The leadership structure, commonly referred to as the “dual-hat” arrangement, assigns a single individual to oversee both organizations.
The post Senate Confirms Joshua Rudd to Lead NSA and US Cyber Command appeared first on SecurityWeek.
Stryker was targeted by the Handala group, which claims to have wiped more than 200,000 of the company’s devices.
The post MedTech Giant Stryker Crippled by Iran-Linked Hacker Attack appeared first on SecurityWeek.
Google has completed its $32 billion acquisition of the cloud security giant, which will maintain its brand.
The post Wiz Joins Google Cloud as Landmark Acquisition Closes appeared first on SecurityWeek.
Cardwell started her career at Netscape, become a VP of engineering at American Express, CISO at UnitedHealth Group, and now CISO in Residence at Transcend.
The post CISO Conversations: Aimee Cardwell appeared first on SecurityWeek.
Hackers stole personal information such as names, Social Security numbers, and driver’s license numbers.
The post 238,000 Impacted by Bell Ambulance Data Breach appeared first on SecurityWeek.
The company connects AI agents to security data lakes for interactive investigations, detection engineering, and autonomous response.
The post Scanner Raises $22 Million for AI-Powered Threat Hunting appeared first on SecurityWeek.
Promptfoo has raised more than $23 million in funding for a platform that helps developers secure LLMs and AI agents.
The post OpenAI to Acquire AI Security Startup Promptfoo appeared first on SecurityWeek.
The bugs could lead to arbitrary code execution, privilege escalation, or authentication rate-limit bypass.
The post Fortinet, Ivanti, Intel Patch High-Severity Vulnerabilities appeared first on SecurityWeek.
The evolution of vulnerability management in the agentic era is characterized by continuous telemetry, contextual prioritization and the ultimate goal of agentic remediation.
The post How to 10x Your Vulnerability Management Program in the Agentic Era appeared first on SecurityWeek.
The cybercriminals have leaked more than 300GB of files allegedly stolen from the tire giant.
The post Michelin Confirms Data Breach Linked to Oracle EBS Attack appeared first on SecurityWeek.
Recently, we uncovered BeatBanker, an Android‑based malware campaign targeting Brazil. It spreads primarily through phishing attacks via a website disguised as the Google Play Store. To achieve their goals, the malicious APKs carry multiple components, including a cryptocurrency miner and a banking Trojan capable of completely hijacking the device and spoofing screens, among other things. In a more recent campaign, the attackers switched from the banker to a known RAT.
This blog post outlines each phase of the malware’s activity on the victim’s handset, explains how it ensures long‑term persistence, and describes its communication with mining pools.
The campaign begins with a counterfeit website, cupomgratisfood[.]shop, that looks exactly like the Google Play Store. This fake app store contains the “INSS Reembolso” app, which is in fact a Trojan. There are also other apps that are most likely Trojans too, but we haven’t obtained them.
The INSS Reembolso app poses as the official mobile portal of Brazil’s Instituto Nacional do Seguro Social (INSS), a government service that citizens can use to perform more than 90 social security tasks, from retirement applications and medical exam scheduling to viewing CNIS (National Registry of Social Information), tax, and payment statements, as well as tracking request statuses. By masquerading as this trusted platform, the fake page tricks users into downloading the malicious APK.
The initial APK file is packed and makes use of a native shared library (ELF) named libludwwiuh.so that is included in the application. Its main task is to decrypt another ELF file that will ultimately load the original DEX file.
First, libludwwiuh.so decrypts an embedded encrypted ELF file and drops it to a temporary location on the device under the name l.so. The same code that loaded the libludwwiuh.so library then loads this file, which uses the Java Native Interface (JNI) to continue execution.
The library does not have calls to its functions; instead, it directly calls the Java methods whose names are encrypted in the stack using XOR (stack strings technique) and restored at runtime:
Initially, the loader makes a request to collect some network information using https://ipapi.is to determine whether the infected device is a mobile device, if a VPN is being used, and to obtain the IP address and other details.
This loader is engineered to bypass mobile antivirus products by utilizing dalvik.system.InMemoryDexClassLoader. It loads malicious DEX code directly into memory, avoiding the creation of any files on the device’s file system. The necessary DEX files can be extracted using dynamic analysis tools like Frida.
Furthermore, the sample incorporates anti-analysis techniques, including runtime checks for emulated or analysis environments. When such an environment is detected (or when specific checks fail, such as verification of the supported CPU_ABI), the malware can immediately terminate its own process by invoking android.os.Process.killProcess(android.os.Process.myPid()), effectively self-destructing to hinder dynamic analysis.
After execution, the malware displays a user interface that mimics the Google Play Store page, showing an update available for the INSS Reembolso app. This is intended to trick victims into granting installation permissions by tapping the “Update” button, which allows the download of additional hidden malicious payloads.
The payload delivery process mimics the application update. The malware uses the REQUEST_INSTALL_PACKAGES permission to install APK files directly into its memory, bypassing Google Play. To ensure persistence, the malware keeps a notification about a system update pinned to the foreground and activates a foreground service with silent media playback, a tactic designed to prevent the operating system from terminating the malicious process.
When UPDATE is clicked on a fake Play Store screen, the malicious application downloads and executes an ELF file containing a cryptomining payload. It starts by issuing a GET request to the C2 server at either hxxps://accessor.fud2026.com/libmine-<arch>.so or hxxps://fud2026.com/libmine-<arch>.so. The downloaded file is then decrypted using CipherInputStream(), with the decryption key being derived from the SHA-1 hash of the downloaded file’s name, ensuring that each version of the file is encrypted with a unique key. The resulting file is renamed d-miner.
The decrypted payload is an ARM-compiled XMRig 6.17.0 binary. At runtime, it attempts to create a direct TCP connection to pool.fud2026[.]com:9000. If successful, it uses this endpoint; otherwise, it automatically switches to the proxy endpoint pool-proxy.fud2026[.]com:9000. The final command-line arguments passed to XMRig are as follows:
Login
-o pool.fud2026[.]com:9000 or pool-proxy.fud2026[.]com:9000 (selected dynamically)-k (keepalive)--tls (encrypted connection)--no-color (disable colored output)--nicehash (NiceHash protocol support)
The malware uses Google’s legitimate Firebase Cloud Messaging (FCM) as its primary command‑and‑control (C2) channel. In the analyzed sample, each FCM message received triggers a check of the battery status, temperature, installation date, and user presence. A hidden cryptocurrency miner is then started or stopped as needed. These mechanisms ensure that infected devices remain permanently accessible and responsive to the attacker’s instructions, which are sent through the FCM infrastructure. The attacker monitors the following information:
The KeepAliveServiceMediaPlayback component ensures continuous operation by initiating uninterrupted playback via MediaPlayer. It keeps the service active in the foreground using a notification and loads a small, continuous audio file. This constant activity prevents the system from suspending or terminating the process due to inactivity.
The identified audio output8.mp3 is five seconds long and plays on a loop. It contains some Chinese words.
BeatBanker compromises the machine with a cryptocurrency miner and introduces another malicious APK that acts as a banking Trojan. This Trojan uses previously obtained permission to install an additional APK called INSS Reebolso, which is associated with the package com.destination.cosmetics.
