For the latest discoveries in cyber research for the week of 5th January, please download our Threat Intelligence Bulletin.
TOP ATTACKS AND BREACHES
Two US banks, Artisans’ Bank and VeraBank, disclosed that customer data was exposed in an August ransomware attack on their vendor, Marquis Software. The vendor was breached via SonicWall vulnerability, and while the banks’ own systems were not compromised, researchers estimate the incident may have affected in total up to 1.35 million people across dozens of financial institutions.
Romania’s largest coal-based power producer, Oltenia Energy Complex, has faced a ransomware attack attributed to the Gentlemen group. The company said files were encrypted and Enterprise Resource Planning systems, email, and the website were disrupted, partially affecting operations, while power supply remained stable and recovery continues.
Emurasoft, maker of EmEditor software, reported a website compromise that redirected the homepage download button to a fake installer for 4 days. The installer deployed infostealer malware that harvested credentials and added a rogue extension enabling remote control and cryptocurrency swapping.
US-based Sedgwick Government Solutions, which manages claims, workforce health, risk, and productivity for government agencies and federal employees, has experienced a cybersecurity incident. The incident was limited to an isolated file transfer system, with no evidence of access to claims servers. The company notified law enforcement and clients after the TridentLocker ransomware group claimed an attack on December 31.
Korean Air, South Korean airline, has suffered a data breach via KC&D Service, a vendor managing inflight catering and duty free. The incident exposed personal data of roughly 30,000 employees, including names and bank account numbers, while customer information was not affected. Cl0p claimed responsibility and reportedly exploited an Oracle E-Business Suite flaw.
Check Point IPS, Threat Emulation and Harmony Endpoint provide protection against this threat(Oracle Multiple Products Remote Code Execution; Ransomware.Win.Clop; Ransomware.Wins.Clop; Ransomware.Wins.Clop.ta.*)
Trust Wallet, a cryptocurrency wallet provider, has disclosed a second Shai-Hulud supply-chain compromise of its Chrome extension, resulting in approximately $8.5 million in losses. Using a leaked Chrome store key, attackers published tampered v2.68 which exfiltrated wallet recovery phrases upon unlock.
European Space Agency (ESA), has confirmed a cybersecurity incident affecting a very small number of external servers outside its corporate network. ESA began forensic analysis and secured potentially affected devices after a threat actor claimed to have stolen 200GB of source code and access credentials in mid-December.
VULNERABILITIES AND PATCHES
Researchers highlighted CVE-2025-14346, a critical missing-authentication flaw in WHILL Model C2 and Model F power wheelchairs that enables attackers within Bluetooth range to take control. CISA urged immediate mitigations, warning that compromise could manipulate wheelchair movements and cause physical harm in healthcare and public settings. No public exploitation has been reported yet.
Security researchers disclosed CVE-2025-20700, CVE-2025-20701 (CVSS 8.8) and CVE-2025-20702 (CVSS 9.6) affecting Airoha Bluetooth SoCs. The flaws enabling unauthenticated access to the RACE protocol, arbitrary memory operations, and nearby takeover of headphones to extract link keys and impersonate devices to access paired smartphones.
A patch has been released for CVE-2025-47411, an important privilege escalation in Apache StreamPipes 0.69.0 to 0.97.0 caused by flawed user ID creation enabling JWT token manipulation. Attackers can impersonate existing administrators to gain full control.
IBM API Connect, an enterprise API management platform, is affected by a critical authentication bypass vulnerability (CVE-2025-13915, CVSS 9.8) enabling remote unauthorized access without credentials. The flaw impacts versions 10.0.8.0 through 10.0.8.5 and 10.0.11.0, with patches and iFixes available; no exploitation has been reported.
THREAT INTELLIGENCE REPORTS
Researchers exposed a new APT36 cyber espionage campaign targeting Indian government, academic, and strategic institutions. The Pakistan affiliated group delivers ZIP attachments disguised as PDFs that install ReadOnly and WriteOnly malware, which enables remote control, steals data, monitors clipboards, captures screenshots, and maintains access.
DarkSpectre, a Chinese affiliated threat actor, has compromised 8.8 million Chrome, Edge, and Firefox users globally via campaigns including ShadyPanda, Zoom Stealer, and GhostPoster. The group employs malicious browser extensions with tactics such as time-bomb activation, dormant sleepers, PNG steganography, and heavy JavaScript obfuscation, exfiltrating corporate meeting data while impersonating videoconferencing tools and abusing browser platform permissions.
Security researchers discovered two Chrome Web Store extensions, Chat GPT for Chrome with GPT-5 and AI Sidebar, that exfiltrate ChatGPT and DeepSeek chat histories, along with users’ browsing activity, every 30 minutes. The extensions collectively have over 900,000 installations, and one holds a Google Featured badge.
Researchers identified the rapid expansion of the Kimwolf botnet, which has infected more than 2 million devices globally by abusing residential proxy networks to reach local devices behind home routers. The campaign leverages insecure Android TV boxes and digital photo frames to enable DDoS, ad fraud, account takeover, and mass scraping.
For the latest discoveries in cyber research for the week of 29th December, please download our Threat Intelligence Bulletin.
TOP ATTACKS AND BREACHES
Romanian Waters, the country’s national water management authority, was hit by a ransomware attack that resulted in nearly 1,000 computer systems across national and regional offices being encrypted. The attack affected geographic information systems, databases, email, web servers, and Windows workstations. Operational technology controlling water infrastructure was not impacted, and no data leakage has been reported, but key IT services were disrupted across the organization.
France’s postal service La Poste has suffered a cyber-attack that disrupted key digital systems, impacting online parcel tracking, mail distribution, and banking services for customers of both the postal service and La Banque Postale. Some services were temporarily unavailable, with no evidence of data compromise. The attack was claimed by the pro-Russian hacktivist group NoName057(16).
Insurance giant Aflac has confirmed a data breach they experienced in June that resulted in the theft of sensitive files containing insurance claims, health data and Social Security numbers. The breach affected personal details of approximately 22.7 million individuals in its US business. The attack has been attributed to Scattered Spider threat group.
Check Point Harmony Endpoint provides protection against this threat.
Japan’s leading carmaker Nissan Motor Corporation has acknowledged a data breach that resulted in the exposure of personal information for approximately 21,000 customers from Nissan Fukuoka Sales Corporation including names, addresses, phone numbers, email addresses, and sales operation data. The incident occurred after unauthorized access to Red Hat data servers led to the leak, but financial data was not affected. The Crimson Collective threat actor claimed responsibility for the initial breach, with ShinyHunters later hosting samples of the stolen data.
Trust Wallet, a popular non-custodial cryptocurrency wallet, has disclosed a cyber-attack involving a compromised Chrome extension update. The attack exfiltrated sensitive wallet data, including seed phrases, to a malicious domain, resulting in at least $7 million in losses. The incident primarily affected users of Chrome extension version 2.68.0, allowing attackers to drain wallets.
Ubisoft’s live service game Rainbow Six Siege (R6) has confirmed a cyber-attack in which threat actors abused internal systems to manipulate bans, unlock all cosmetics and developer-only skins, and distribute around $13.33 million worth of in-game currency worldwide.
Baker University has encountered a data breach that resulted in attackers accessing its network and stealing sensitive information belongs to 53,624 students, alumni, staff, and affiliates of the university, such as names, Social Security numbers, financial account details, and medical records.
VULNERABILITIES AND PATCHES
A high-severity memory-read vulnerability, CVE-2025-14847, dubbed “MongoBleed” has been identified in multiple MongoDB Server versions, allowing unauthenticated remote attackers to exploit a zlib implementation flaw and potentially access uninitialized heap memory. The issue, caused by improper handling of length parameter inconsistency (CWE-130), may permit arbitrary code execution and system compromise. Affected versions include MongoDB 4.0 through 8.2.3.
Details on a critical serialization injection vulnerability in LangChain Core were disclosed. CVE-2025-68664 (CVSS 9.3) affects langchain-core, where unescaped user-controlled dictionaries with lc keys are treated as trusted objects during deserialization, enabling secret extraction, prompt injection, and potentially arbitrary code execution.
A critical buffer overflow vulnerability, CVE-2025-68615, in Net-SNMP’s snmptrapd daemon can be triggered remotely via a specially crafted packet. The issue has a CVSS score of 9.8 and may allow unauthenticated attackers to achieve remote code execution or cause service crashes. Patches are available, and the vulnerability is addressed in Net-SNMP versions 5.9.5 and 5.10.pre2.
THREAT INTELLIGENCE REPORTS
Check Point researchers describe a phishing campaign in which attackers abused Google Cloud Application Integration’s “Send Email” workflow to send over 9,000 spoofed Google notification emails from a Google address. The messages targeted manufacturing, technology, and finance sectors and used multi-step redirection through Google domains to lead victims to a Microsoft-themed credential harvesting site. Most victims located in the US, Asia-Pacific, and Europe.
Researchers uncovered a two-year Evasive Panda campaign using adversary-in-the-middle DNS poisoning to deliver MgBot via fake updaters and stealthy loaders. The chain used multi-stage shellcode, hybrid encryption, and DLL sideloading to run MgBot in memory, with victim-specific payloads tied to machines via DPAPI and RC5. Attackers poisoned legitimate domains, injected into signed system processes for persistence, and updated configs with hardcoded C2s.
Check Point Harmony Endpoint provides protection against this threat (Infostealer.Win.MgBot)
A Webrat campaign leveraged fake GitHub repositories masquerading as exploit and proof-of-concept code for high-severity CVEs, targeting gamers, students, and inexperienced security researchers. The attack uses droppers to elevate privileges, disable Windows Defender, and deploy the Webrat backdoor, enabling remote control, credential theft, keylogging, and device surveillance.
Researchers found lotusbail, a malicious npm package masquerading as a WhatsApp Web API library that intercepts messages and steals session/auth data, contacts, and media via WebSocket tampering and device-pairing hijack. Separately, 14 malicious NuGet packages were found redirecting crypto funds and stealing Google Ads OAuth tokens.
In mid-2025, we identified a malicious driver file on computer systems in Asia. The driver file is signed with an old, stolen, or leaked digital certificate and registers as a mini-filter driver on infected machines. Its end-goal is to inject a backdoor Trojan into the system processes and provide protection for malicious files, user-mode processes, and registry keys.
Our analysis indicates that the final payload injected by the driver is a new sample of the ToneShell backdoor, which connects to the attacker’s servers and provides a reverse shell, along with other capabilities. The ToneShell backdoor is a tool known to be used exclusively by the HoneyMyte (aka Mustang Panda or Bronze President) APT actor and is often used in cyberespionage campaigns targeting government organizations, particularly in Southeast and East Asia.
The command-and-control servers for the ToneShell backdoor used in this campaign were registered in September 2024 via NameCheap services, and we suspect the attacks themselves to have begun in February 2025. We’ve observed through our telemetry that the new ToneShell backdoor is frequently employed in cyberespionage campaigns against government organizations in Southeast and East Asia, with Myanmar and Thailand being the most heavily targeted.
Notably, nearly all affected victims had previously been infected with other HoneyMyte tools, including the ToneDisk USB worm, PlugX, and older variants of ToneShell. Although the initial access vector remains unclear, it’s suspected that the threat actor leveraged previously compromised machines to deploy the malicious driver.
Compromised digital certificate
The driver file is signed with a digital certificate from Guangzhou Kingteller Technology Co., Ltd., with a serial number of 08 01 CC 11 EB 4D 1D 33 1E 3D 54 0C 55 A4 9F 7F. The certificate was valid from August 2012 until 2015.
We found multiple other malicious files signed with the same certificate which didn’t show any connections to the attacks described in this article. Therefore, we believe that other threat actors have been using it to sign their malicious tools as well. The following image shows the details of the certificate.
Technical details of the malicious driver
The filename used for the driver on the victim’s machine is ProjectConfiguration.sys. The registry key created for the driver’s service uses the same name, ProjectConfiguration.
The malicious driver contains two user-mode shellcodes, which are embedded into the .data section of the driver’s binary file. The shellcodes are executed as separate user-mode threads. The rootkit functionality protects both the driver’s own module and the user-mode processes into which the backdoor code is injected, preventing access by any process on the system.
API resolution
To obfuscate the actual behavior of the driver module, the attackers used dynamic resolution of the required API addresses from hash values.
The malicious driver first retrieves the base address of the ntoskrnl.exe and fltmgr.sys by calling ZwQuerySystemInformation with the SystemInformationClass set to SYSTEM_MODULE_INFORMATION. It then iterates through this system information and searches for the desired DLLs by name, noting the ImageBaseAddress of each.
Once the base addresses of the libraries are obtained, the driver uses a simple hashing algorithm to dynamically resolve the required API addresses from ntoskrnl.exe and fltmgr.sys.
The hashing algorithm is shown below. The two variants of the seed value provided in the comment are used in the shellcodes and the final payload of the attack.
Protection of the driver file
The malicious driver registers itself with the Filter Manager using FltRegisterFilter and sets up a pre-operation callback. This callback inspects I/O requests for IRP_MJ_SET_INFORMATION and triggers a malicious handler when certain FileInformationClass values are detected. The handler then checks whether the targeted file object is associated with the driver; if it is, it forces the operation to fail by setting IOStatus to STATUS_ACCESS_DENIED. The relevant FileInformationClass values include:
FileRenameInformation
FileDispositionInformation
FileRenameInformationBypassAccessCheck
FileDispositionInformationEx
FileRenameInformationEx
FileRenameInformationExBypassAccessCheck
These classes correspond to file-delete and file-rename operations. By monitoring them, the driver prevents itself from being removed or renamed – actions that security tools might attempt when trying to quarantine it.
