazil, exploiting WhatsApp as its main vector to compromise thousands of unsuspecting users. First detected in mid-October 2025, Maverick has already triggered over 62,000 blocked infection attempts within just the first ten days, signaling a significant threat to Brazilian internet users

Maverick targets Brazilian users by sending Portuguese-language WhatsApp messages that include malicious ZIP archives. These archives cleverly bypass WhatsApp’s security filters and contain a weaponized .LNK file, a Windows shortcut file that acts as the Trojan’s entry point.

The infection process begins when victims open these ZIP files, often disguised as bank notifications or important documents. The .LNK file then launches a chain of commands through cmd.exe and PowerShell, connecting stealthily to command-and-control (C2) servers. These servers validate the malware with stringent authentication protocols before downloading additional payloads.

What makes Maverick especially dangerous is that it operates in a fully fileless manner, all malicious components run directly in the system’s memory without writing files to the disk. This approach makes detection by traditional antivirus tools extremely difficult.

An Evolution in Malware Development: AI-Assisted Coding

Researchers at Securelist have found that Maverick shares significant code overlap with Coyote, a Brazilian banking Trojan documented in 2024. However, Maverick is more sophisticated, notably due to its use of artificial intelligence during its development. AI techniques are leveraged particularly for decrypting security certificates and optimizing the malware’s code-writing process.

This marks a worrying trend where cybercriminals are integrating AI tools to create more potent and evasive malware, raising the stakes in the cybersecurity battle.

Confirming the Victim’s Location

Maverick includes stringent geographic targeting measures to avoid detection and ensure attacks focus solely on Brazilian users. The malware checks:

  • System timezone
  • System language
  • Region settings
  • Date format

If any of these indicators do not confirm a Brazilian environment, the malware immediately terminates, preventing analysis by security researchers outside Brazil.

Spyware Capabilities and Data Theft

Once activated, Maverick unleashes an arsenal of surveillance tools:

  • Capturing screenshots
  • Monitoring browsers
  • Logging keystrokes
  • Controlling the mouse
  • Displaying overlay phishing pages

These tactics target credentials from 26 Brazilian banks, six cryptocurrency exchanges, and one payment platform, aiming to steal sensitive financial information and take over accounts.

Self-Propagation Through WhatsApp Account Hijacking

Perhaps the most alarming feature of Maverick is its ability to self-propagate by hijacking infected users’ WhatsApp accounts. Using WPPConnect, an open-source WhatsApp Web automation framework, the malware automatically sends malicious messages to all contacts in the victim’s list.

This worm-like behavior allows the Trojan to spread rapidly and exponentially, leveraging WhatsApp’s massive user base as a distribution network.

Advanced Command-and-Control Security

Maverick’s C2 infrastructure employs sophisticated security measures to avoid detection and tampering:

  • HMAC-256 signatures authenticate each request with a hardcoded secret key: "MaverickZapBot2025SecretKey12345".
  • Validation of User-Agent headers ensures that only genuine malware clients connect.
  • API endpoints deliver payloads encrypted as shellcodes using Donut loaders, with XOR encryption keys cleverly hidden within the payload’s final bytes.

The malware’s decryption method extracts the encryption key from the payload itself by reading the last four bytes to determine key size and then applying XOR operations to decrypt the entire code. Additionally, heavy code obfuscation techniques, such as Control Flow Flattening, make reverse engineering and analysis extremely challenging.

building-Cisco-Systems

Rootkits hitting unpatched Cisco switches

Security teams got a rude reminder this month where older, unpatched Cisco switches are being actively targeted and backdoored with a Linux rootkit in a campaign Trend Micro has named Operation ZeroDisco. If you manage campus or branch network gear, especially older 9400/9300/3750G-series hardware, read this now — the attackers are using a recent SNMP zero‑day plus a reused Telnet exploit to get persistent, stealthy access.

What the attackers are exploiting

Vulnerability: CVE‑2025‑20352 (CVSS 7.7), a stack‑overflow bug in SNMP on Cisco IOS / IOS XE. Cisco released a patch in late September after confirming in‑the‑wild exploitation.

Secondary exploit: A modified exploit for CVE‑2017‑3881 (a Telnet RCE) that the adversary uses to read/write memory on affected devices.

Targets: Older Linux‑based devices (Trend Micro observed Cisco 9400, 9300 and legacy 3750G models).

Campaign name: Operation ZeroDisco, the malware sets a universal password containing the text “disco” (one-letter change from “Cisco”), hence the name.

How the attack works (high level)

On 32‑bit devices the attackers send malicious SNMP packets to execute commands and use the Telnet exploit to obtain arbitrary memory read/write.

On 64‑bit devices they deploy a rootkit via the SNMP bug, set the universal “disco” password in memory, then log in and install a fileless backdoor. They can also connect different VLANs to move laterally.

The rootkit monitors UDP packets (even to closed ports) so specific packets can trigger backdoor functionality. It also tampers with IOSd memory to:

  • install the universal password across many auth methods,
  • hide running‑config items in memory,
  • bypass ACLs applied to VTY,
  • disable or tamper with log history,
  • reset running‑config timestamps to conceal changes.

Why this is bad

This isn’t just a noisy DoS exploit it’s a stealthy, persistent compromise that actively hides from blue teams. Because the malware modifies device memory and running config in ways that don’t always show in persistent storage, standard checks can miss it. Trend Micro warns there’s currently no reliable universal automated tool to detect ZeroDisco infections across switches.