Similar to the initial malicious APK, it establishes persistence by creating and displaying a fixed notification in the foreground to hinder removal. Furthermore, BeatBanker attempts to trick the user into granting accessibility permissions to the package.
Leveraging the acquired accessibility permissions, the malware establishes comprehensive control over the device’s user interface.
The Trojan constantly monitors the foreground application. It targets the official Binance application (com.binance.dev) and the Trust Wallet application (com.wallet.crypto.trustapp), focusing on USDT transactions. When a user tries to withdraw USDT, the Trojan instantly overlays the target app’s transaction confirmation screen with a highly realistic page sourced from Base64-encoded HTML stored in the banking module.
The module captures the original withdrawal address and amount, then surreptitiously substitutes the destination address with an attacker-controlled one using AccessibilityNodeInfo.ACTION_SET_TEXT. The overlay page shows the victim the address they copied (for Binance) or just shows a loading icon (for Trust Wallet), leading them to believe they are remitting funds to the intended wallet when, in fact, the cryptocurrency is transferred to the attacker’s designated address.
Fake overlay pages: Binance (left) and Trust Wallet (right)
BeatBanker’s banking module monitors the following browsers installed on the victim’s device:
Its aim is to collect the URLs accessed by the victim using the regular expression ^(?:https?://)?(?:[^:/\\\\]+\\\\.)?([^:/\\\\]+\\\\.[^:/\\\\]+). It also offers management functionalities (add, edit, delete, list) for links saved in the device’s default browser, as well as the ability to open links provided by the attacker.
BeatBanker is also designed to receive commands from the C2. These commands aim to collect the victim’s personal information and gain complete control of the device.
| Command | Description |
| 0 | Starts dynamic loading of the DEX class |
| Update | Simulates software update and locks the screen |
| msg: | Displays a Toast message with the provided text |
| goauth<*> | Opens Google Authenticator (if installed) and enables the AccessService.SendGoogleAuth flag used to monitor and retrieve authentication codes |
| kill<*> | Sets the protection bypass flag AccessService.bypass to “True”and sets the initializeService.uninstall flag to “Off” |
| srec<*> | Starts or stops audio recording (microphone), storing the recorded data in a file with an automatically generated filename. The following path format is used to store the recording: /Config/sys/apps/rc/<timestamp>_0REC<last5digits>.wav |
| pst<*> | Pastes text from the clipboard (via Accessibility Services) |
| GRC<*> | Lists all existing audio recording files |
| gtrc<*> | Sends a specific audio recording file to the C2 |
| lcm<*> | Lists supported front camera resolutions |
| usdtress<*> | Sets a USDT cryptocurrency address when a transaction is detected |
| lnk<*> | Opens a link in the browser |
| EHP<*> | Updates login credentials (host, port, name) and restarts the application |
| ssms<*> | Sends an SMS message (individually or to all contacts) |
| CRD<*> | Adds (E>) or removes (D>) packages from the list of blocked/disabled applications |
| SFD<*> | Deletes files (logs, recordings, tones) or uninstalls itself |
| adm<>lck<> | Immediately locks the screen using Device Administrator permissions |
| adm<>wip<> | Performs a complete device data wipe (factory reset) |
| Aclk<*> | Executes a sequence of automatic taps (auto-clicker) or lists existing macros |
| KBO<*>lod | Checks the status of the keylogger and virtual keyboard |
| KBO<*>AKP/AKA | Requests permission to activate a custom virtual keyboard or activates one |
| KBO<*>ENB: | Enables (1) or disables (0) the keylogger |
| RPM<*>lod | Checks the status of all critical permissions |
| RPM<*>ACC | Requests Accessibility Services permission |
| RPM<*>DOZ | Requests Doze/App Standby permission (battery optimization) |
| RPM<*>DRW | Requests Draw Over Other Apps permission (overlay) |
| RPM<*>INST | Requests permission to install apps from unknown sources (Android 8+) |
| ussd<*> | Executes a USSD code (e.g., *#06# for IMEI) |
| Blkt<*> | Sets the text for the lock overlay |
| BLKV<*> | Enables or disables full-screen lock using WindowManager.LayoutParams.TYPE_APPLICATION_OVERLAY to display a black FrameLayout element over the entire screen |
| SCRD<> / SCRD2<> | Enables/disables real-time screen text submission to the C2 (screen reading) |
| rdall<*> | Clears or sends all keylogger logs |
| rdd<*> | Deletes a specific log file |
| rd<*> | Sends the content of a specific keylogger file |
| MO<*> | Manages application monitoring (add, remove, list, screenshot, etc.) |
| FW<*> | Controls VPN and firewall (status, block/allow apps, enable/disable) |
| noti<*> | Creates persistent and custom notifications |
| sp<*> | Executes a sequence of swipes/taps (gesture macro) |
| lodp<*> | Manages saved links in the internal browser (add, edit, delete, list) |
| scc: | Starts screen capture/streaming |
Our recent detection efforts uncovered a campaign leveraging a fraudulent StarLink application that we assess as being a new BeatBanker variant. The infection chain mirrored previous instances, employing identical persistence methods – specifically, looped audio and fixed notifications. Furthermore, this variant included a crypto miner similar to those seen previously. However, rather than deploying the banking module, it was observed distributing the BTMOB remote administration tool.
The BTMOB APK is highly obfuscated and contains a class responsible for configuration. Despite this, it’s possible to identify a parser used to define the application’s behavior on the device, as well as persistence features, such as protection against restart, deletion, lock reset, and the ability to perform real-time screen recording.
The simple decryption routine uses repetitive XOR between the encrypted data and a short key. It iterates through the encrypted text byte by byte, repeating the key from the beginning whenever it reaches the end. At each position, the sample XORs the encrypted byte with the corresponding byte of the key, overwriting the original. Ultimately, the modified byte array contains the original text, which is then converted to UTF-8 and returned as a string.
BTMOB is an Android remote administration tool that evolved from the CraxsRAT, CypherRAT, and SpySolr families. It provides full remote control of the victim’s device and is sold in a Malware-as-a-Service (MaaS) model. On July 26, 2025, a threat actor posted a screenshot of the BTMOB RAT in action on GitHub under the username “brmobrats”, along with a link to the website btmob[.]xyz. The website contains information about the BTMOB RAT, including its version history, features, and other relevant details. It also redirects to a Telegram contact. Cyfirma has already linked this account to CraxsRAT and CypherRAT.
Recently, a YouTube channel was created by a different threat actor that features videos demonstrating how to use the malware and facilitate its sale via Telegram.
We also saw the distribution and sale of leaked BTMOB source code on some dark web forums. This may suggest that the creator of BeatBanker acquired BTMOB from its original author or the source of the leak and is utilizing it as the final payload, replacing the banking module observed in the INSS Reebolso incident.
In terms of functionality, BTMOB maintains a set of intrusive capabilities, including: automatic granting of permissions, especially on Android 13–15 devices; use of a black FrameLayout overlay to hide system notifications similar to the one observed in the banking module; silent installation; persistent background execution; and mechanisms designed to capture screen lock credentials, including PINs, patterns, and passwords. The malware also provides access to front and rear cameras, captures keystrokes in real time, monitors GPS location, and constantly collects sensitive data. Together, these functionalities provide the operator with comprehensive remote control, persistent access, and extensive surveillance capabilities over compromised devices.