Protection of registry keys
The driver also builds a global list of registry paths and parameter names that it intends to protect. This list contains the following entries:
ProjectConfiguration
ProjectConfiguration\Instances
ProjectConfiguration Instance
To guard these keys, the malware sets up a RegistryCallback routine, registering it through CmRegisterCallbackEx. To do so, it must assign itself an altitude value. Microsoft governs altitude assignments for mini-filters, grouping them into Load Order categories with predefined altitude ranges. A filter driver with a low numerical altitude is loaded into the I/O stack below filters with higher altitudes. The malware uses a hardcoded starting point of 330024 and creates altitude strings in the format 330024.%l, where %l ranges from 0 to 10,000.
The malware then begins attempting to register the callback using the first generated altitude. If the registration fails with STATUS_FLT_INSTANCE_ALTITUDE_COLLISION, meaning the altitude is already taken, it increments the value and retries. It repeats this process until it successfully finds an unused altitude.
The callback monitors four specific registry operations. Whenever one of these operations targets a key from its protected list, it responds with 0xC0000022 (STATUS_ACCESS_DENIED), blocking the action. The monitored operations are:
RegNtPreCreateKey
RegNtPreOpenKey
RegNtPreCreateKeyEx
RegNtPreOpenKeyEx
Microsoft designates the 320000–329999 altitude range for the FSFilter Anti-Virus Load Order Group. The malware’s chosen altitude exceeds this range. Since filters with lower altitudes sit deeper in the I/O stack, the malicious driver intercepts file operations before legitimate low-altitude filters like antivirus components, allowing it to circumvent security checks.
Finally, the malware tampers with the altitude assigned to WdFilter, a key Microsoft Defender driver. It locates the registry entry containing the driver’s altitude and changes it to 0, effectively preventing WdFilter from being loaded into the I/O stack.
Protection of user-mode processes
The malware sets up a list intended to hold protected process IDs (PIDs). It begins with 32 empty slots, which are filled as needed during execution. A status flag is also initialized and set to 1 to indicate that the list starts out empty.
Next, the malware uses ObRegisterCallbacks to register two callbacks that intercept process-related operations. These callbacks apply to both OB_OPERATION_HANDLE_CREATE and OB_OPERATION_HANDLE_DUPLICATE, and both use a malicious pre-operation routine.
This routine checks whether the process involved in the operation has a PID that appears in the protected list. If so, it sets the DesiredAccess field in the OperationInformation structure to 0, effectively denying any access to the process.
The malware also registers a callback routine by calling PsSetCreateProcessNotifyRoutine. These callbacks are triggered during every process creation and deletion on the system. This malware’s callback routine checks whether the parent process ID (PPID) of a process being deleted exists in the protected list; if it does, the malware removes that PPID from the list. This eventually removes the rootkit protection from a process with an injected backdoor, once the backdoor has fulfilled its responsibilities.
Payload injection
The driver delivers two user-mode payloads.
The first payload spawns an svchost process and injects a small delay-inducing shellcode. The PID of this new svchost instance is written to a file for later use.
The second payload is the final component – the ToneShell backdoor – and is later injected into that same svchost process.
Injection workflow:
The malicious driver searches for a high-privilege target process by iterating through PIDs and checking whether each process exists and runs under SeLocalSystemSid. Once it finds one, it customizes the first payload using random event names, file names, and padding bytes, then creates a named event and injects the payload by attaching its current thread to the process, allocating memory, and launching a new thread.
After injection, it waits for the payload to signal the event, reads the PID of the newly created svchost process from the generated file, and adds it to its protected process list. It then similarly customizes the second payload (ToneShell) using random event name and random padding bytes, then creates a named event and injects the payload by attaching to the process, allocating memory, and launching a new thread.
Once the ToneShell backdoor finishes execution, it signals the event. The malware then removes the svchost PID from the protected list, waits 10 seconds, and attempts to terminate the process.
ToneShell backdoor
The final stage of the attack deploys ToneShell, a backdoor previously linked to operations by the HoneyMyte APT group and discussed in earlier reporting (see Malpedia and MITRE). Notably, this is the first time we’ve seen ToneShell delivered through a kernel-mode loader, giving it protection from user-mode monitoring and benefiting from the rootkit capabilities of the driver that hides its activity from security tools.
Earlier ToneShell variants generated a 16-byte GUID using CoCreateGuid and stored it as a host identifier. In contrast, this version checks for a file named C:\ProgramData\MicrosoftOneDrive.tlb, validating a 4-byte marker inside it. If the file is absent or the marker is invalid, the backdoor derives a new pseudo-random 4-byte identifier using system-specific values (computer name, tick count, and PRNG), then creates the file and writes the marker. This becomes the unique ID for the infected host.
The samples we have analyzed contact two command-and-control servers:
avocadomechanism[.]com
potherbreference[.]com
ToneShell communicates with its C2 over raw TCP on port 443 while disguising traffic using fake TLS headers. This version imitates the first bytes of a TLS 1.3 record (0x17 0x03 0x04) instead of the TLS 1.2 pattern used previously. After this three-byte marker, each packet contains a size field and an encrypted payload.
Packet layout:
Header (3 bytes): Fake TLS marker
Size (2 bytes): Payload length
Payload: Encrypted with a rolling XOR key
The backdoor supports a set of remote operations, including file upload/download, remote shell functionality, and session control. The command set includes:
Command ID
Description
0x1
Create temporary file for incoming data
0x2 / 0x3
Download file
0x4
Cancel download
0x7
Establish remote shell via pipe
0x8
Receive operator command
0x9
Terminate shell
0xA / 0xB
Upload file
0xC
Cancel upload
0xD
Close connection
Conclusion
We assess with high confidence that the activity described in this report is linked to the HoneyMyte threat actor. This conclusion is supported by the use of the ToneShell backdoor as the final-stage payload, as well as the presence of additional tools long associated with HoneyMyte – such as PlugX, and the ToneDisk USB worm – on the impacted systems.
HoneyMyte’s 2025 operations show a noticeable evolution toward using kernel-mode injectors to deploy ToneShell, improving both stealth and resilience. In this campaign, we observed a new ToneShell variant delivered through a kernel-mode driver that carries and injects the backdoor directly from its embedded payload. To further conceal its activity, the driver first deploys a small user-mode component that handles the final injection step. It also uses multiple obfuscation techniques, callback routines, and notification mechanisms to hide its API usage and track process and registry activity, ultimately strengthening the backdoor’s defenses.
Because the shellcode executes entirely in memory, memory forensics becomes essential for uncovering and analyzing this intrusion. Detecting the injected shellcode is a key indicator of ToneShell’s presence on compromised hosts.
Recommendations
To protect themselves against this threat, organizations should:
Implement robust network security measures, such as firewalls and intrusion detection systems.
In mid-2025, we identified a malicious driver file on computer systems in Asia. The driver file is signed with an old, stolen, or leaked digital certificate and registers as a mini-filter driver on infected machines. Its end-goal is to inject a backdoor Trojan into the system processes and provide protection for malicious files, user-mode processes, and registry keys.
Our analysis indicates that the final payload injected by the driver is a new sample of the ToneShell backdoor, which connects to the attacker’s servers and provides a reverse shell, along with other capabilities. The ToneShell backdoor is a tool known to be used exclusively by the HoneyMyte (aka Mustang Panda or Bronze President) APT actor and is often used in cyberespionage campaigns targeting government organizations, particularly in Southeast and East Asia.
The command-and-control servers for the ToneShell backdoor used in this campaign were registered in September 2024 via NameCheap services, and we suspect the attacks themselves to have begun in February 2025. We’ve observed through our telemetry that the new ToneShell backdoor is frequently employed in cyberespionage campaigns against government organizations in Southeast and East Asia, with Myanmar and Thailand being the most heavily targeted.
Notably, nearly all affected victims had previously been infected with other HoneyMyte tools, including the ToneDisk USB worm, PlugX, and older variants of ToneShell. Although the initial access vector remains unclear, it’s suspected that the threat actor leveraged previously compromised machines to deploy the malicious driver.
Compromised digital certificate
The driver file is signed with a digital certificate from Guangzhou Kingteller Technology Co., Ltd., with a serial number of 08 01 CC 11 EB 4D 1D 33 1E 3D 54 0C 55 A4 9F 7F. The certificate was valid from August 2012 until 2015.
We found multiple other malicious files signed with the same certificate which didn’t show any connections to the attacks described in this article. Therefore, we believe that other threat actors have been using it to sign their malicious tools as well. The following image shows the details of the certificate.
Technical details of the malicious driver
The filename used for the driver on the victim’s machine is ProjectConfiguration.sys. The registry key created for the driver’s service uses the same name, ProjectConfiguration.
The malicious driver contains two user-mode shellcodes, which are embedded into the .data section of the driver’s binary file. The shellcodes are executed as separate user-mode threads. The rootkit functionality protects both the driver’s own module and the user-mode processes into which the backdoor code is injected, preventing access by any process on the system.
API resolution
To obfuscate the actual behavior of the driver module, the attackers used dynamic resolution of the required API addresses from hash values.
The malicious driver first retrieves the base address of the ntoskrnl.exe and fltmgr.sys by calling ZwQuerySystemInformation with the SystemInformationClass set to SYSTEM_MODULE_INFORMATION. It then iterates through this system information and searches for the desired DLLs by name, noting the ImageBaseAddress of each.
Once the base addresses of the libraries are obtained, the driver uses a simple hashing algorithm to dynamically resolve the required API addresses from ntoskrnl.exe and fltmgr.sys.
The hashing algorithm is shown below. The two variants of the seed value provided in the comment are used in the shellcodes and the final payload of the attack.
Protection of the driver file
The malicious driver registers itself with the Filter Manager using FltRegisterFilter and sets up a pre-operation callback. This callback inspects I/O requests for IRP_MJ_SET_INFORMATION and triggers a malicious handler when certain FileInformationClass values are detected. The handler then checks whether the targeted file object is associated with the driver; if it is, it forces the operation to fail by setting IOStatus to STATUS_ACCESS_DENIED. The relevant FileInformationClass values include:
FileRenameInformation
FileDispositionInformation
FileRenameInformationBypassAccessCheck
FileDispositionInformationEx
FileRenameInformationEx
FileRenameInformationExBypassAccessCheck
These classes correspond to file-delete and file-rename operations. By monitoring them, the driver prevents itself from being removed or renamed – actions that security tools might attempt when trying to quarantine it.
Protection of registry keys
The driver also builds a global list of registry paths and parameter names that it intends to protect. This list contains the following entries:
ProjectConfiguration
ProjectConfiguration\Instances
ProjectConfiguration Instance
To guard these keys, the malware sets up a RegistryCallback routine, registering it through CmRegisterCallbackEx. To do so, it must assign itself an altitude value. Microsoft governs altitude assignments for mini-filters, grouping them into Load Order categories with predefined altitude ranges. A filter driver with a low numerical altitude is loaded into the I/O stack below filters with higher altitudes. The malware uses a hardcoded starting point of 330024 and creates altitude strings in the format 330024.%l, where %l ranges from 0 to 10,000.
The malware then begins attempting to register the callback using the first generated altitude. If the registration fails with STATUS_FLT_INSTANCE_ALTITUDE_COLLISION, meaning the altitude is already taken, it increments the value and retries. It repeats this process until it successfully finds an unused altitude.
The callback monitors four specific registry operations. Whenever one of these operations targets a key from its protected list, it responds with 0xC0000022 (STATUS_ACCESS_DENIED), blocking the action. The monitored operations are:
RegNtPreCreateKey
RegNtPreOpenKey
RegNtPreCreateKeyEx
RegNtPreOpenKeyEx
Microsoft designates the 320000–329999 altitude range for the FSFilter Anti-Virus Load Order Group. The malware’s chosen altitude exceeds this range. Since filters with lower altitudes sit deeper in the I/O stack, the malicious driver intercepts file operations before legitimate low-altitude filters like antivirus components, allowing it to circumvent security checks.
Finally, the malware tampers with the altitude assigned to WdFilter, a key Microsoft Defender driver. It locates the registry entry containing the driver’s altitude and changes it to 0, effectively preventing WdFilter from being loaded into the I/O stack.
Protection of user-mode processes
The malware sets up a list intended to hold protected process IDs (PIDs). It begins with 32 empty slots, which are filled as needed during execution. A status flag is also initialized and set to 1 to indicate that the list starts out empty.
Next, the malware uses ObRegisterCallbacks to register two callbacks that intercept process-related operations. These callbacks apply to both OB_OPERATION_HANDLE_CREATE and OB_OPERATION_HANDLE_DUPLICATE, and both use a malicious pre-operation routine.
This routine checks whether the process involved in the operation has a PID that appears in the protected list. If so, it sets the DesiredAccess field in the OperationInformation structure to 0, effectively denying any access to the process.
The malware also registers a callback routine by calling PsSetCreateProcessNotifyRoutine. These callbacks are triggered during every process creation and deletion on the system. This malware’s callback routine checks whether the parent process ID (PPID) of a process being deleted exists in the protected list; if it does, the malware removes that PPID from the list. This eventually removes the rootkit protection from a process with an injected backdoor, once the backdoor has fulfilled its responsibilities.
Payload injection
The driver delivers two user-mode payloads.
The first payload spawns an svchost process and injects a small delay-inducing shellcode. The PID of this new svchost instance is written to a file for later use.
The second payload is the final component – the ToneShell backdoor – and is later injected into that same svchost process.