All variants of BeatBanker – those with the banking module and those with the BTMOB RAT – were detected on victims in Brazil. Some of the samples that deliver BTMOB appear to use WhatsApp to spread, as well as phishing pages.
BeatBanker is an excellent example of how mobile threats are becoming more sophisticated and multi-layered. Initially focused in Brazil, this Trojan operates a dual campaign, acting as a Monero cryptocurrency miner, discreetly draining your device’s battery life while also stealing banking credentials and tampering with cryptocurrency transactions. Moreover, the most recent version goes even further, substituting the banking module with a full-fledged BTMOB RAT.
The attackers have devised inventive tricks to maintain persistence. They keep the process alive by looping an almost inaudible audio track, which prevents the operating system from terminating it and allows BeatBanker to remain active for extended periods.
Furthermore, the threat demonstrates an obsession with staying hidden. It monitors device usage, battery level and temperature. It even uses Google’s legitimate system (FCM) to receive commands. The threat’s banking module is capable of overlaying Binance and Trust Wallet screens and diverting USDT funds to the criminals’ wallets before the victim even notices.
The lesson here is clear: distrust is your best defense. BeatBanker spreads through fake websites that mimic Google Play, disguising itself as trustworthy government applications. To protect yourself against threats like this, it is essential to:
Our solutions detect this threat as HEUR:Trojan-Dropper.AndroidOS.BeatBanker and HEUR:Trojan-Dropper.AndroidOS.Banker.*
Additional IoCs, TTPs and detection rules are available to customers of our Threat Intelligence Reporting service. For more details, contact us at crimewareintel@kaspersky.com.
Host-based (MD5 hashes)
F6C979198809E13859196B135D21E79B – INSS Reebolso
D3005BF1D52B40B0B72B3C3B1773336B – StarLink
Domains
cupomgratisfood[.]shop
fud2026[.]com
accessor.fud2026[.]com
pool.fud2026[.]com
pool-proxy.fud2026[.]com
aptabase.fud2026[.]com
aptabase.khwdji319[.]xyz
btmob[.]xyz
bt-mob[.]net
The fourth quarter of 2025 went down as one of the most intense periods on record for high-profile, critical vulnerability disclosures, hitting popular libraries and mainstream applications. Several of these vulnerabilities were picked up by attackers and exploited in the wild almost immediately.
In this report, we dive into the statistics on published vulnerabilities and exploits, as well as the known vulnerabilities leveraged with popular C2 frameworks throughout Q4 2025.
This section contains statistics on registered vulnerabilities. The data is taken from cve.org.
Let’s take a look at the number of registered CVEs for each month over the last five years, up to and including the end of 2025. As predicted in our last report, Q4 saw a higher number of registered vulnerabilities than the same period in 2024, and the year-end totals also cleared the bar set the previous year.
Total published vulnerabilities by month from 2021 through 2025 (download)
Now, let’s look at the number of new critical vulnerabilities (CVSS > 8.9) for that same period.
Total number of published critical vulnerabilities by month from 2021 to 2025< (download)
The graph shows that the volume of critical vulnerabilities remains quite substantial; however, in the second half of the year, we saw those numbers dip back down to levels seen in 2023. This was due to vulnerability churn: a handful of published security issues were revoked. The widespread adoption of secure development practices and the move toward safer languages also pushed those numbers down, though even that couldn’t stop the overall flood of vulnerabilities.
This section contains statistics on the use of exploits in Q4 2025. The data is based on open sources and our telemetry.
In Q4 2025, the most prevalent exploits targeted the exact same vulnerabilities that dominated the threat landscape throughout the rest of the year. These were exploits targeting Microsoft Office products with unpatched security flaws.
Kaspersky solutions detected the most exploits on the Windows platform for the following vulnerabilities:
The list has remained unchanged for years.
We also see that attackers continue to adapt exploits for directory traversal vulnerabilities (CWE-35) when unpacking archives in WinRAR. They are being heavily leveraged to gain initial access via malicious archives on the Windows operating system:
As in the previous quarter, we see a rise in the use of archiver exploits, with fresh vulnerabilities increasingly appearing in attacks.
Below are the exploit detection trends for Windows users over the last two years.
Dynamics of the number of Windows users encountering exploits, Q1 2024 – Q4 2025. The number of users who encountered exploits in Q1 2024 is taken as 100% (download)
The vulnerabilities listed here can be used to gain initial access to a vulnerable system. This highlights the critical importance of timely security updates for all affected software.
On Linux-based devices, the most frequently detected exploits targeted the following vulnerabilities:
Dynamics of the number of Linux users encountering exploits, Q1 2024 – Q4 2025. The number of users who encountered exploits in Q1 2024 is taken as 100% (download)
We are seeing a massive surge in Linux-based exploit attempts: in Q4, the number of affected users doubled compared to Q3. Our statistics show that the final quarter of the year accounted for more than half of all Linux exploit attacks recorded for the entire year. This surge is primarily driven by the rapidly growing number of Linux-based consumer devices. This trend naturally attracts the attention of threat actors, making the installation of security patches critically important.
The distribution of published exploits by software type in Q4 2025 largely mirrors the patterns observed in the previous quarter. The majority of exploits we investigate through our monitoring of public research, news, and PoCs continue to target vulnerabilities within operating systems.
Distribution of published exploits by platform, Q1 2025 (download)
Distribution of published exploits by platform, Q2 2025 (download)
Distribution of published exploits by platform, Q3 2025 (download)
Distribution of published exploits by platform, Q4 2025 (download)
In Q4 2025, no public exploits for Microsoft Office products emerged; the bulk of the vulnerabilities were issues discovered in system components. When calculating our statistics, we placed these in the OS category.
We analyzed which vulnerabilities were utilized in APT attacks during Q4 2025. The following rankings draw on our telemetry, research, and open-source data.
TOP 10 vulnerabilities exploited in APT attacks, Q4 2025 (download)
In Q4 2025, APT attacks most frequently exploited fresh vulnerabilities published within the last six months. We believe that these CVEs will remain favorites among attackers for a long time, as fixing them may require significant structural changes to the vulnerable applications or the user’s system. Often, replacing or updating the affected components requires a significant amount of resources. Consequently, the probability of an attack through such vulnerabilities may persist. Some of these new vulnerabilities are likely to become frequent tools for lateral movement within user infrastructure, as the corresponding security flaws have been discovered in network services that are accessible without authentication. This heavy exploitation of very recently registered vulnerabilities highlights the ability of threat actors to rapidly implement new techniques and adapt old ones for their attacks. Therefore, we strongly recommend applying the security patches provided by vendors.
In this section, we will look at the most popular C2 frameworks used by threat actors and analyze the vulnerabilities whose exploits interacted with C2 agents in APT attacks.
The chart below shows the frequency of known C2 framework usage in attacks against users during Q4 2025, according to open sources.
TOP 10 C2 frameworks used by APTs to compromise user systems in Q4 2025 (download)
Despite the significant footprints it can leave when used in its default configuration, Sliver continues to hold the top spot among the most common C2 frameworks in our Q4 2025 analysis. Mythic and Havoc were second and third, respectively. After reviewing open sources and analyzing malicious C2 agent samples that contained exploits, we found that the following vulnerabilities were used in APT attacks involving the C2 frameworks mentioned above:
The set of vulnerabilities described above suggests that attackers have been using them for initial access and early-stage maneuvers in vulnerable systems to create a springboard for deploying a C2 agent. The list of vulnerabilities includes both zero-days and well-known, established security issues.
This section highlights the most noteworthy vulnerabilities that were publicly disclosed in Q4 2025 and have a publicly available description.
We typically describe vulnerabilities affecting a specific application. CVE-2025-55182 stood out as an exception, as it was discovered in React, a library primarily used for building web applications. This means that exploiting the vulnerability could potentially disrupt a vast number of applications that rely on the library. The vulnerability itself lies in the interaction mechanism between the client and server components, which is built on sending serialized objects. If an attacker sends serialized data containing malicious functionality, they can execute JavaScript commands directly on the server, bypassing all client-side request validation. Technical details about this vulnerability and an example of how Kaspersky solutions detect it can be found in our article.
This vulnerability represents a data-handling flaw that occurs when retrieving information from a remote server: when executing the curl or Invoke-WebRequest command, Windows launches Internet Explorer in the background. This can lead to a cross-site scripting (XSS) attack.
This vulnerability reinforces the trend of exploiting security flaws found in file archivers. The core of CVE-2025-11001 lies in the incorrect handling of symbolic links. An attacker can craft an archive so that when it is extracted into an arbitrary directory, its contents end up in the location pointed to by a symbolic link. The likelihood of exploiting this vulnerability is significantly reduced because utilizing such functionality requires the user opening the archive to possess system administrator privileges.
This vulnerability was associated with a wave of misleading news reports claiming it was being used in real-world attacks against end users. This misconception stemmed from an error in the security bulletin.
The year 2025 saw a surge in high-profile vulnerabilities, several of which were significant enough to earn a unique nickname. This was the case with CVE-2025-49844, also known as RediShell, which was unveiled during a hacking competition. This vulnerability is a use-after-free issue related to how the load command functions within Lua interpreter scripts. To execute the attack, an attacker needs to prepare a malicious script and load it into the interpreter.
As with any named vulnerability, RediShell was immediately weaponized by threat actors and spammers, albeit in a somewhat unconventional manner. Because technical details were initially scarce following its disclosure, the internet was flooded with fake PoC exploits and scanners claiming to test for the vulnerability. In the best-case scenario, these tools were non-functional; in the worst, they infected the system. Notably, these fraudulent projects were frequently generated using LLMs. They followed a standardized template and often cross-referenced source code from other identical fake repositories.
Driver vulnerabilities are often discovered in legitimate third-party applications that have been part of the official OS distribution for a long time. Thus, CVE-2025-24990 has existed within code shipped by Microsoft throughout nearly the entire history of Windows. The vulnerable driver has been shipped since at least Windows 7 as a third-party driver for Agere Modem. According to Microsoft, this driver is no longer supported and, following the discovery of the flaw, was removed from the OS distribution entirely.
The vulnerability itself is straightforward: insecure handling of IOCTL codes leading to a null pointer dereference. Successful exploitation can lead to arbitrary command execution or a system crash resulting in a blue screen of death (BSOD) on modern systems.
CVE-2025-59287 represents a textbook case of insecure deserialization. Exploitation is possible without any form of authentication; due to its ease of use, this vulnerability rapidly gained traction among threat actors. Technical details and detection methodologies for our product suite have been covered in our previous advisories.
In Q4 2025, the rate of vulnerability registration has shown no signs of slowing down. Consequently, consistent monitoring and the timely application of security patches have become more critical than ever. To ensure resilient defense, it is vital to regularly assess and remediate known vulnerabilities while implementing technology designed to mitigate the impact of potential exploits.
Continuous monitoring of infrastructure, including the network perimeter, allows for the timely identification of threats and prevents them from escalating. Effective security also demands tracking the current threat landscape and applying preventative measures to minimize risks associated with system flaws. Kaspersky Next serves as a reliable partner in this process, providing real-time identification and detailed mapping of vulnerabilities within the environment.
Securing the workplace remains a top priority. Protecting corporate devices requires the adoption of solutions capable of blocking malware and preventing it from spreading. Beyond basic measures, organizations should implement adaptive systems that allow for the rapid deployment of security updates and the automation of patch management workflows.
Starting from the third quarter of 2025, we have updated our statistical methodology based on the Kaspersky Security Network. These changes affect all sections of the report except for the installation package statistics, which remain unchanged.
To illustrate trends between reporting periods, we have recalculated the previous year’s data; consequently, these figures may differ significantly from previously published numbers. All subsequent reports will be generated using this new methodology, ensuring accurate data comparisons with the findings presented in this article.
Kaspersky Security Network (KSN) is a global network for analyzing anonymized threat intelligence, voluntarily shared by Kaspersky users. The statistics in this report are based on KSN data unless explicitly stated otherwise.
According to Kaspersky Security Network, in 2025:
In 2025, cybercriminals launched an average of approximately 1.17 million attacks per month against mobile devices using malicious, advertising, or unwanted software. In total, Kaspersky solutions blocked 14,059,465 attacks throughout the year.
Attacks on Kaspersky mobile users in 2025 (download)
Beyond the malware mentioned in previous quarterly reports, 2025 saw the discovery of several other notable Trojans. Among these, in Q4 we uncovered the Keenadu preinstalled backdoor. This malware is integrated into device firmware during the manufacturing stage. The malicious code is injected into libandroid_runtime.so – a core library for the Android Java runtime environment – allowing a copy of the backdoor to enter the address space of every app running on the device. Depending on the specific app, the malware can then perform actions such as inflating ad views, displaying banners on behalf of other apps, or hijacking search queries. The functionality of Keenadu is virtually unlimited, as its malicious modules are downloaded dynamically and can be updated remotely.
Cybersecurity researchers also identified the Kimwolf IoT botnet, which specifically targets Android TV boxes. Infected devices are capable of launching DDoS attacks, operating as reverse proxies, and executing malicious commands via a reverse shell. Subsequent analysis revealed that Kimwolf’s reverse proxy functionality was being leveraged by proxy providers to use compromised home devices as residential proxies.
Another notable discovery in 2025 was the LunaSpy Trojan.
LunaSpy Trojan, distributed under the guise of an antivirus app
Disguised as antivirus software, this spyware exfiltrates browser passwords, messaging app credentials, SMS messages, and call logs. Furthermore, it is capable of recording audio via the device’s microphone and capturing video through the camera. This threat primarily targeted users in Russia.
815,735 new unique installation packages were observed in 2025, showing a decrease compared to the previous year. While the decline in 2024 was less pronounced, this past year saw the figure drop by nearly one-third.
Detected Android-specific malware and unwanted software installation packages in 2022–2025 (download)
The overall decrease in detected packages is primarily due to a reduction in apps categorized as not-a-virus. Conversely, the number of Trojans has increased significantly, a trend clearly reflected in the distribution data below.