Injection workflow:
The malicious driver searches for a high-privilege target process by iterating through PIDs and checking whether each process exists and runs under SeLocalSystemSid. Once it finds one, it customizes the first payload using random event names, file names, and padding bytes, then creates a named event and injects the payload by attaching its current thread to the process, allocating memory, and launching a new thread.
After injection, it waits for the payload to signal the event, reads the PID of the newly created svchost process from the generated file, and adds it to its protected process list. It then similarly customizes the second payload (ToneShell) using random event name and random padding bytes, then creates a named event and injects the payload by attaching to the process, allocating memory, and launching a new thread.
Once the ToneShell backdoor finishes execution, it signals the event. The malware then removes the svchost PID from the protected list, waits 10 seconds, and attempts to terminate the process.
ToneShell backdoor
The final stage of the attack deploys ToneShell, a backdoor previously linked to operations by the HoneyMyte APT group and discussed in earlier reporting (see Malpedia and MITRE). Notably, this is the first time we’ve seen ToneShell delivered through a kernel-mode loader, giving it protection from user-mode monitoring and benefiting from the rootkit capabilities of the driver that hides its activity from security tools.
Earlier ToneShell variants generated a 16-byte GUID using CoCreateGuid and stored it as a host identifier. In contrast, this version checks for a file named C:\ProgramData\MicrosoftOneDrive.tlb, validating a 4-byte marker inside it. If the file is absent or the marker is invalid, the backdoor derives a new pseudo-random 4-byte identifier using system-specific values (computer name, tick count, and PRNG), then creates the file and writes the marker. This becomes the unique ID for the infected host.
The samples we have analyzed contact two command-and-control servers:
avocadomechanism[.]com
potherbreference[.]com
ToneShell communicates with its C2 over raw TCP on port 443 while disguising traffic using fake TLS headers. This version imitates the first bytes of a TLS 1.3 record (0x17 0x03 0x04) instead of the TLS 1.2 pattern used previously. After this three-byte marker, each packet contains a size field and an encrypted payload.
Packet layout:
Header (3 bytes): Fake TLS marker
Size (2 bytes): Payload length
Payload: Encrypted with a rolling XOR key
The backdoor supports a set of remote operations, including file upload/download, remote shell functionality, and session control. The command set includes:
Command ID
Description
0x1
Create temporary file for incoming data
0x2 / 0x3
Download file
0x4
Cancel download
0x7
Establish remote shell via pipe
0x8
Receive operator command
0x9
Terminate shell
0xA / 0xB
Upload file
0xC
Cancel upload
0xD
Close connection
Conclusion
We assess with high confidence that the activity described in this report is linked to the HoneyMyte threat actor. This conclusion is supported by the use of the ToneShell backdoor as the final-stage payload, as well as the presence of additional tools long associated with HoneyMyte – such as PlugX, and the ToneDisk USB worm – on the impacted systems.
HoneyMyte’s 2025 operations show a noticeable evolution toward using kernel-mode injectors to deploy ToneShell, improving both stealth and resilience. In this campaign, we observed a new ToneShell variant delivered through a kernel-mode driver that carries and injects the backdoor directly from its embedded payload. To further conceal its activity, the driver first deploys a small user-mode component that handles the final injection step. It also uses multiple obfuscation techniques, callback routines, and notification mechanisms to hide its API usage and track process and registry activity, ultimately strengthening the backdoor’s defenses.
Because the shellcode executes entirely in memory, memory forensics becomes essential for uncovering and analyzing this intrusion. Detecting the injected shellcode is a key indicator of ToneShell’s presence on compromised hosts.
Recommendations
To protect themselves against this threat, organizations should:
Implement robust network security measures, such as firewalls and intrusion detection systems.
The Evasive Panda APT group (also known as Bronze Highland, Daggerfly, and StormBamboo) has been active since 2012, targeting multiple industries with sophisticated, evolving tactics. Our latest research (June 2025) reveals that the attackers conducted highly-targeted campaigns, which started in November 2022 and ran until November 2024.
The group mainly performed adversary-in-the-middle (AitM) attacks on specific victims. These included techniques such as dropping loaders into specific locations and storing encrypted parts of the malware on attacker-controlled servers, which were resolved as a response to specific website DNS requests. Notably, the attackers have developed a new loader that evades detection when infecting its targets, and even employed hybrid encryption practices to complicate analysis and make implants unique to each victim.
Furthermore, the group has developed an injector that allows them to execute their MgBot implant in memory by injecting it into legitimate processes. It resides in the memory space of a decade-old signed executable by using DLL sideloading and enables them to maintain a stealthy presence in compromised systems for extended periods.
The threat actor commonly uses lures that are disguised as new updates to known third-party applications or popular system applications trusted by hundreds of users over the years.
In this campaign, the attackers used an executable disguised as an update package for SohuVA, which is a streaming app developed by Sohu Inc., a Chinese internet company. The malicious package, named sohuva_update_10.2.29.1-lup-s-tp.exe, clearly impersonates a real SohuVA update to deliver malware from the following resource, as indicated by our telemetry:
There is a possibility that the attackers used a DNS poisoning attack to alter the DNS response of p2p.hd.sohu.com[.]cn to an attacker-controlled server’s IP address, while the genuine update module of the SohuVA application tries to update its binaries located in appdata\roaming\shapp\7.0.18.0\package. Although we were unable to verify this at the time of analysis, we can make an educated guess, given that it is still unknown what triggered the update mechanism.
Furthermore, our analysis of the infection process has identified several additional campaigns pursued by the same group. For example, they utilized a fake updater for the iQIYI Video application, a popular platform for streaming Asian media content similar to SohuVA. This fake updater was dropped into the application’s installation folder and executed by the legitimate service qiyiservice.exe. Upon execution, the fake updater initiated malicious activity on the victim’s system, and we have identified that the same method is used for IObit Smart Defrag and Tencent QQ applications.
The initial loader was developed in C++ using the Windows Template Library (WTL). Its code bears a strong resemblance to Wizard97Test, a WTL sample application hosted on Microsoft’s GitHub. The attackers appear to have embedded malicious code within this project to effectively conceal their malicious intentions.
The loader first decrypts the encrypted configuration buffer by employing an XOR-based decryption algorithm:
for ( index = 0; index < v6; index = (index + 1) )
{
if ( index >= 5156 )
break;
mw_configindex ^= (&mw_deflated_config + (index & 3));
}
After decryption, it decompresses the LZMA-compressed buffer into the allocated buffer, and all of the configuration is exposed, including several components:
The malware also checks the name of the logged-in user in the system and performs actions accordingly. If the username is SYSTEM, the malware copies itself with a different name by appending the ext.exe suffix inside the current working directory. Then it uses the ShellExecuteW API to execute the newly created version. Notably, all relevant strings in the malware, such as SYSTEM and ext.exe, are encrypted, and the loader decrypts them with a specific XOR algorithm.
Decryption routine of encrypted strings
If the username is not SYSTEM, the malware first copies explorer.exe into %TEMP%, naming the instance as tmpX.tmp (where X is an incremented decimal number), and then deletes the original file. The purpose of this activity is unclear, but it consumes high system resources. Next, the loader decrypts the kernel32.dll and VirtualProtect strings to retrieve their base addresses by calling the GetProcAddress API. Afterwards, it uses a single-byte XOR key to decrypt the shellcode, which is 9556 bytes long, and stores it at the same address in the .data section. Since the .data section does not have execute permission, the malware uses the VirtualProtect API to set the permission for the section. This allows for the decrypted shellcode to be executed without alerting security products by allocating new memory blocks. Before executing the shellcode, the malware prepares a 16-byte-long parameter structure that contains several items, with the most important one being the address of the encrypted MgBot configuration buffer.
Multi-stage shellcode execution
As mentioned above, the loader follows a unique delivery scheme, which includes at least two stages of payload. The shellcode employs a hashing algorithm known as PJW to resolve Windows APIs at runtime in a stealthy manner.
The shellcode first searches for a specific DAT file in the malware’s primary installation directory. If it is found, the shellcode decrypts it using the CryptUnprotectData API, a Windows API that decrypts protected data into allocated heap memory, and ensures that the data can only be decrypted on the particular machine by design. After decryption, the shellcode deletes the file to avoid leaving any traces of the valuable part of the attack chain.
If, however, the DAT file is not present, the shellcode initiates the next-stage shellcode installation process. It involves retrieving encrypted data from a web source that is actually an attacker-controlled server, by employing a DNS poisoning attack. Our telemetry shows that the attackers successfully obtained the encrypted second-stage shellcode, disguised as a PNG file, from the legitimate website dictionary[.]com. However, upon further investigation, it was discovered that the IP address associated with dictionary[.]com had been manipulated through a DNS poisoning technique. As a result, victims’ systems were resolving the website to different attacker-controlled IP addresses depending on the victims’ geographical location and internet service provider.
To retrieve the second-stage shellcode, the first-stage shellcode uses the RtlGetVersion API to obtain the current Windows version number and then appends a predefined string to the HTTP header:
sec-ch-ua-platform: windows %d.%d.%d.%d.%d.%d
This implies that the attackers needed to be able to examine request headers and respond accordingly. We suspect that the attackers’ collection of the Windows version number and its inclusion in the request headers served a specific purpose, likely allowing them to target specific operating system versions and even tailor their payload to different operating systems. Given that the Evasive Panda threat actor has been known to use distinct implants for Windows (MgBot) and macOS (Macma) in previous campaigns, it is likely that the malware uses the retrieved OS version string to determine which implant to deploy. This enables the threat actor to adapt their attack to the victim’s specific operating system by assessing results on the server side.
Downloading a payload from the web resource
From this point on, the first-stage shellcode proceeds to decrypt the retrieved payload with a XOR decryption algorithm:
key = *(mw_decryptedDataFromDatFile + 92);
index = 0;
if ( sz_shellcode )
{
mw_decryptedDataFromDatFile_1 = Heap;
do
{
*(index + mw_decryptedDataFromDatFile_1) ^= *(&key + (index & 3));
++index;
}
while ( index < sz_shellcode );
}
The shellcode uses a 4-byte XOR key, consistent with the one used in previous stages, to decrypt the new shellcode stored in the DAT file. It then creates a structure for the decrypted second-stage shellcode, similar to the first stage, including a partially decrypted configuration buffer and other relevant details.
Next, the shellcode resolves the VirtualProtect API to change the protection flag of the new shellcode buffer, allowing it to be executed with PAGE_EXECUTE_READWRITE permissions. The second-stage shellcode is then executed, with the structure passed as an argument. After the shellcode has finished running, its return value is checked to see if it matches 0x9980. Depending on the outcome, the shellcode will either terminate its own process or return control to the caller.
Although we were unable to retrieve the second-stage payload from the attackers’ web server during our analysis, we were able to capture and examine the next stage of the malware, which was to be executed afterwards. Our analysis suggests that the attackers may have used the CryptProtectData API during the execution of the second shellcode to encrypt the entire shellcode and store it as a DAT file in the malware’s main installation directory. This implies that the malware writes an encrypted DAT file to disk using the CryptProtectData API, which can then be decrypted and executed by the first-stage shellcode. Furthermore, it appears that the attacker attempted to generate a unique encrypted second shellcode file for each victim, which we believe is another technique used to evade detection and defense mechanisms in the attack chain.
Secondary loader
We identified a secondary loader, named libpython2.4.dll, which was disguised as a legitimate Windows library and used by the Evasive Panda group to achieve a stealthier loading mechanism. Notably, this malicious DLL loader relies on a legitimate, signed executable named evteng.exe (MD5: 1c36452c2dad8da95d460bee3bea365e), which is an older version of python.exe. This executable is a Python wrapper that normally imports the libpython2.4.dll library and calls the Py_Main function.
The secondary loader retrieves the full path of the current module (libpython2.4.dll) and writes it to a file named status.dat, located in C:\ProgramData\Microsoft\eHome, but only if a file with the same name does not already exist in that directory. We believe with a low-to-medium level of confidence that this action is intended to allow the attacker to potentially update the secondary loader in the future. This suggests that the attacker may be planning for future modifications or upgrades to the malware.
The malware proceeds to decrypt the next stage by reading the entire contents of C:\ProgramData\Microsoft\eHome\perf.dat. This file contains the previously downloaded and XOR-decrypted data from the attacker-controlled server, which was obtained through the DNS poisoning technique as described above. Notably, the implant downloads the payload several times and moves it between folders by renaming it. It appears that the attacker used a complex process to obtain this stage from a resource, where it was initially XOR-encrypted. The attacker then decrypted this stage with XOR and subsequently encrypted and saved it to perf.dat using a custom hybrid of Microsoft’s Data Protection Application Programming Interface (DPAPI) and the RC5 algorithm.
General overview of storing payload on disk by using hybrid encryption
This custom encryption algorithm works as follows. The RC5 encryption key is itself encrypted using Microsoft’s DPAPI and stored in the first 16 bytes of perf.dat. The RC5-encrypted payload is then appended to the file, following the encrypted key. To decrypt the payload, the process is reversed: the encrypted RC5 key is first decrypted with DPAPI, and then used to decrypt the remaining contents of perf.dat, which contains the next-stage payload.
The attacker uses this approach to ensure that a crucial part of the attack chain is secured, and the encrypted data can only be decrypted on the specific system where the encryption was initially performed. This is because the DPAPI functions used to secure the RC5 key tie the decryption process to the individual system, making it difficult for the encrypted data to be accessed or decrypted elsewhere. This makes it more challenging for defenders to intercept and analyze the malicious payload.
After completing the decryption process, the secondary loader initiates the runtime injection method, which likely involves the use of a custom runtime DLL injector for the decrypted data. The injector first calls the DLL entry point and then searches for a specific export function named preload. Although we were unable to determine which encrypted module was decrypted and executed in memory due to a lack of available data on the attacker-controlled server, our telemetry reveals that an MgBot variant is injected into the legitimate svchost.exe process after the secondary loader is executed. Fortunately, this allowed us to analyze these implants further and gain additional insights into the attack, as well as reveal that the encrypted initial configuration was passed through the infection chain, ultimately leading to the execution of MgBot. The configuration file was decrypted with a single-byte XOR key, 0x58, and this would lead to the full exposure of the configuration.
Our analysis suggests that the configuration includes a campaign name, hardcoded C2 server IP addresses, and unknown bytes that may serve as encryption or decryption keys, although our confidence in this assessment is limited. Interestingly, some of the C2 server addresses have been in use for multiple years, indicating a potential long-term operation.
Decryption of the configuration in the injected MgBot implant
Victims
Our telemetry has detected victims in Türkiye, China, and India, with some systems remaining compromised for over a year. The attackers have shown remarkable persistence, sustaining the campaign for two years (from November 2022 to November 2024) according to our telemetry, which indicates a substantial investment of resources and dedication to the operation.
Attribution
The techniques, tactics, and procedures (TTPs) employed in this compromise indicate with high confidence that the Evasive Panda threat actor is responsible for the attack. Despite the development of a new loader, which has been added to their arsenal, the decade-old MgBot implant was still identified in the final stage of the attack with new elements in its configuration. Consistent with previous research conducted by several vendors in the industry, the Evasive Panda threat actor is known to commonly utilize various techniques, such as supply-chain compromise, Adversary-in-the-Middle attacks, and watering-hole attacks, which enable them to distribute their payloads without raising suspicion.
Conclusion
The Evasive Panda threat actor has once again showcased its advanced capabilities, evading security measures with new techniques and tools while maintaining long-term persistence in targeted systems. Our investigation suggests that the attackers are continually improving their tactics, and it is likely that other ongoing campaigns exist. The introduction of new loaders may precede further updates to their arsenal.
As for the AitM attack, we do not have any reliable sources on how the threat actor delivers the initial loader, and the process of poisoning DNS responses for legitimate websites, such as dictionary[.]com, is still unknown. However, we are considering two possible scenarios based on prior research and the characteristics of the threat actor: either the ISPs used by the victims were selectively targeted, and some kind of network implant was installed on edge devices, or one of the network devices of the victims — most likely a router or firewall appliance — was targeted for this purpose. However, it is difficult to make a precise statement, as this campaign requires further attention in terms of forensic investigation, both on the ISPs and the victims.
The configuration file’s numerous C2 server IP addresses indicate a deliberate effort to maintain control over infected systems running the MgBot implant. By using multiple C2 servers, the attacker aims to ensure prolonged persistence and prevents loss of control over compromised systems, suggesting a strategic approach to sustaining their operations.
The Evasive Panda APT group (also known as Bronze Highland, Daggerfly, and StormBamboo) has been active since 2012, targeting multiple industries with sophisticated, evolving tactics. Our latest research (June 2025) reveals that the attackers conducted highly-targeted campaigns, which started in November 2022 and ran until November 2024.
The group mainly performed adversary-in-the-middle (AitM) attacks on specific victims. These included techniques such as dropping loaders into specific locations and storing encrypted parts of the malware on attacker-controlled servers, which were resolved as a response to specific website DNS requests. Notably, the attackers have developed a new loader that evades detection when infecting its targets, and even employed hybrid encryption practices to complicate analysis and make implants unique to each victim.
Furthermore, the group has developed an injector that allows them to execute their MgBot implant in memory by injecting it into legitimate processes. It resides in the memory space of a decade-old signed executable by using DLL sideloading and enables them to maintain a stealthy presence in compromised systems for extended periods.
The threat actor commonly uses lures that are disguised as new updates to known third-party applications or popular system applications trusted by hundreds of users over the years.
In this campaign, the attackers used an executable disguised as an update package for SohuVA, which is a streaming app developed by Sohu Inc., a Chinese internet company. The malicious package, named sohuva_update_10.2.29.1-lup-s-tp.exe, clearly impersonates a real SohuVA update to deliver malware from the following resource, as indicated by our telemetry:
There is a possibility that the attackers used a DNS poisoning attack to alter the DNS response of p2p.hd.sohu.com[.]cn to an attacker-controlled server’s IP address, while the genuine update module of the SohuVA application tries to update its binaries located in appdata\roaming\shapp\7.0.18.0\package. Although we were unable to verify this at the time of analysis, we can make an educated guess, given that it is still unknown what triggered the update mechanism.
Furthermore, our analysis of the infection process has identified several additional campaigns pursued by the same group. For example, they utilized a fake updater for the iQIYI Video application, a popular platform for streaming Asian media content similar to SohuVA. This fake updater was dropped into the application’s installation folder and executed by the legitimate service qiyiservice.exe. Upon execution, the fake updater initiated malicious activity on the victim’s system, and we have identified that the same method is used for IObit Smart Defrag and Tencent QQ applications.
The initial loader was developed in C++ using the Windows Template Library (WTL). Its code bears a strong resemblance to Wizard97Test, a WTL sample application hosted on Microsoft’s GitHub. The attackers appear to have embedded malicious code within this project to effectively conceal their malicious intentions.
The loader first decrypts the encrypted configuration buffer by employing an XOR-based decryption algorithm:
for ( index = 0; index < v6; index = (index + 1) )
{
if ( index >= 5156 )
break;
mw_configindex ^= (&mw_deflated_config + (index & 3));
}
After decryption, it decompresses the LZMA-compressed buffer into the allocated buffer, and all of the configuration is exposed, including several components:
The malware also checks the name of the logged-in user in the system and performs actions accordingly. If the username is SYSTEM, the malware copies itself with a different name by appending the ext.exe suffix inside the current working directory. Then it uses the ShellExecuteW API to execute the newly created version. Notably, all relevant strings in the malware, such as SYSTEM and ext.exe, are encrypted, and the loader decrypts them with a specific XOR algorithm.
Decryption routine of encrypted strings
If the username is not SYSTEM, the malware first copies explorer.exe into %TEMP%, naming the instance as tmpX.tmp (where X is an incremented decimal number), and then deletes the original file. The purpose of this activity is unclear, but it consumes high system resources. Next, the loader decrypts the kernel32.dll and VirtualProtect strings to retrieve their base addresses by calling the GetProcAddress API. Afterwards, it uses a single-byte XOR key to decrypt the shellcode, which is 9556 bytes long, and stores it at the same address in the .data section. Since the .data section does not have execute permission, the malware uses the VirtualProtect API to set the permission for the section. This allows for the decrypted shellcode to be executed without alerting security products by allocating new memory blocks. Before executing the shellcode, the malware prepares a 16-byte-long parameter structure that contains several items, with the most important one being the address of the encrypted MgBot configuration buffer.
Multi-stage shellcode execution
As mentioned above, the loader follows a unique delivery scheme, which includes at least two stages of payload. The shellcode employs a hashing algorithm known as PJW to resolve Windows APIs at runtime in a stealthy manner.
The shellcode first searches for a specific DAT file in the malware’s primary installation directory. If it is found, the shellcode decrypts it using the CryptUnprotectData API, a Windows API that decrypts protected data into allocated heap memory, and ensures that the data can only be decrypted on the particular machine by design. After decryption, the shellcode deletes the file to avoid leaving any traces of the valuable part of the attack chain.
If, however, the DAT file is not present, the shellcode initiates the next-stage shellcode installation process. It involves retrieving encrypted data from a web source that is actually an attacker-controlled server, by employing a DNS poisoning attack. Our telemetry shows that the attackers successfully obtained the encrypted second-stage shellcode, disguised as a PNG file, from the legitimate website dictionary[.]com. However, upon further investigation, it was discovered that the IP address associated with dictionary[.]com had been manipulated through a DNS poisoning technique. As a result, victims’ systems were resolving the website to different attacker-controlled IP addresses depending on the victims’ geographical location and internet service provider.
To retrieve the second-stage shellcode, the first-stage shellcode uses the RtlGetVersion API to obtain the current Windows version number and then appends a predefined string to the HTTP header:
sec-ch-ua-platform: windows %d.%d.%d.%d.%d.%d
This implies that the attackers needed to be able to examine request headers and respond accordingly. We suspect that the attackers’ collection of the Windows version number and its inclusion in the request headers served a specific purpose, likely allowing them to target specific operating system versions and even tailor their payload to different operating systems. Given that the Evasive Panda threat actor has been known to use distinct implants for Windows (MgBot) and macOS (Macma) in previous campaigns, it is likely that the malware uses the retrieved OS version string to determine which implant to deploy. This enables the threat actor to adapt their attack to the victim’s specific operating system by assessing results on the server side.
Downloading a payload from the web resource
From this point on, the first-stage shellcode proceeds to decrypt the retrieved payload with a XOR decryption algorithm:
key = *(mw_decryptedDataFromDatFile + 92);
index = 0;
if ( sz_shellcode )
{
mw_decryptedDataFromDatFile_1 = Heap;
do
{
*(index + mw_decryptedDataFromDatFile_1) ^= *(&key + (index & 3));
++index;
}
while ( index < sz_shellcode );
}
The shellcode uses a 4-byte XOR key, consistent with the one used in previous stages, to decrypt the new shellcode stored in the DAT file. It then creates a structure for the decrypted second-stage shellcode, similar to the first stage, including a partially decrypted configuration buffer and other relevant details.
Next, the shellcode resolves the VirtualProtect API to change the protection flag of the new shellcode buffer, allowing it to be executed with PAGE_EXECUTE_READWRITE permissions. The second-stage shellcode is then executed, with the structure passed as an argument. After the shellcode has finished running, its return value is checked to see if it matches 0x9980. Depending on the outcome, the shellcode will either terminate its own process or return control to the caller.
Although we were unable to retrieve the second-stage payload from the attackers’ web server during our analysis, we were able to capture and examine the next stage of the malware, which was to be executed afterwards. Our analysis suggests that the attackers may have used the CryptProtectData API during the execution of the second shellcode to encrypt the entire shellcode and store it as a DAT file in the malware’s main installation directory. This implies that the malware writes an encrypted DAT file to disk using the CryptProtectData API, which can then be decrypted and executed by the first-stage shellcode. Furthermore, it appears that the attacker attempted to generate a unique encrypted second shellcode file for each victim, which we believe is another technique used to evade detection and defense mechanisms in the attack chain.
Secondary loader
We identified a secondary loader, named libpython2.4.dll, which was disguised as a legitimate Windows library and used by the Evasive Panda group to achieve a stealthier loading mechanism. Notably, this malicious DLL loader relies on a legitimate, signed executable named evteng.exe (MD5: 1c36452c2dad8da95d460bee3bea365e), which is an older version of python.exe. This executable is a Python wrapper that normally imports the libpython2.4.dll library and calls the Py_Main function.
The secondary loader retrieves the full path of the current module (libpython2.4.dll) and writes it to a file named status.dat, located in C:\ProgramData\Microsoft\eHome, but only if a file with the same name does not already exist in that directory. We believe with a low-to-medium level of confidence that this action is intended to allow the attacker to potentially update the secondary loader in the future. This suggests that the attacker may be planning for future modifications or upgrades to the malware.
The malware proceeds to decrypt the next stage by reading the entire contents of C:\ProgramData\Microsoft\eHome\perf.dat. This file contains the previously downloaded and XOR-decrypted data from the attacker-controlled server, which was obtained through the DNS poisoning technique as described above. Notably, the implant downloads the payload several times and moves it between folders by renaming it. It appears that the attacker used a complex process to obtain this stage from a resource, where it was initially XOR-encrypted. The attacker then decrypted this stage with XOR and subsequently encrypted and saved it to perf.dat using a custom hybrid of Microsoft’s Data Protection Application Programming Interface (DPAPI) and the RC5 algorithm.
General overview of storing payload on disk by using hybrid encryption
This custom encryption algorithm works as follows. The RC5 encryption key is itself encrypted using Microsoft’s DPAPI and stored in the first 16 bytes of perf.dat. The RC5-encrypted payload is then appended to the file, following the encrypted key. To decrypt the payload, the process is reversed: the encrypted RC5 key is first decrypted with DPAPI, and then used to decrypt the remaining contents of perf.dat, which contains the next-stage payload.
The attacker uses this approach to ensure that a crucial part of the attack chain is secured, and the encrypted data can only be decrypted on the specific system where the encryption was initially performed. This is because the DPAPI functions used to secure the RC5 key tie the decryption process to the individual system, making it difficult for the encrypted data to be accessed or decrypted elsewhere. This makes it more challenging for defenders to intercept and analyze the malicious payload.
After completing the decryption process, the secondary loader initiates the runtime injection method, which likely involves the use of a custom runtime DLL injector for the decrypted data. The injector first calls the DLL entry point and then searches for a specific export function named preload. Although we were unable to determine which encrypted module was decrypted and executed in memory due to a lack of available data on the attacker-controlled server, our telemetry reveals that an MgBot variant is injected into the legitimate svchost.exe process after the secondary loader is executed. Fortunately, this allowed us to analyze these implants further and gain additional insights into the attack, as well as reveal that the encrypted initial configuration was passed through the infection chain, ultimately leading to the execution of MgBot. The configuration file was decrypted with a single-byte XOR key, 0x58, and this would lead to the full exposure of the configuration.