Distribution* of detected mobile software by type, 2024–2025 (download)
* The data for the previous year may differ from previously published data due to some verdicts being retrospectively revised.
A significant increase in Trojan-Banker and Trojan-Spy apps was accompanied by a decline in AdWare and RiskTool files. The most prevalent banking Trojans were Mamont (accounting for 49.8% of apps) and Creduz (22.5%). Leading the persistent adware category were MobiDash (39%), Adlo (27%), and HiddenAd (20%).
Share* of users attacked by each type of malware or unwanted software out of all users of Kaspersky mobile solutions attacked in 2024–2025 (download)
* The total may exceed 100% if the same users encountered multiple attack types.
Trojan-Banker malware saw a significant surge in 2025, not only in terms of unique file counts but also in the total number of attacks. Nevertheless, this category ranked fourth overall, trailing far behind the Trojan file category, which was dominated by various modifications of Triada and Fakemoney.
Note that the malware rankings below exclude riskware and potentially unwanted apps, such as RiskTool and adware.
| Verdict | % 2024* | % 2025* | Difference in p.p. | Change in ranking |
| Trojan.AndroidOS.Triada.fe | 0.04 | 9.84 | +9.80 | |
| Trojan.AndroidOS.Triada.gn | 2.94 | 8.14 | +5.21 | +6 |
| Trojan.AndroidOS.Fakemoney.v | 7.46 | 7.97 | +0.51 | +1 |
| DangerousObject.Multi.Generic | 7.73 | 5.83 | –1.91 | –2 |
| Trojan.AndroidOS.Triada.ii | 0.00 | 5.25 | +5.25 | |
| Trojan-Banker.AndroidOS.Mamont.da | 0.10 | 4.12 | +4.02 | |
| Trojan.AndroidOS.Triada.ga | 10.56 | 3.75 | –6.81 | –6 |
| Trojan-Banker.AndroidOS.Mamont.db | 0.01 | 3.53 | +3.51 | |
| Backdoor.AndroidOS.Triada.z | 0.00 | 2.79 | +2.79 | |
| Trojan-Banker.AndroidOS.Coper.c | 0.81 | 2.54 | +1.72 | +35 |
| Trojan-Clicker.AndroidOS.Agent.bh | 0.34 | 2.48 | +2.14 | +74 |
| Trojan-Dropper.Linux.Agent.gen | 1.82 | 2.37 | +0.55 | +4 |
| Trojan.AndroidOS.Boogr.gsh | 5.41 | 2.06 | –3.35 | –8 |
| DangerousObject.AndroidOS.GenericML | 2.42 | 1.97 | –0.45 | –3 |
| Trojan.AndroidOS.Triada.gs | 3.69 | 1.93 | –1.76 | –9 |
| Trojan-Downloader.AndroidOS.Agent.no | 0.00 | 1.87 | +1.87 | |
| Trojan.AndroidOS.Triada.hf | 0.00 | 1.75 | +1.75 | |
| Trojan-Banker.AndroidOS.Mamont.bc | 1.13 | 1.65 | +0.51 | +8 |
| Trojan.AndroidOS.Generic. | 2.13 | 1.47 | –0.66 | –6 |
| Trojan.AndroidOS.Triada.hy | 0.00 | 1.44 | +1.44 |
* Unique users who encountered this malware as a percentage of all attacked users of Kaspersky mobile solutions.
The list is largely dominated by the Triada family, which is distributed via malicious modifications of popular messaging apps. Another infection vector involves tricking victims into installing an official messaging app within a “customized virtual environment” that supposedly offers enhanced configuration options. Fakemoney scam applications, which promise fraudulent investment opportunities or fake payouts, continue to target users frequently, ranking third in our statistics. Meanwhile, the Mamont banking Trojan variants occupy the 6th, 8th, and 18th positions by number of attacks. The Triada backdoor preinstalled in the firmware of certain devices reached the 9th spot.
This section describes malware families whose attack campaigns are concentrated within specific countries.
| Verdict | Country* | %** |
| Trojan-Banker.AndroidOS.Coper.a | Türkiye | 95.74 |
| Trojan-Dropper.AndroidOS.Hqwar.bj | Türkiye | 94.96 |
| Trojan.AndroidOS.Thamera.bb | India | 94.71 |
| Trojan-Proxy.AndroidOS.Agent.q | Germany | 93.70 |
| Trojan-Banker.AndroidOS.Coper.c | Türkiye | 93.42 |
| Trojan-Banker.AndroidOS.Rewardsteal.lv | India | 92.44 |
| Trojan-Banker.AndroidOS.Rewardsteal.jp | India | 92.31 |
| Trojan-Banker.AndroidOS.Rewardsteal.ib | India | 91.91 |
| Trojan-Dropper.AndroidOS.Rewardsteal.h | India | 91.45 |
| Trojan-Banker.AndroidOS.Rewardsteal.nk | India | 90.98 |
| Trojan-Dropper.AndroidOS.Agent.sm | Türkiye | 90.34 |
| Trojan-Dropper.AndroidOS.Rewardsteal.ac | India | 89.38 |
| Trojan-Banker.AndroidOS.Rewardsteal.oa | India | 89.18 |
| Trojan-Banker.AndroidOS.Rewardsteal.ma | India | 88.58 |
| Trojan-Spy.AndroidOS.SmForw.ko | India | 88.48 |
| Trojan-Dropper.AndroidOS.Pylcasa.c | Brazil | 88.25 |
| Trojan-Dropper.AndroidOS.Hqwar.bf | Türkiye | 88.15 |
| Trojan-Banker.AndroidOS.Agent.pp | India | 87.85 |
* Country where the malware was most active.
** Unique users who encountered the malware in the indicated country as a percentage of all users of Kaspersky mobile solutions who were attacked by the same malware.
Türkiye saw the highest concentration of attacks from Coper banking Trojans and their associated Hqwar droppers. In India, Rewardsteal Trojans continued to proliferate, exfiltrating victims’ payment data under the guise of monetary giveaways. Additionally, India saw a resurgence of the Thamera Trojan, which we previously observed frequently attacking users in 2023. This malware hijacks the victim’s device to illicitly register social media accounts.
The Trojan-Proxy.AndroidOS.Agent.q campaign, concentrated in Germany, utilized a compromised third-party application designed for tracking discounts at a major German retail chain. Attackers monetized these infections through unauthorized use of the victims’ devices as residential proxies.
In Brazil, 2025 saw a concentration of Pylcasa Trojan attacks. This malware is primarily used to redirect users to phishing pages or illicit online casino sites.
The number of new banking Trojan installation packages surged to 255,090, representing a several-fold increase over previous years.
Mobile banking Trojan installation packages detected by Kaspersky in 2022–2025 (download)
Notably, the total number of attacks involving bankers grew by 1.5 times, maintaining the same growth rate seen in the previous year. Given the sharp spike in the number of unique malicious packages, we can conclude that these attacks yield significant profit for cybercriminals. This is further evidenced by the fact that threat actors continue to diversify their delivery channels and accelerate the production of new variants in an effort to evade detection by security solutions.
| Verdict | % 2024* | % 2025* | Difference in p.p. | Change in ranking |
| Trojan-Banker.AndroidOS.Mamont.da | 0.86 | 15.65 | +14.79 | +28 |
| Trojan-Banker.AndroidOS.Mamont.db | 0.12 | 13.41 | +13.29 | |
| Trojan-Banker.AndroidOS.Coper.c | 7.19 | 9.65 | +2.46 | +2 |
| Trojan-Banker.AndroidOS.Mamont.bc | 10.03 | 6.26 | –3.77 | –3 |
| Trojan-Banker.AndroidOS.Mamont.ev | 0.00 | 4.10 | +4.10 | |
| Trojan-Banker.AndroidOS.Coper.a | 9.04 | 4.00 | –5.04 | –4 |
| Trojan-Banker.AndroidOS.Mamont.ek | 0.00 | 3.73 | +3.73 | |
| Trojan-Banker.AndroidOS.Mamont.cb | 0.64 | 3.04 | +2.40 | +26 |
| Trojan-Banker.AndroidOS.Faketoken.pac | 2.17 | 2.95 | +0.77 | +5 |
| Trojan-Banker.AndroidOS.Mamont.hi | 0.00 | 2.75 | +2.75 |
* Unique users who encountered this malware as a percentage of all users of Kaspersky mobile solutions who encountered banking threats.
In 2025, we observed a massive surge in activity from Mamont banking Trojans. They accounted for approximately half of all new apps in their category and also were utilized in half of all banking Trojan attacks.
The year 2025 saw a continuing trend toward a decline in total unique unwanted software installation packages. However, we noted a significant year-over-year increase in specific threats – most notably mobile banking Trojans and spyware – even though adware remained the most frequently detected threat overall.
Among the mobile threats detected, we have seen an increased prevalence of preinstalled backdoors, such as Triada and Keenadu. Consistent with last year’s findings, certain mobile malware families continue to proliferate via official app stores. Finally, we have observed a growing interest among threat actors in leveraging compromised devices as proxies.
Security researchers have seen the vulnerabilities being exploited to deliver shells, conduct reconnaissance, and download malware.
The post Ivanti Exploitation Surges as Zero-Day Attacks Traced Back to July 2025 appeared first on SecurityWeek.
Tenga confirmed reports published by several outlets that the company notified customers of a data breach.
The Japanese manufacturer of adult products appears to have fallen victim to a phishing attack targeting one of its employees. Tenga reportedly wrote in the data breach notification:
“An unauthorized party gained access to the professional email account of one of our employees.”
This unauthorized access exposed the contents of said account’s inbox, potentially including customer names, email addresses, past correspondence, order details, and customer service inquiries.
In its official statement, Tenga said a “limited segment” of US customers who interacted with the company were impacted by the incident. Regarding the scope of the stolen data, it stated:
“The information involved was limited to customer email addresses and related correspondence history. No sensitive personal data, such as Social Security numbers, billing/credit card information, or TENGA/iroha Store passwords were jeopardized in this incident.”
From the wording of Tenga’s online statement, it seems the compromised account was used to send spam emails that included an attachment.
“Attachment Safety: We want to state clearly that there is no risk to your device or data if the suspicious attachment was not opened. The risk was limited to the potential execution of the attachment within the specific ‘spam’ window (February 12, 2026, between 12am and 1am PT).”
We reached out to Tenga about this “suspicious attachment” but have not heard back at the time of writing. We’ll keep you posted.
Tenga proactively contacted potentially affected customers. It advises them to change passwords and remain vigilant about any unusual activity. We would add that affected customers should be on the lookout for sextortion-themed phishing attempts.
If you think you have been affected by a data breach, here are steps you can take to protect yourself:
What do cybercriminals know about you?
Use Malwarebytes’ free Digital Footprint scan to see whether your personal information has been exposed online.
The UK is bracketing "intimate images shared without a victim's consent" along with terror and child sexual abuse material, and demanding that online platforms remove them within two days.…
Tech bros have been wanting to become immortal for years. Until they get there, their fallback might be continuing to post nonsense on social media from the afterlife.
On December 30, 2025, Meta was granted US patent 12513102B2: Simulation of a user of a social networking system using a language model. It describes a system that trains an AI on a user’s posts, comments, chats, voice messages, and likes, then deploys a bot to respond to newsfeeds, DMs, and even simulated audio or video calls.
Filed in November 2023 by Meta CTO Andrew Bosworth, it sounds innocuous enough. Perhaps some people would use it to post their political hot takes while they’re asleep.
Dig deeper, though, and the patent veers from absurd to creepy. It’s designed to be used not just from beyond the pillow but beyond the grave.
From the patent:
“The language model may be used for simulating the user when the user is absent from the social networking system, for example, when the user takes a long break or if the user is deceased.”
A Meta spokesperson told Business Insider that the company has no plans to act on the patent. And tech companies have a habit of laying claim to bizarre ideas that never materialize. But Facebook’s user numbers have stalled, and it presumably needs all the engagement it can get. We already know that the company loves the idea of AI ‘users’, having reportedly piloted them in late 2024, much to human users’ annoyance.
If the company ever did decide to pull the trigger on this technology, it would be a departure from its own memorialization policy, which preserves accounts without changes. One reason the company might not be willing to step over the line is that the world simply isn’t ready for AI conversations with the dead. Other companies have considered and even tested similar systems. Microsoft patented a chatbot that would allow you to talk to AI versions of deceased individuals in 2020; its own AI general manager called it disturbing, and it never went into production. Amazon demonstrated Alexa mimicking a dead grandmother’s voice from under a minute of audio in 2022, framing it as preserving memories. That never launched either.
Some projects that did ship left people wishing they hadn’t. Startup 2Wai’s avatar app originally offered the chance to preserve loved ones as AI avatars. Users called it “nightmare fuel” and “demonic”. The company seems to have pivoted to safer ground like social avatars and personal AI coaches now.
The other thing holding Meta back could be the legal questions. Unsurprisingly for such a new idea, there isn’t a uniform US framework on the use of AI to represent the dead. Several states recognize post-mortem right of publicity, although states like New York limit that to people whose voices and images have commercial value (typically meaning celebrities). California’s AB 1836 specifically targets AI-generated impersonations of the deceased, though.
Meta would also need to tiptoe carefully around the law in Europe. The company had to pause AI training on European users in 2024 under regulatory pressure, but then launched it anyway in March last year. Then it refused to sign the EU’s GPAI Code of Practice last July (the only major AI firm to do so). Meta’s relationship with EU regulators is strained at best.
Europe’s General Data Protection Regulation (GDPR) excludes deceased persons’ data, but Article 85 of the French Data Protection law lets anyone leave instructions about the retention, deletion and communication of their personal data after death. The EU AI Act’s Article 50 (fully applicable this August) will also require AI systems to disclose they are AI, with penalties up to €15 million or 3% of worldwide turnover for companies that don’t comply.
Hopefully Meta really will file this in the “just because we can do it doesn’t mean we should” drawer, and leave erstwhile social media sharers to rest in peace.
We don’t just report on threats – we help protect your social media
Cybersecurity risks should never spread beyond a headline. Protect your social media accounts by using Malwarebytes Identity Theft Protection.