Our analysis suggests that the configuration includes a campaign name, hardcoded C2 server IP addresses, and unknown bytes that may serve as encryption or decryption keys, although our confidence in this assessment is limited. Interestingly, some of the C2 server addresses have been in use for multiple years, indicating a potential long-term operation.
Decryption of the configuration in the injected MgBot implant
Victims
Our telemetry has detected victims in Türkiye, China, and India, with some systems remaining compromised for over a year. The attackers have shown remarkable persistence, sustaining the campaign for two years (from November 2022 to November 2024) according to our telemetry, which indicates a substantial investment of resources and dedication to the operation.
Attribution
The techniques, tactics, and procedures (TTPs) employed in this compromise indicate with high confidence that the Evasive Panda threat actor is responsible for the attack. Despite the development of a new loader, which has been added to their arsenal, the decade-old MgBot implant was still identified in the final stage of the attack with new elements in its configuration. Consistent with previous research conducted by several vendors in the industry, the Evasive Panda threat actor is known to commonly utilize various techniques, such as supply-chain compromise, Adversary-in-the-Middle attacks, and watering-hole attacks, which enable them to distribute their payloads without raising suspicion.
Conclusion
The Evasive Panda threat actor has once again showcased its advanced capabilities, evading security measures with new techniques and tools while maintaining long-term persistence in targeted systems. Our investigation suggests that the attackers are continually improving their tactics, and it is likely that other ongoing campaigns exist. The introduction of new loaders may precede further updates to their arsenal.
As for the AitM attack, we do not have any reliable sources on how the threat actor delivers the initial loader, and the process of poisoning DNS responses for legitimate websites, such as dictionary[.]com, is still unknown. However, we are considering two possible scenarios based on prior research and the characteristics of the threat actor: either the ISPs used by the victims were selectively targeted, and some kind of network implant was installed on edge devices, or one of the network devices of the victims — most likely a router or firewall appliance — was targeted for this purpose. However, it is difficult to make a precise statement, as this campaign requires further attention in terms of forensic investigation, both on the ISPs and the victims.
The configuration file’s numerous C2 server IP addresses indicate a deliberate effort to maintain control over infected systems running the MgBot implant. By using multiple C2 servers, the attacker aims to ensure prolonged persistence and prevents loss of control over compromised systems, suggesting a strategic approach to sustaining their operations.
For the latest discoveries in cyber research for the week of 22nd December, please download our Threat Intelligence Bulletin.
TOP ATTACKS AND BREACHES
An adult content platform PornHub has disclosed a data breach linked to analytics provider Mixpanel. The breach exposed more than 200 million records related to Premium users, including email addresses, search, watch, and download histories, locations, and associated video details collected prior to 2021. Pornhub stated that no passwords, payment information, or government-issued IDs were compromised. OpenAI also acknowledged a related incident that was caused by compromise of Mixpanel. The breach has been attributed to the ShinyHunters extortion group.
SoundCloud, an online audio streaming platform, has confirmed a cyber attack that resulted in threat actors gaining unauthorized access to a database containing users’ email addresses and public profile information. The breach affected approximately 20% of SoundCloud’s users, which might impact 28 million accounts, and caused outages and VPN connection issues. The ShinyHunters extortion gang has claimed responsibility for this attack.
Autoparts giant LKQ has acknowledged a cyberattack tied to the Oracle E-Business Suite compromise. The company said personal data of over 9,070 people, including Employer Identification Numbers and Social Security numbers, was exposed.
Check Point IPS provides protection against this threat (Oracle Multiple Products Remote Code Execution)
DXS International, a British NHS technology supplier, has encountered a cyber-attack on December 14th that resulted in unauthorized access to its internal office servers, affecting internal systems but not disrupting clinical services. It remains unclear whether NHS patient data was compromised.
The University of Sydney has suffered a data breach that resulted in hackers gaining access to an online coding repository and stealing files containing personal information of staff and students. Over 27,000 individuals were affected, including names, dates of birth, phone numbers, home addresses, and job details for current and former staff, students, alumni, and affiliates.
Petróleos de Venezuela (PDVSA), Venezuela’s state oil company, has experienced a cyberattack that resulted in disruptions to its export operations and offline systems managing the country’s main crude terminal. The incident affected administrative and operational network systems, leading to a halt in cargo deliveries. The scope of data or user information compromised has not been disclosed.
Denmark’s water utility has experienced a cyber attack that resulted in a disruption of critical water infrastructure systems. The attack impacted operational control systems supporting essential services, forming part of a broader campaign of attacks targeting Denmark’s critical infrastructure and electoral environment. The Danish Defence Intelligence Service attributed the incident to the Russia affiliated group Z-Pentest.
VULNERABILITIES AND PATCHES
Critical severity vulnerability with a CVSS score of 10.0 was disclosed in HPE OneView Software. The flaw, CVE-2025-37164, allows unauthenticated remote code execution and affects all versions prior to 11.00, including versions 5.20 through 10.20. Successful exploitation could enable a remote attacker to execute arbitrary code on affected centralized IT infrastructure management systems.
Check Point IPS provides protection against this threat (HPE OneView Remote Code Execution (CVE-2025-37164))
A critical remote code execution vulnerability, CVE-2025-14733, in WatchGuard Firebox firewalls running Fireware OS 11.x and later is being actively exploited. The out-of-bounds write flaw enables unauthenticated remote code execution on unpatched devices with IKEv2, without user interaction.
Researchers spotted active exploitation of CVE-2025-59718 and CVE-2025-59719, critical authentication bypass flaws in Fortinet FortiGate, FortiOS, FortiWeb, FortiProxy, and FortiSwitchManager. Attackers can log in without credentials and export full device configurations, risking cracked passwords.
THREAT INTELLIGENCE REPORTS
Check Point Research revealed a sophisticated wave of attacks attributed to the Chinese threat actor Ink Dragon, which targets European governments while continuing campaigns in Southeast Asia and South America. The threat actor converts compromised IIS servers into relay nodes with ShadowPad, exploits predictable configuration keys for access, and deploys a new FinalDraft backdoor for exfiltration and lateral movement.
Check Point Threat Emulation and Harmony Endpoint provide protection against this threat
Check Point Research analyzed GachiLoader, a Node.js–based malware loader observed in a campaign linked to the YouTube Ghost Network. The campaign is notable for extensive obfuscation and a previously undocumented PE injection technique. GachiLoader deploys a second-stage loader, Kidkadi, which abuses Vectored Exception Handling (VEH) in a novel method, dubbed Vectored Overloading, to load its malicious payload.
Check Point Threat Emulation and Harmony Endpoint provide protection against this threat
Check Point Research noticed a surge in darknet campaigns recruiting insiders at banks, crypto exchanges, telecoms, and major tech firms to sell access and data. Listings advertise payouts of $3,000 to $15,000, offer datasets like 37 million records for $25,000, and solicit telecom staff for SIM swapping to bypass two-factor authentication.
Check Point researchers updated on a global surge in AI-driven holiday scams across phishing, fake retail sites, and social media giveaways. They recorded 33,502 phishing emails in two weeks and over 10,000 daily ads impersonating delivery brands like Royal Mail, FedEx, UPS and DPD, while AI chatbots help fraudulent stores appear credible.
Known since 2014, the Cloud Atlas group targets countries in Eastern Europe and Central Asia. Infections occur via phishing emails containing a malicious document that exploits an old vulnerability in the Microsoft Office Equation Editor process (CVE-2018-0802) to download and execute malicious code. In this report, we describe the infection chain and tools that the group used in the first half of 2025, with particular focus on previously undescribed implants.
The starting point is typically a phishing email with a malicious DOC(X) attachment. When the document is opened, a malicious template is downloaded from a remote server. The document has the form of an RTF file containing an exploit for the formula editor, which downloads and executes an HTML Application (HTA) file.
Fpaylo
Malicious template with the exploit loaded by Word when opening the document
We were unable to obtain the actual RTF template with the exploit. We assume that after a successful infection of the victim, the link to this file becomes inaccessible. In the given example, the malicious RTF file containing the exploit was downloaded from the URL hxxps://securemodem[.]com?tzak.html_anacid.
Template files, like HTA files, are located on servers controlled by the group, and their downloading is limited both in time and by the IP addresses of the victims. The malicious HTA file extracts and creates several VBS files on disk that are parts of the VBShower backdoor. VBShower then downloads and installs other backdoors: PowerShower, VBCloud, and CloudAtlas.
Several implants remain the same, with insignificant changes in file names, and so on. You can find more details in our previous article on the following implants:
In this research, we’ll focus on new and updated components.
VBShower
VBShower::Backdoor
Compared to the previous version, the backdoor runs additional downloaded VB scripts in the current context, regardless of the size. A previous modification of this script checked the size of the payload, and if it exceeded 1 MB, instead of executing it in the current context, the backdoor wrote it to disk and used the wscript utility to launch it.
VBShower::Payload (1)
The script collects information about running processes, including their creation time, caption, and command line. The collected information is encrypted and sent to the C2 server by the parent script (VBShower::Backdoor) via the v_buff variable.
VBShower::Payload (1)
VBShower::Payload (2)
The script is used to install the VBCloud implant. First, it downloads a ZIP archive from the hardcoded URL and unpacks it into the %Public% directory. Then, it creates a scheduler task named “MicrosoftEdgeUpdateTask” to run the following command line:
It renames the unzipped file %Public%\Libraries\v.log to %Public%\Libraries\MicrosoftEdgeUpdate.vbs, iterates through the files in the %Public%\Libraries directory, and collects information about the filenames and sizes. The data, in the form of a buffer, is collected in the v_buff variable. The malware gets information about the task by executing the following command line:
The specified command line is executed, with the output redirected to the TMP file. Both the TMP file and the content of the v_buff variable will be sent to the C2 server by the parent script (VBShower::Backdoor).
Here is an example of the information present in the v_buff variable:
The file MicrosoftEdgeUpdate.vbs is a launcher for VBCloud, which reads the encrypted body of the backdoor from the file upgrade.mds, decrypts it, and executes it.
VBShower::Payload (2) used to install VBCloud
Almost the same script is used to install the CloudAtlas backdoor on an infected system. The script only downloads and unpacks the ZIP archive to "%LOCALAPPDATA%", and sends information about the contents of the directories "%LOCALAPPDATA%\vlc\plugins\access" and "%LOCALAPPDATA%\vlc" as output.
In this case, the file renaming operation is not applied, and there is no code for creating a scheduler task.
Here is an example of information to be sent to the C2 server:
In fact, a.xml, d.xml, and e.xml are the executable file and libraries, respectively, of VLC Media Player. The c.xml file is a malicious library used in a DLL hijacking attack, where VLC acts as a loader, and the b.xml file is an encrypted body of the CloudAtlas backdoor, read from disk by the malicious library, decrypted, and executed.
VBShower::Payload (2) used to install CloudAtlas
VBShower::Payload (3)
This script is the next component for installing CloudAtlas. It is downloaded by VBShower from the C2 server as a separate file and executed after the VBShower::Payload (2) script. The script renames the XML files unpacked by VBShower::Payload (2) from the archive to the corresponding executables and libraries, and also renames the file containing the encrypted backdoor body.
These files are copied by VBShower::Payload (3) to the following paths:
Additionally, VBShower::Payload (3) creates a scheduler task to execute the command line: "%LOCALAPPDATA%\vlc\vlc.exe". The script then iterates through the files in the "%LOCALAPPDATA%\vlc" and "%LOCALAPPDATA%\vlc\plugins\access" directories, collecting information about filenames and sizes. The data, in the form of a buffer, is collected in the v_buff variable. The script also retrieves information about the task by executing the following command line, with the output redirected to a TMP file:
This script is used to check access to various cloud services and executed before installing VBCloud or CloudAtlas. It consistently accesses the URLs of cloud services, and the received HTTP responses are saved to the v_buff variable for subsequent sending to the C2 server. A truncated example of the information sent to the C2 server:
This is a small script for checking the accessibility of PowerShower’s C2 from an infected system.
VBShower::Payload (7)
VBShower::Payload (8)
This script is used to install PowerShower, another backdoor known to be employed by Cloud Atlas. The script does so by performing the following steps in sequence:
Creates registry keys to make the console window appear off-screen, effectively hiding it:
Decrypts the contents of the embedded data block with XOR and saves the resulting script to the file "%APPDATA%\Adobe\p.txt". Then, renames the file "p.txt" to "AdobeMon.ps1".
Collects information about file names and sizes in the path "%APPDATA%\Adobe". Gets information about the task by executing the following command line, with the output redirected to a TMP file:
cmd.exe /c schtasks /query /v /fo LIST /tn MicrosoftAdobeUpdateTaskMachine
VBShower::Payload (8) used to install PowerShower
The decrypted PowerShell script is disguised as one of the standard modules, but at the end of the script, there is a command to launch the PowerShell interpreter with another script encoded in Base64.
Content of AdobeMon.ps1 (PowerShower)
VBShower::Payload (9)
This is a small script for collecting information about the system proxy settings.
VBShower::Payload (9)
VBCloud
On an infected system, VBCloud is represented by two files: a VB script (VBCloud::Launcher) and an encrypted main body (VBCloud::Backdoor). In the described case, the launcher is located in the file MicrosoftEdgeUpdate.vbs, and the payload — in upgrade.mds.
VBCloud::Launcher
The launcher script reads the contents of the upgrade.mds file, decodes characters delimited with “%H”, uses the RC4 stream encryption algorithm with a key built into the script to decrypt it, and transfers control to the decrypted content. It is worth noting that the implementation of RC4 uses PRGA (pseudo-random generation algorithm), which is quite rare, since most malware implementations of this algorithm skip this step.