Bork!Bork!Bork! Today's bork is entirely human-generated and will send a shiver down the spine of security pros. No matter how secure a system is, a user's ability to undo an administrator's best efforts should not be underestimated.…
In October 2025, we discovered a series of forum posts advertising a previously unknown stealer, dubbed “Arkanix Stealer” by its authors. It operated under a MaaS (malware-as-a-service) model, providing users not only with the implant but also with access to a control panel featuring configurable payloads and statistics. The set of implants included a publicly available browser post-exploitation tool known as ChromElevator, which was delivered by a native C++ version of the stealer. This version featured a wide range of capabilities, from collecting system information to stealing cryptocurrency wallet data. Alongside that, we have also discovered Python implementation of the stealer capable of dynamically modifying its configuration. The Python version was often packed, thus giving the adversary multiple methods for distributing their malware. It is also worth noting that Arkanix was rather a one-shot malicious campaign: at the time of writing this article, the affiliate program appears to be already taken down.
Kaspersky products detect this threat as Trojan-PSW.Win64.Coins.*, HEUR:Trojan-PSW.Multi.Disco.gen, Trojan.Python.Agent.*.
In October 2025, a series of posts was discovered on various dark web forums, advertising a stealer referred to by its author as “Arkanix Stealer”. These posts detail the features of the stealer and include a link to a Discord server, which serves as the primary communication channel between the author and the users of the stealer.
Example of an Arkanix Stealer advertisement
Upon further research utilizing public resources, we identified a set of implants associated with this stealer.
The initial infection vector remains unknown. However, based on some of the file names (such as steam_account_checker_pro_v1.py, discord_nitro_checker.py, and TikTokAccountBotter.exe) of the loader scripts we obtained, it can be concluded with high confidence that the initial infection vector involved phishing.
| MD5 | 208fa7e01f72a50334f3d7607f6b82bf |
| File name | discord_nitro_code_validator_right_aligned.py |
The Python loader is the script responsible for downloading and executing the Python-based version of the Arkanix infostealer. We have observed both plaintext Python scripts and those bundled using PyInstaller or Nuitka, all of which share a common execution vector and are slightly obfuscated. These scripts often serve as decoys, initially appearing to contain legitimate code. Some of them do have useful functionality, and others do nothing apart from loading the stealer. Additionally, we have encountered samples that employ no obfuscation at all, in which the infostealer is launched in a separate thread via Python’s built-in threading module.
Variants of Python loaders executing the next stage
Upon execution, the loader first installs the required packages — namely, requests, pycryptodome, and psutil — via the pip package manager, utilizing the subprocess module. On Microsoft Windows systems, the loader also installs pywin32. In some of the analyzed samples, this process is carried out twice. Since the loader does not perform any output validation of the module installation command, it proceeds to make a POST request to hxxps://arkanix[.]pw/api/session/create to register the current compromised machine on the panel with a predefined set of parameters even if the installation failed. After that, the stealer makes a GET request to hxxps://arkanix[.]pw/stealer.py and executes the downloaded payload.
| MD5 | af8fd03c1ec81811acf16d4182f3b5e1 |
| File name | – |
During our research, we obtained a sample of the Python implementation of the Arkanix stealer, which was downloaded from the endpoint hxxps://arkanix[.]pw/stealer.py by the previous stage.
The stealer’s capabilities — or features, as referred to by the author — in this version are configurable, with the default configuration predefined within the script file. To dynamically update the feature list, the stealer makes a GET request to hxxps://arkanix[.]pw/api/features/{payload_id}, indicating that these capabilities can be modified on the panel side. The feature list is identical to the one that was described in the GDATA report.
Configurable options
Prior to executing the information retrieval-related functions, the stealer makes a request to hxxps://arkanix[.]pw/upload_dropper.py, saves the response to %TEMP%\upd_{random 8-byte name}.py, and executes it. We do not have access to the contents of this script, which is referred to as the “dropper” by the attackers.
During its main information retrieval routine, at the end of each processing stage, the collected information is serialized into JSON format and saved to a predefined path, such as %LOCALAPPDATA\Arkanix_lol\%info_class%.json.
In the following, we will provide a more detailed description of the Python version’s data collection features.
Arkanix Stealer is capable of collecting a set of info about the compromised system. This info includes:
Information collection is performed using standard shell commands with the exception of the VPN check. The latter is implemented by querying the endpoint hxxps://ipapi[.]co/json/ and verifying whether the associated IP address belongs to a known set of VPNs, proxies, or Tor exit nodes.
This stealer is capable of extracting various types of data from supported browsers (22 in total, ranging from the widely popular Google Chrome to the Tor Browser). The list of supported browsers is hardcoded, and unlike other parameters, it cannot be modified during execution. In addition to a separate Chrome grabber module (which we’ll discuss later), the stealer itself supports the extraction of diverse information, such as:
All information is decrypted using either the Windows DPAPI or AES, where applicable, and searched for relevant keywords. In the case of browser information collection, the stealer searches exclusively for keywords related to banking (e.g., “revolut”, “stripe”, “bank”) and cryptocurrencies (e.g., “binance”, “metamask”, “wallet”). In addition to this, the stealer is capable of extracting extension data from a hardcoded list of extensions associated with cryptocurrencies.
Part of the extension list which the stealer utilizes to extract data from
Telegram data collection begins with terminating the Telegram.exe process using the taskkill command. Subsequently, if the telegram_optimized feature is set to False, the malware zips the entire tdata directory (typically located at %APPDATA%\Roaming\Telegram Desktop\tdata) and transmits it to the attacker. Otherwise, it selectively copies and zips only the subdirectories containing valuable info, such as message log. The generated archive is sent to the endpoint /delivery with the filename tdata_session.zip.
The stealer includes two features connected with Discord: credentials stealing and self-spreading. The first one can be utilized to acquire credentials both from the standard client and custom clients. If the client is Chromium-based, the stealer employs the same data exfiltration mechanism as during browser credentials stealing.
The self-spreading feature is configurable (meaning it can be disabled in the config). The stealer acquires the list of user’s friends and channels via the Discord API and sends a message provided by the attacker. This stealer does not support attaching files to such messages.
The VPN collector is searching for a set of known VPN software to extract account credentials from the credentials file with a known path that gets parsed with a regular expression. The extraction occurs from the following set of applications:
File retrieval is performed regardless of the configuration. The script relies on a predefined set of paths associated with the current user (such as Desktop, Download, etc.) and file extensions mainly connected with documents and media. The script also has a predefined list of filenames to exfiltrate. The extracted files are packed into a ZIP archive which is later sent to the C2 asynchronously. An interesting aspect is that the filename list includes several French words, such as “motdepasse” (French for “password”), “banque” (French for “bank”), “secret” (French for “secret”), and “compte” (French for “account”).