VBCloud::Launcher
VBCloud::Backdoor
The backdoor performs several actions in a loop to eventually download and execute additional malicious scripts, as described in the previous research.
VBCloud::Payload (FileGrabber)
Unlike VBShower, which uses a global variable to save its output or a temporary file to be sent to the C2 server, each VBCloud payload communicates with the C2 server independently. One of the most commonly used payloads for the VBCloud backdoor is FileGrabber. The script exfiltrates files and documents from the target system as described before.
The FileGrabber payload has the following limitations when scanning for files:
It ignores the following paths:
Program Files
Program Files (x86)
%SystemRoot%
The file size for archiving must be between 1,000 and 3,000,000 bytes.
The file’s last modification date must be less than 30 days before the start of the scan.
Files containing the following strings in their names are ignored:
“intermediate.txt”
“FlightingLogging.txt”
“log.txt”
“thirdpartynotices”
“ThirdPartyNotices”
“easylist.txt”
“acroNGLLog.txt”
“LICENSE.txt”
“signature.txt”
“AlternateServices.txt”
“scanwia.txt”
“scantwain.txt”
“SiteSecurityServiceState.txt”
“serviceworker.txt”
“SettingsCache.txt”
“NisLog.txt”
“AppCache”
“backupTest”
Part of VBCloud::Payload (FileGrabber)
PowerShower
As mentioned above, PowerShower is installed via one of the VBShower payloads. This script launches the PowerShell interpreter with another script encoded in Base64. Running in an infinite loop, it attempts to access the C2 server to retrieve an additional payload, which is a PowerShell script twice encoded with Base64. This payload is executed in the context of the backdoor, and the execution result is sent to the C2 server via an HTTP POST request.
Decoded PowerShower script
In previous versions of PowerShower, the payload created a sapp.xtx temporary file to save its output, which was sent to the C2 server by the main body of the backdoor. No intermediate files are created anymore, and the result of execution is returned to the backdoor by a normal call to the "return" operator.
PowerShower::Payload (1)
This script was previously described as PowerShower::Payload (2). This payload is unique to each victim.
PowerShower::Payload (2)
This script is used for grabbing files with metadata from a network share.
PowerShower::Payload (2)
CloudAtlas
As described above, the CloudAtlas backdoor is installed via VBShower from a downloaded archive delivered through a DLL hijacking attack. The legitimate VLC application acts as a loader, accompanied by a malicious library that reads the encrypted payload from the file and transfers control to it. The malicious DLL is located at "%LOCALAPPDATA%\vlc\plugins\access", while the file with the encrypted payload is located at "%LOCALAPPDATA%\vlc\".
When the malicious DLL gains control, it first extracts another DLL from itself, places it in the memory of the current process, and transfers control to it. The unpacked DLL uses a byte-by-byte XOR operation to decrypt the block with the loader configuration. The encrypted config immediately follows the key. The config specifies the name of the event that is created to prevent a duplicate payload launch. The config also contains the name of the file where the encrypted payload is located — "chambranle" in this case — and the decryption key itself.
Encrypted and decrypted loader configuration
The library reads the contents of the "chambranle" file with the payload, uses the key from the decrypted config and the IV located at the very end of the "chambranle" file to decrypt it with AES-256-CBC. The decrypted file is another DLL with its size and SHA-1 hash embedded at the end, added to verify that the DLL is decrypted correctly. The DLL decrypted from "chambranle" is the main body of the CloudAtlas backdoor, and control is transferred to it via one of the exported functions, specifically the one with ordinal 2.
Main routine that processes the payload file
When the main body of the backdoor gains control, the first thing it does is decrypt its own configuration. Decryption is done in a similar way, using AES-256-CBC. The key for AES-256 is located before the configuration, and the IV is located right after it. The most useful information in the configuration file includes the URL of the cloud service, paths to directories for receiving payloads and unloading results, and credentials for the cloud service.
Encrypted and decrypted CloudAtlas backdoor config
Immediately after decrypting the configuration, the backdoor starts interacting with the C2 server, which is a cloud service, via WebDAV. First, the backdoor uses the MKCOL HTTP method to create two directories: one ("/guessed/intershop/Euskalduns/") will regularly receive a beacon in the form of an encrypted file containing information about the system, time, user name, current command line, and volume information. The other directory ("/cancrenate/speciesists/") is used to retrieve payloads. The beacon file and payload files are AES-256-CBC encrypted with the key that was used for backdoor configuration decryption.
HTTP requests of the CloudAtlas backdoor
The backdoor uses the HTTP PROPFIND method to retrieve the list of files. Each of these files will be subsequently downloaded, deleted from the cloud service, decrypted, and executed.
HTTP requests from the CloudAtlas backdoor
The payload consists of data with a binary block containing a command number and arguments at the beginning, followed by an executable plugin in the form of a DLL. The structure of the arguments depends on the type of command. After the plugin is loaded into memory and configured, the backdoor calls the exported function with ordinal 1, passing several arguments: a pointer to the backdoor function that implements sending files to the cloud service, a pointer to the decrypted backdoor configuration, and a pointer to the binary block with the command and arguments from the beginning of the payload.
Plugin setup and execution routine
Before calling the plugin function, the backdoor saves the path to the current directory and restores it after the function is executed. Additionally, after execution, the plugin is removed from memory.
CloudAtlas::Plugin (FileGrabber)
FileGrabber is the most commonly used plugin. As the name suggests, it is designed to steal files from an infected system. Depending on the command block transmitted, it is capable of:
Stealing files from all local disks
Stealing files from the specified removable media
Stealing files from specified folders
Using the selected username and password from the command block to mount network resources and then steal files from them
For each detected file, a series of rules are generated based on the conditions passed within the command block, including:
Checking for minimum and maximum file size
Checking the file’s last modification time
Checking the file path for pattern exclusions. If a string pattern is found in the full path to a file, the file is ignored
Checking the file name or extension against a list of patterns
Resource scanning
If all conditions match, the file is sent to the C2 server, along with its metadata, including attributes, creation time, last access time, last modification time, size, full path to the file, and SHA-1 of the file contents. Additionally, if a special flag is set in one of the rule fields, the file will be deleted after a copy is sent to the C2 server. There is also a limit on the total amount of data sent, and if this limit is exceeded, scanning of the resource stops.
Generating data for sending to C2
CloudAtlas::Plugin (Common)
This is a general-purpose plugin, which parses the transferred block, splits it into commands, and executes them. Each command has its own ID, ranging from 0 to 6. The list of commands is presented below.
Command ID 0: Creates, sets and closes named events.
Command ID 1: Deletes the selected list of files.
Command ID 2: Drops a file on disk with content and a path selected in the command block arguments.
Command ID 3: Capable of performing several operations together or independently, including:
Dropping several files on disk with content and paths selected in the command block arguments
Dropping and executing a file at a specified path with selected parameters. This operation supports three types of launch:
Using the WinExec function
Using the ShellExecuteW function
Using the CreateProcessWithLogonW function, which requires that the user’s credentials be passed within the command block to launch the process on their behalf
Command ID 4: Uses the StdRegProv COM interface to perform registry manipulations, supporting key creation, value deletion, and value setting (both DWORD and string values).
Command ID 5: Calls the ExitProcess function.
Command ID 6: Uses the credentials passed within the command block to connect a network resource, drops a file to the remote resource under the name specified within the command block, creates and runs a VB script on the local system to execute the dropped file on the remote system. The VB script is created at "%APPDATA%\ntsystmp.vbs". The path to launch the file dropped on the remote system is passed to the launched VB script as an argument.
Content of the dropped VBS
CloudAtlas::Plugin (PasswordStealer)
This plugin is used to steal cookies and credentials from browsers. This is an extended version of the Common Plugin, which is used for more specific purposes. It can also drop, launch, and delete files, but its primary function is to drop files belonging to the “Chrome App-Bound Encryption Decryption” open-source project onto the disk, and run the utility to steal cookies and passwords from Chromium-based browsers. After launching the utility, several files ("cookies.txt" and "passwords.txt") containing the extracted browser data are created on disk. The plugin then reads JSON data from the selected files, parses the data, and sends the extracted information to the C2 server.
Part of the function for parsing JSON and sending the extracted data to C2
CloudAtlas::Plugin (InfoCollector)
This plugin is used to collect information about the infected system. The list of commands is presented below.
Command ID 0xFFFFFFF0: Collects the computer’s NetBIOS name and domain information.
Command ID 0xFFFFFFF1: Gets a list of processes, including full paths to executable files of processes, and a list of modules (DLLs) loaded into each process.
Command ID 0xFFFFFFF2: Collects information about installed products.
Command ID 0xFFFFFFF3: Collects device information.
Command ID 0xFFFFFFF4: Collects information about logical drives.
Command ID 0xFFFFFFF5: Executes the command with input/output redirection, and sends the output to the C2 server. If the command line for execution is not specified, it sequentially launches the following utilities and sends their output to the C2 server:
net group "Exchange servers" /domain
Ipconfig
arp -a
Python script
As mentioned in one of our previous reports, Cloud Atlas uses a custom Python script named get_browser_pass.py to extract saved credentials from browsers on infected systems. If the Python interpreter is not present on the victim’s machine, the group delivers an archive that includes both the script and a bundled Python interpreter to ensure execution.
During one of the latest incidents we investigated, we once again observed traces of this tool in action, specifically the presence of the file "C:\ProgramData\py\pytest.dll".
The pytest.dll library is called from within get_browser_pass.py and used to extract credentials from Yandex Browser. The data is then saved locally to a file named y3.txt.
Victims
According to our telemetry, the identified targets of the malicious activities described here are located in Russia and Belarus, with observed activity dating back to the beginning of 2025. The industries being targeted are diverse, encompassing organizations in the telecommunications sector, construction, government entities, and plants.
Conclusion
For more than ten years, the group has carried on its activities and expanded its arsenal. Now the attackers have four implants at their disposal (PowerShower, VBShower, VBCloud, CloudAtlas), each of them a full-fledged backdoor. Most of the functionality in the backdoors is duplicated, but some payloads provide various exclusive capabilities. The use of cloud services to manage backdoors is a distinctive feature of the group, and it has proven itself in various attacks.
Indicators of compromise
Note: The indicators in this section are valid at the time of publication.
Known since 2014, the Cloud Atlas group targets countries in Eastern Europe and Central Asia. Infections occur via phishing emails containing a malicious document that exploits an old vulnerability in the Microsoft Office Equation Editor process (CVE-2018-0802) to download and execute malicious code. In this report, we describe the infection chain and tools that the group used in the first half of 2025, with particular focus on previously undescribed implants.
The starting point is typically a phishing email with a malicious DOC(X) attachment. When the document is opened, a malicious template is downloaded from a remote server. The document has the form of an RTF file containing an exploit for the formula editor, which downloads and executes an HTML Application (HTA) file.
Fpaylo
Malicious template with the exploit loaded by Word when opening the document
We were unable to obtain the actual RTF template with the exploit. We assume that after a successful infection of the victim, the link to this file becomes inaccessible. In the given example, the malicious RTF file containing the exploit was downloaded from the URL hxxps://securemodem[.]com?tzak.html_anacid.
Template files, like HTA files, are located on servers controlled by the group, and their downloading is limited both in time and by the IP addresses of the victims. The malicious HTA file extracts and creates several VBS files on disk that are parts of the VBShower backdoor. VBShower then downloads and installs other backdoors: PowerShower, VBCloud, and CloudAtlas.
Several implants remain the same, with insignificant changes in file names, and so on. You can find more details in our previous article on the following implants:
In this research, we’ll focus on new and updated components.
VBShower
VBShower::Backdoor
Compared to the previous version, the backdoor runs additional downloaded VB scripts in the current context, regardless of the size. A previous modification of this script checked the size of the payload, and if it exceeded 1 MB, instead of executing it in the current context, the backdoor wrote it to disk and used the wscript utility to launch it.
VBShower::Payload (1)
The script collects information about running processes, including their creation time, caption, and command line. The collected information is encrypted and sent to the C2 server by the parent script (VBShower::Backdoor) via the v_buff variable.
VBShower::Payload (1)
VBShower::Payload (2)
The script is used to install the VBCloud implant. First, it downloads a ZIP archive from the hardcoded URL and unpacks it into the %Public% directory. Then, it creates a scheduler task named “MicrosoftEdgeUpdateTask” to run the following command line:
It renames the unzipped file %Public%\Libraries\v.log to %Public%\Libraries\MicrosoftEdgeUpdate.vbs, iterates through the files in the %Public%\Libraries directory, and collects information about the filenames and sizes. The data, in the form of a buffer, is collected in the v_buff variable. The malware gets information about the task by executing the following command line:
The specified command line is executed, with the output redirected to the TMP file. Both the TMP file and the content of the v_buff variable will be sent to the C2 server by the parent script (VBShower::Backdoor).
Here is an example of the information present in the v_buff variable:
The file MicrosoftEdgeUpdate.vbs is a launcher for VBCloud, which reads the encrypted body of the backdoor from the file upgrade.mds, decrypts it, and executes it.
VBShower::Payload (2) used to install VBCloud
Almost the same script is used to install the CloudAtlas backdoor on an infected system. The script only downloads and unpacks the ZIP archive to "%LOCALAPPDATA%", and sends information about the contents of the directories "%LOCALAPPDATA%\vlc\plugins\access" and "%LOCALAPPDATA%\vlc" as output.
In this case, the file renaming operation is not applied, and there is no code for creating a scheduler task.
Here is an example of information to be sent to the C2 server:
In fact, a.xml, d.xml, and e.xml are the executable file and libraries, respectively, of VLC Media Player. The c.xml file is a malicious library used in a DLL hijacking attack, where VLC acts as a loader, and the b.xml file is an encrypted body of the CloudAtlas backdoor, read from disk by the malicious library, decrypted, and executed.
VBShower::Payload (2) used to install CloudAtlas
VBShower::Payload (3)
This script is the next component for installing CloudAtlas. It is downloaded by VBShower from the C2 server as a separate file and executed after the VBShower::Payload (2) script. The script renames the XML files unpacked by VBShower::Payload (2) from the archive to the corresponding executables and libraries, and also renames the file containing the encrypted backdoor body.
These files are copied by VBShower::Payload (3) to the following paths:
Additionally, VBShower::Payload (3) creates a scheduler task to execute the command line: "%LOCALAPPDATA%\vlc\vlc.exe". The script then iterates through the files in the "%LOCALAPPDATA%\vlc" and "%LOCALAPPDATA%\vlc\plugins\access" directories, collecting information about filenames and sizes. The data, in the form of a buffer, is collected in the v_buff variable. The script also retrieves information about the task by executing the following command line, with the output redirected to a TMP file:
This script is used to check access to various cloud services and executed before installing VBCloud or CloudAtlas. It consistently accesses the URLs of cloud services, and the received HTTP responses are saved to the v_buff variable for subsequent sending to the C2 server. A truncated example of the information sent to the C2 server:
This is a small script for checking the accessibility of PowerShower’s C2 from an infected system.
VBShower::Payload (7)
VBShower::Payload (8)
This script is used to install PowerShower, another backdoor known to be employed by Cloud Atlas. The script does so by performing the following steps in sequence:
Creates registry keys to make the console window appear off-screen, effectively hiding it:
Decrypts the contents of the embedded data block with XOR and saves the resulting script to the file "%APPDATA%\Adobe\p.txt". Then, renames the file "p.txt" to "AdobeMon.ps1".
Collects information about file names and sizes in the path "%APPDATA%\Adobe". Gets information about the task by executing the following command line, with the output redirected to a TMP file:
cmd.exe /c schtasks /query /v /fo LIST /tn MicrosoftAdobeUpdateTaskMachine
VBShower::Payload (8) used to install PowerShower
The decrypted PowerShell script is disguised as one of the standard modules, but at the end of the script, there is a command to launch the PowerShell interpreter with another script encoded in Base64.
Content of AdobeMon.ps1 (PowerShower)
VBShower::Payload (9)
This is a small script for collecting information about the system proxy settings.
VBShower::Payload (9)
VBCloud
On an infected system, VBCloud is represented by two files: a VB script (VBCloud::Launcher) and an encrypted main body (VBCloud::Backdoor). In the described case, the launcher is located in the file MicrosoftEdgeUpdate.vbs, and the payload — in upgrade.mds.
VBCloud::Launcher
The launcher script reads the contents of the upgrade.mds file, decodes characters delimited with “%H”, uses the RC4 stream encryption algorithm with a key built into the script to decrypt it, and transfers control to the decrypted content. It is worth noting that the implementation of RC4 uses PRGA (pseudo-random generation algorithm), which is quite rare, since most malware implementations of this algorithm skip this step.
VBCloud::Launcher
VBCloud::Backdoor
The backdoor performs several actions in a loop to eventually download and execute additional malicious scripts, as described in the previous research.
VBCloud::Payload (FileGrabber)
Unlike VBShower, which uses a global variable to save its output or a temporary file to be sent to the C2 server, each VBCloud payload communicates with the C2 server independently. One of the most commonly used payloads for the VBCloud backdoor is FileGrabber. The script exfiltrates files and documents from the target system as described before.
The FileGrabber payload has the following limitations when scanning for files:
It ignores the following paths:
Program Files
Program Files (x86)
%SystemRoot%
The file size for archiving must be between 1,000 and 3,000,000 bytes.
The file’s last modification date must be less than 30 days before the start of the scan.
Files containing the following strings in their names are ignored:
“intermediate.txt”
“FlightingLogging.txt”
“log.txt”
“thirdpartynotices”
“ThirdPartyNotices”
“easylist.txt”
“acroNGLLog.txt”
“LICENSE.txt”
“signature.txt”
“AlternateServices.txt”
“scanwia.txt”
“scantwain.txt”
“SiteSecurityServiceState.txt”
“serviceworker.txt”
“SettingsCache.txt”
“NisLog.txt”
“AppCache”
“backupTest”
Part of VBCloud::Payload (FileGrabber)
PowerShower
As mentioned above, PowerShower is installed via one of the VBShower payloads. This script launches the PowerShell interpreter with another script encoded in Base64. Running in an infinite loop, it attempts to access the C2 server to retrieve an additional payload, which is a PowerShell script twice encoded with Base64. This payload is executed in the context of the backdoor, and the execution result is sent to the C2 server via an HTTP POST request.
Decoded PowerShower script
In previous versions of PowerShower, the payload created a sapp.xtx temporary file to save its output, which was sent to the C2 server by the main body of the backdoor. No intermediate files are created anymore, and the result of execution is returned to the backdoor by a normal call to the "return" operator.
PowerShower::Payload (1)
This script was previously described as PowerShower::Payload (2). This payload is unique to each victim.
PowerShower::Payload (2)
This script is used for grabbing files with metadata from a network share.
PowerShower::Payload (2)
CloudAtlas
As described above, the CloudAtlas backdoor is installed via VBShower from a downloaded archive delivered through a DLL hijacking attack. The legitimate VLC application acts as a loader, accompanied by a malicious library that reads the encrypted payload from the file and transfers control to it. The malicious DLL is located at "%LOCALAPPDATA%\vlc\plugins\access", while the file with the encrypted payload is located at "%LOCALAPPDATA%\vlc\".
When the malicious DLL gains control, it first extracts another DLL from itself, places it in the memory of the current process, and transfers control to it. The unpacked DLL uses a byte-by-byte XOR operation to decrypt the block with the loader configuration. The encrypted config immediately follows the key. The config specifies the name of the event that is created to prevent a duplicate payload launch. The config also contains the name of the file where the encrypted payload is located — "chambranle" in this case — and the decryption key itself.
Encrypted and decrypted loader configuration
The library reads the contents of the "chambranle" file with the payload, uses the key from the decrypted config and the IV located at the very end of the "chambranle" file to decrypt it with AES-256-CBC. The decrypted file is another DLL with its size and SHA-1 hash embedded at the end, added to verify that the DLL is decrypted correctly. The DLL decrypted from "chambranle" is the main body of the CloudAtlas backdoor, and control is transferred to it via one of the exported functions, specifically the one with ordinal 2.
Main routine that processes the payload file
When the main body of the backdoor gains control, the first thing it does is decrypt its own configuration. Decryption is done in a similar way, using AES-256-CBC. The key for AES-256 is located before the configuration, and the IV is located right after it. The most useful information in the configuration file includes the URL of the cloud service, paths to directories for receiving payloads and unloading results, and credentials for the cloud service.
Encrypted and decrypted CloudAtlas backdoor config
Immediately after decrypting the configuration, the backdoor starts interacting with the C2 server, which is a cloud service, via WebDAV. First, the backdoor uses the MKCOL HTTP method to create two directories: one ("/guessed/intershop/Euskalduns/") will regularly receive a beacon in the form of an encrypted file containing information about the system, time, user name, current command line, and volume information. The other directory ("/cancrenate/speciesists/") is used to retrieve payloads. The beacon file and payload files are AES-256-CBC encrypted with the key that was used for backdoor configuration decryption.
HTTP requests of the CloudAtlas backdoor
The backdoor uses the HTTP PROPFIND method to retrieve the list of files. Each of these files will be subsequently downloaded, deleted from the cloud service, decrypted, and executed.
HTTP requests from the CloudAtlas backdoor
The payload consists of data with a binary block containing a command number and arguments at the beginning, followed by an executable plugin in the form of a DLL. The structure of the arguments depends on the type of command. After the plugin is loaded into memory and configured, the backdoor calls the exported function with ordinal 1, passing several arguments: a pointer to the backdoor function that implements sending files to the cloud service, a pointer to the decrypted backdoor configuration, and a pointer to the binary block with the command and arguments from the beginning of the payload.
Plugin setup and execution routine
Before calling the plugin function, the backdoor saves the path to the current directory and restores it after the function is executed. Additionally, after execution, the plugin is removed from memory.
CloudAtlas::Plugin (FileGrabber)
FileGrabber is the most commonly used plugin. As the name suggests, it is designed to steal files from an infected system. Depending on the command block transmitted, it is capable of:
Stealing files from all local disks
Stealing files from the specified removable media
Stealing files from specified folders
Using the selected username and password from the command block to mount network resources and then steal files from them
For each detected file, a series of rules are generated based on the conditions passed within the command block, including:
Checking for minimum and maximum file size
Checking the file’s last modification time
Checking the file path for pattern exclusions. If a string pattern is found in the full path to a file, the file is ignored
Checking the file name or extension against a list of patterns
Resource scanning
If all conditions match, the file is sent to the C2 server, along with its metadata, including attributes, creation time, last access time, last modification time, size, full path to the file, and SHA-1 of the file contents. Additionally, if a special flag is set in one of the rule fields, the file will be deleted after a copy is sent to the C2 server. There is also a limit on the total amount of data sent, and if this limit is exceeded, scanning of the resource stops.
Generating data for sending to C2
CloudAtlas::Plugin (Common)
This is a general-purpose plugin, which parses the transferred block, splits it into commands, and executes them. Each command has its own ID, ranging from 0 to 6. The list of commands is presented below.
Command ID 0: Creates, sets and closes named events.
Command ID 1: Deletes the selected list of files.
Command ID 2: Drops a file on disk with content and a path selected in the command block arguments.
Command ID 3: Capable of performing several operations together or independently, including:
Dropping several files on disk with content and paths selected in the command block arguments
Dropping and executing a file at a specified path with selected parameters. This operation supports three types of launch:
Using the WinExec function
Using the ShellExecuteW function
Using the CreateProcessWithLogonW function, which requires that the user’s credentials be passed within the command block to launch the process on their behalf
Command ID 4: Uses the StdRegProv COM interface to perform registry manipulations, supporting key creation, value deletion, and value setting (both DWORD and string values).
Command ID 5: Calls the ExitProcess function.
Command ID 6: Uses the credentials passed within the command block to connect a network resource, drops a file to the remote resource under the name specified within the command block, creates and runs a VB script on the local system to execute the dropped file on the remote system. The VB script is created at "%APPDATA%\ntsystmp.vbs". The path to launch the file dropped on the remote system is passed to the launched VB script as an argument.
Content of the dropped VBS
CloudAtlas::Plugin (PasswordStealer)
This plugin is used to steal cookies and credentials from browsers. This is an extended version of the Common Plugin, which is used for more specific purposes. It can also drop, launch, and delete files, but its primary function is to drop files belonging to the “Chrome App-Bound Encryption Decryption” open-source project onto the disk, and run the utility to steal cookies and passwords from Chromium-based browsers. After launching the utility, several files ("cookies.txt" and "passwords.txt") containing the extracted browser data are created on disk. The plugin then reads JSON data from the selected files, parses the data, and sends the extracted information to the C2 server.
Part of the function for parsing JSON and sending the extracted data to C2
CloudAtlas::Plugin (InfoCollector)
This plugin is used to collect information about the infected system. The list of commands is presented below.
Command ID 0xFFFFFFF0: Collects the computer’s NetBIOS name and domain information.
Command ID 0xFFFFFFF1: Gets a list of processes, including full paths to executable files of processes, and a list of modules (DLLs) loaded into each process.
Command ID 0xFFFFFFF2: Collects information about installed products.
Command ID 0xFFFFFFF3: Collects device information.
Command ID 0xFFFFFFF4: Collects information about logical drives.
Command ID 0xFFFFFFF5: Executes the command with input/output redirection, and sends the output to the C2 server. If the command line for execution is not specified, it sequentially launches the following utilities and sends their output to the C2 server:
net group "Exchange servers" /domain
Ipconfig
arp -a
Python script
As mentioned in one of our previous reports, Cloud Atlas uses a custom Python script named get_browser_pass.py to extract saved credentials from browsers on infected systems. If the Python interpreter is not present on the victim’s machine, the group delivers an archive that includes both the script and a bundled Python interpreter to ensure execution.
During one of the latest incidents we investigated, we once again observed traces of this tool in action, specifically the presence of the file "C:\ProgramData\py\pytest.dll".
The pytest.dll library is called from within get_browser_pass.py and used to extract credentials from Yandex Browser. The data is then saved locally to a file named y3.txt.
Victims
According to our telemetry, the identified targets of the malicious activities described here are located in Russia and Belarus, with observed activity dating back to the beginning of 2025. The industries being targeted are diverse, encompassing organizations in the telecommunications sector, construction, government entities, and plants.
Conclusion
For more than ten years, the group has carried on its activities and expanded its arsenal. Now the attackers have four implants at their disposal (PowerShower, VBShower, VBCloud, CloudAtlas), each of them a full-fledged backdoor. Most of the functionality in the backdoors is duplicated, but some payloads provide various exclusive capabilities. The use of cloud services to manage backdoors is a distinctive feature of the group, and it has proven itself in various attacks.
Indicators of compromise
Note: The indicators in this section are valid at the time of publication.