We were able to identify additional modules that are downloaded from the C2 rather than embedded into the stealer script; however, we weren’t able to obtain them. These modules can be described by the following table, with the “Details” column referring to the information that could be extracted from the main stealer code.
| Module name | Endpoint to download | Details |
| Chrome grabber | /api/chrome-grabber-template/{payload_id} | – |
| Wallet patcher | /api/wallet-patcher/{payload_id} | Checks whether “Exodus” and “Atomic” cryptocurrency wallets are installed |
| Extra collector | /api/extra-collector/{payload_id} | Uses a set of options from the config, such as collect_filezilla, collect_vpn_data, collect_steam, and collect_screenshots |
| HVNC | /hvnc | Is saved to the Startup directory (%APPDATA%\Microsoft\Windows\Start Menu\Programs\Startup\hvnc.py) to execute upon system boot |
The Wallet patcher and Extra collector scripts are received in an encrypted form from the C2 server. To decrypt them, the attackers utilize the AES-GCM algorithm in conjunction with PBKDF2 (HMAC and SHA256). After decryption, the additional payload has its template placeholders replaced and is stored under a partially randomized name within a temporary folder.
Decryption routine and template substitution
Once all operations are completed, the stealer removes itself from the drive, along with the artifacts folder (Arkanix_lol in this case).
| MD5 | a3fc46332dcd0a95e336f6927bae8bb7 |
| File name | ArkanixStealer.exe |
During our analysis, we were able to obtain both the release and debug versions of the native implementation, as both were uploaded to publicly available resources. The following are the key differences between the two:
arkanix[.]pw.Notably, the native implementation explicitly references the name of the stealer in the VersionInfo resources. This naming convention is consistent across both the debug version and certain samples containing the release version of the implant.
Version info
After launching, the stealer implements a series of analysis countermeasures to verify that the application is not being executed within a sandboxed environment or run under a debugger. Following these checks, the sample patches AmsiScanBuffer and EtwEventWrite to prevent the triggering of any unwanted events by the system.
Once the preliminary checks are completed, the sample proceeds to gather information about the system. The list of capabilities is hardcoded and cannot be modified from the server side, in contrast to the Python version. What is more, the feature list is quite similar to the Python version except a few ones.
The stealer is capable of collecting information about known RDP connections that the compromised user has. To achieve this, it searches for .rdp files in %USERPROFILE%\Documents and extracts the full server address, password, username and server port.
The stealer also targets gamers and is capable to steal credentials from the popular gaming platform clients, including:
The native version, unlike its Python counterpart, is capable of capturing screenshots for each monitor via capCreateCaptureWindowA WinAPI.
In conclusion, this sample communicates with the C2 server through the same endpoints as the Python version. However, in this instance, all data is encrypted using the same AES-GCM + PBKDF2 (HMAC and SHA256) scheme as partially employed in the Python variant. In some observed samples, the key used was arkanix_secret_key_v20_2024. Alongside that, the C++ sample explicitly sets the User-Agent to ArkanixStealer/1.0.
| MD5 | 3283f8c54a3ddf0bc0d4111cc1f950c0 |
| File name | – |
This is an implant embedded within the resources of the C++ implementation. The author incorporated it into the resource section without applying any obfuscation or encryption. Subsequently, the stealer extracts the payload to a temporary folder with a randomly generated name composed of hexadecimal digits (0-9 and A-F) and executes it using the CreateProcess WinAPI. The payload itself is the unaltered publicly available project known as “ChromElevator”. To summarize, this tool consists of two components: an injector and the main payload. The injector initializes a direct syscall engine, spawns a suspended target browser process, and injects the decrypted code into it via Nt syscalls. The injected payload then decrypts the browser master key and exfiltrates data such as cookies, login information, web data, and so on.
During the Arkanix campaign, two domains used in the attacks were identified. Although these domains were routed through Cloudflare, a real IP address was successfully discovered for one of them, namely, arkanix[.]pw. For the second one we only obtained a Cloudflare IP address.
| Domain | IP | First seen | ASN |
| arkanix[.]pw | 195.246.231[.]60 | Oct 09, 2025 | – |
| arkanix[.]ru | 172.67.186[.]193 | Oct 19, 2025 | – |
Both servers were also utilized to host the stealer panel, which allows attackers to monitor their victims. The contents of the panel are secured behind a sign-in page. Closer to the end of our research, the panel was seemingly taken down with no message or notice.
Stealer panel sign-in page
During the research of this campaign, we noticed that the forum posts advertising the stealer contained a link leading to a Discord server dubbed “Arkanix” by the authors. The server posed as a forum where authors posted various content and clients could ask various questions regarding this malicious software. While users mainly thank and ask about when the feature promised by the authors will be released and added into the stealer, the content made by the authors is broader. The adversary builds up the communication with potential buyers using the same marketing and communication methods real companies employ. To begin with, they warm up the audience by posting surveys about whether they should implement specific features, such as Discord injection and binding with a legitimate application (sic!).
Feature votes
Additionally, the author promised to release a crypter as a side project in four to six weeks, at the end of October. As of now, the stealer seems to have been taken down without any notice while the crypter was never released.
Arkanix Crypter
Furthermore, the Arkanix Stealer authors decided to implement a referral program to attract new customers. Referrers were promised an additional free hour to their premium license, while invited customers received seven days of free “premium” trial use. As stated in forum posts, the premium plan included the following features:
Referral program ad and corresponding panel interface
Speaking of technical details, based on the screenshot of the Visual Studio stealer project that was sent to the Discord server, we can conclude that the author is German-speaking.
This same screenshot also serves as a probable indicator of AI-assisted development as it shares the common patterns of such assistants, e.g. the presence of the utils.cpp file. What provides even more confidence is the overall code structure, the presence of comments and extensive debugging log output.
Example of LLM-specific patterns
Information stealers have always posed as a serious threat to users’ data. Arkanix is no exception as it targets a wide range of users, from those interested in cryptocurrencies and gaming to those using online banking. It collects a vast amount of information including highly sensitive personal data. While being quite functional, it contains probable traces of LLM-assisted development which suggests that such assistance might have drastically reduced development time and costs. Hence it follows that this campaign tends to be more of a one-shot campaign for quick financial gains rather than a long-running infection. The panel and the Discord chat were taken down around December 2025, leaving no message or traces of further development or a resurgence.
In addition, the developers behind the Arkanix Stealer decided to address the public, implementing a forum where they posted development insights, conducted surveys and even ran a referral program where you could get bonuses for “bringing a friend”. This behavior makes Arkanix more of a public software product than a shady stealer.
Additional IoCs are available to customers of our Threat Intelligence Reporting service. For more details, contact us at crimewareintel@kaspersky.com.
File hashes
752e3eb5a9c295ee285205fb39b67fc4
c1e4be64f80bc019651f84ef852dfa6c
a8eeda4ae7db3357ed2ee0d94b963eff
c0c04df98b7d1ca9e8c08dd1ffbdd16b
88487ab7a666081721e1dd1999fb9fb2
d42ba771541893eb047a0e835bd4f84e
5f71b83ca752cb128b67dbb1832205a4
208fa7e01f72a50334f3d7607f6b82bf
e27edcdeb44522a9036f5e4cd23f1f0c
ea50282fa1269836a7e87eddb10f95f7
643696a052ea1963e24cfb0531169477
f5765930205719c2ac9d2e26c3b03d8d
576de7a075637122f47d02d4288e3dd6
7888eb4f51413d9382e2b992b667d9f5
3283f8c54a3ddf0bc0d4111cc1f950c0
Domains and IPs
arkanix[.]pw
arkanix[.]ru