In March 2025, we discovered Operation ForumTroll, a series of sophisticated cyberattacks exploiting the CVE-2025-2783 vulnerability in Google Chrome. We previously detailed the malicious implants used in the operation: the LeetAgent backdoor and the complex spyware Dante, developed by Memento Labs (formerly Hacking Team). However, the attackers behind this operation didn’t stop at their spring campaign and have continued to infect targets within the Russian Federation.
In October 2025, just days before we presented our report detailing the ForumTroll APT group’s attack at the Security Analyst Summit, we detected a new targeted phishing campaign by the same group. However, while the spring cyberattacks focused on organizations, the fall campaign honed in on specific individuals: scholars in the field of political science, international relations, and global economics, working at major Russian universities and research institutions.
The emails received by the victims were sent from the address support@e-library[.]wiki. The campaign purported to be from the scientific electronic library, eLibrary, whose legitimate website is elibrary.ru. The phishing emails contained a malicious link in the format: https://e-library[.]wiki/elib/wiki.php?id=<8 pseudorandom letters and digits>. Recipients were prompted to click the link to download a plagiarism report. Clicking that link triggered the download of an archive file. The filename was personalized, using the victim’s own name in the format: <LastName>_<FirstName>_<Patronymic>.zip.
A well-prepared attack
The attackers did their homework before sending out the phishing emails. The malicious domain, e-library[.]wiki, was registered back in March 2025, over six months before the email campaign started. This was likely done to build the domain’s reputation, as sending emails from a suspicious, newly registered domain is a major red flag for spam filters.
Furthermore, the attackers placed a copy of the legitimate eLibrary homepage on https://e-library[.]wiki. According to the information on the page, they accessed the legitimate website from the IP address 193.65.18[.]14 back in December 2024.
A screenshot of the malicious site elements showing the IP address and initial session date
The attackers also carefully personalized the phishing emails for their targets, specific professionals in the field. As mentioned above, the downloaded archive was named with the victim’s last name, first name, and patronymic.
Another noteworthy technique was the attacker’s effort to hinder security analysis by restricting repeat downloads. When we attempted to download the archive from the malicious site, we received a message in Russian, indicating the download link was likely for one-time use only:
The message that was displayed when we attempted to download the archive
Our investigation found that the malicious site displayed a different message if the download was attempted from a non-Windows device. In that case, it prompted the user to try again from a Windows computer.
The message that was displayed when we attempted to download the archive from a non-Windows OS
The malicious archive
The malicious archives downloaded via the email links contained the following:
A malicious shortcut file named after the victim: <LastName>_<FirstName>_<Patronymic>.lnk;
A .Thumbs directory containing approximately 100 image files with names in Russian. These images were not used during the infection process and were likely added to make the archives appear less suspicious to security solutions.
A portion of the .Thumbs directory contents
When the user clicked the shortcut, it ran a PowerShell script. The script’s primary purpose was to download and execute a PowerShell-based payload from a malicious server.
The script that was launched by opening the shortcut
The downloaded payload then performed the following actions:
Contacted a URL in the format: https://e-library[.]wiki/elib/query.php?id=<8 pseudorandom letters and digits>&key=<32 hexadecimal characters> to retrieve the final payload, a DLL file.
Saved the downloaded file to %localappdata%\Microsoft\Windows\Explorer\iconcache_<4 pseudorandom digits>.dll.
Established persistence for the payload using COM Hijacking. This involved writing the path to the DLL file into the registry key HKCR\CLSID\{1f486a52-3cb1-48fd-8f50-b8dc300d9f9d}\InProcServer32. Notably, the attackers had used that same technique in their spring attacks.
Downloaded a decoy PDF from a URL in the format: https://e-library[.]wiki/pdf/<8 pseudorandom letters and digits>.pdf. This PDF was saved to the user’s Downloads folder with a filename in the format: <LastName>_<FirstName>_<Patronymic>.pdf and then opened automatically.
The decoy PDF contained no valuable information. It was merely a blurred report generated by a Russian plagiarism-checking system.
A screenshot of a page from the downloaded report
At the time of our investigation, the links for downloading the final payloads didn’t work. Attempting to access them returned error messages in English: “You are already blocked…” or “You have been bad ended” (sic). This likely indicates the use of a protective mechanism to prevent payloads from being downloaded more than once. Despite this, we managed to obtain and analyze the final payload.
The final payload: the Tuoni framework
The DLL file deployed to infected devices proved to be an OLLVM-obfuscated loader, which we described in our previous report on Operation ForumTroll. However, while this loader previously delivered rare implants like LeetAgent and Dante, this time the attackers opted for a better-known commercial red teaming framework: Tuoni. Portions of the Tuoni code are publicly available on GitHub. By deploying this tool, the attackers gained remote access to the victim’s device along with other capabilities for further system compromise.
As in the previous campaign, the attackers used fastly.net as C2 servers.
Conclusion
The cyberattacks carried out by the ForumTroll APT group in the spring and fall of 2025 share significant similarities. In both campaigns, infection began with targeted phishing emails, and persistence for the malicious implants was achieved with the COM Hijacking technique. The same loader was used to deploy the implants both in the spring and the fall.
Despite these similarities, the fall series of attacks cannot be considered as technically sophisticated as the spring campaign. In the spring, the ForumTroll APT group exploited zero-day vulnerabilities to infect systems. By contrast, the autumn attacks relied entirely on social engineering, counting on victims not only clicking the malicious link but also downloading the archive and launching the shortcut file. Furthermore, the malware used in the fall campaign, the Tuoni framework, is less rare.
ForumTroll has been targeting organizations and individuals in Russia and Belarus since at least 2022. Given this lengthy timeline, it is likely this APT group will continue to target entities and individuals of interest within these two countries. We believe that investigating ForumTroll’s potential future campaigns will allow us to shed light on shadowy malicious implants created by commercial developers – much as we did with the discovery of the Dante spyware.
In March 2025, we discovered Operation ForumTroll, a series of sophisticated cyberattacks exploiting the CVE-2025-2783 vulnerability in Google Chrome. We previously detailed the malicious implants used in the operation: the LeetAgent backdoor and the complex spyware Dante, developed by Memento Labs (formerly Hacking Team). However, the attackers behind this operation didn’t stop at their spring campaign and have continued to infect targets within the Russian Federation.
In October 2025, just days before we presented our report detailing the ForumTroll APT group’s attack at the Security Analyst Summit, we detected a new targeted phishing campaign by the same group. However, while the spring cyberattacks focused on organizations, the fall campaign honed in on specific individuals: scholars in the field of political science, international relations, and global economics, working at major Russian universities and research institutions.
The emails received by the victims were sent from the address support@e-library[.]wiki. The campaign purported to be from the scientific electronic library, eLibrary, whose legitimate website is elibrary.ru. The phishing emails contained a malicious link in the format: https://e-library[.]wiki/elib/wiki.php?id=<8 pseudorandom letters and digits>. Recipients were prompted to click the link to download a plagiarism report. Clicking that link triggered the download of an archive file. The filename was personalized, using the victim’s own name in the format: <LastName>_<FirstName>_<Patronymic>.zip.
A well-prepared attack
The attackers did their homework before sending out the phishing emails. The malicious domain, e-library[.]wiki, was registered back in March 2025, over six months before the email campaign started. This was likely done to build the domain’s reputation, as sending emails from a suspicious, newly registered domain is a major red flag for spam filters.
Furthermore, the attackers placed a copy of the legitimate eLibrary homepage on https://e-library[.]wiki. According to the information on the page, they accessed the legitimate website from the IP address 193.65.18[.]14 back in December 2024.
A screenshot of the malicious site elements showing the IP address and initial session date
The attackers also carefully personalized the phishing emails for their targets, specific professionals in the field. As mentioned above, the downloaded archive was named with the victim’s last name, first name, and patronymic.
Another noteworthy technique was the attacker’s effort to hinder security analysis by restricting repeat downloads. When we attempted to download the archive from the malicious site, we received a message in Russian, indicating the download link was likely for one-time use only:
The message that was displayed when we attempted to download the archive
Our investigation found that the malicious site displayed a different message if the download was attempted from a non-Windows device. In that case, it prompted the user to try again from a Windows computer.
The message that was displayed when we attempted to download the archive from a non-Windows OS
The malicious archive
The malicious archives downloaded via the email links contained the following:
A malicious shortcut file named after the victim: <LastName>_<FirstName>_<Patronymic>.lnk;
A .Thumbs directory containing approximately 100 image files with names in Russian. These images were not used during the infection process and were likely added to make the archives appear less suspicious to security solutions.
A portion of the .Thumbs directory contents
When the user clicked the shortcut, it ran a PowerShell script. The script’s primary purpose was to download and execute a PowerShell-based payload from a malicious server.
The script that was launched by opening the shortcut
The downloaded payload then performed the following actions:
Contacted a URL in the format: https://e-library[.]wiki/elib/query.php?id=<8 pseudorandom letters and digits>&key=<32 hexadecimal characters> to retrieve the final payload, a DLL file.
Saved the downloaded file to %localappdata%\Microsoft\Windows\Explorer\iconcache_<4 pseudorandom digits>.dll.
Established persistence for the payload using COM Hijacking. This involved writing the path to the DLL file into the registry key HKCR\CLSID\{1f486a52-3cb1-48fd-8f50-b8dc300d9f9d}\InProcServer32. Notably, the attackers had used that same technique in their spring attacks.
Downloaded a decoy PDF from a URL in the format: https://e-library[.]wiki/pdf/<8 pseudorandom letters and digits>.pdf. This PDF was saved to the user’s Downloads folder with a filename in the format: <LastName>_<FirstName>_<Patronymic>.pdf and then opened automatically.
The decoy PDF contained no valuable information. It was merely a blurred report generated by a Russian plagiarism-checking system.
A screenshot of a page from the downloaded report
At the time of our investigation, the links for downloading the final payloads didn’t work. Attempting to access them returned error messages in English: “You are already blocked…” or “You have been bad ended” (sic). This likely indicates the use of a protective mechanism to prevent payloads from being downloaded more than once. Despite this, we managed to obtain and analyze the final payload.
The final payload: the Tuoni framework
The DLL file deployed to infected devices proved to be an OLLVM-obfuscated loader, which we described in our previous report on Operation ForumTroll. However, while this loader previously delivered rare implants like LeetAgent and Dante, this time the attackers opted for a better-known commercial red teaming framework: Tuoni. Portions of the Tuoni code are publicly available on GitHub. By deploying this tool, the attackers gained remote access to the victim’s device along with other capabilities for further system compromise.
As in the previous campaign, the attackers used fastly.net as C2 servers.
Conclusion
The cyberattacks carried out by the ForumTroll APT group in the spring and fall of 2025 share significant similarities. In both campaigns, infection began with targeted phishing emails, and persistence for the malicious implants was achieved with the COM Hijacking technique. The same loader was used to deploy the implants both in the spring and the fall.
Despite these similarities, the fall series of attacks cannot be considered as technically sophisticated as the spring campaign. In the spring, the ForumTroll APT group exploited zero-day vulnerabilities to infect systems. By contrast, the autumn attacks relied entirely on social engineering, counting on victims not only clicking the malicious link but also downloading the archive and launching the shortcut file. Furthermore, the malware used in the fall campaign, the Tuoni framework, is less rare.
ForumTroll has been targeting organizations and individuals in Russia and Belarus since at least 2022. Given this lengthy timeline, it is likely this APT group will continue to target entities and individuals of interest within these two countries. We believe that investigating ForumTroll’s potential future campaigns will allow us to shed light on shadowy malicious implants created by commercial developers – much as we did with the discovery of the Dante spyware.
Every year, the cloud is becoming more distributed, automated and tightly wired into the business. Every day, adversaries compress the timeline between compromise and data exfiltration. What once took them 44 days now takes minutes. For the fifth year in a row, Palo Alto Networks State of Cloud Security Report 2025 captures the changes both big and small that security leaders are navigating in the market today. Our report reveals that the rapid adoption of enterprise AI is fueling an unprecedented surge in cloud security risks, driving a massive expansion of the attack surface. We found that 99% of organizations experienced at least one attack on their AI systems within the past year, and the acceleration of GenAI-assisted coding is outstripping security teams' capacity to keep pace. What’s missing isn't just visibility, it’s alignment.
Our research, drawing on insights from more than 2,800 security leaders, surfaces the critical cost of misalignment across teams, tools and workflows. This report provides key benchmarks to help inform the decisions that shape your cloud strategy as we track where teams gain ground, where they struggle, and how the threat landscape, now accelerated by AI, is evolving.
The Cloud Attack Surface Is Expanding with AI
The biggest shift in the cloud landscape is the acceleration of risk driven by AI adoption. As cloud infrastructure expands to host the growing number of AI workloads, it has become a critical target. The introduction of GenAI into development pipelines is also compounding the problem by increasing the volume of insecure code going into production.
Of those surveyed in the 2025 report, 75% of organizations stated that they are running AI in their production environments today. That level is significant, as it points to the growing adoption and use of AI as businesses are locked in what looks like a modern arms race to bring the latest capabilities and benefits to their organizations and customers. In addition, as stated earlier, our findings confirm that 99% of organizations reported at least one attack on their AI systems within the past year. This number proves that AI needs human guardrails, as well as to be secured to contain the risk of critical data exposure by adversaries.
The prevalence of AI use and attacks on AI.
The AppSec Pipeline Is Not Secure Enough Yet
As AI expands the cloud attack surface and has been proven to be a significant target, we can see that code development pipelines are also being stressed by the same forces. An important trend from the 2025 report is the rise of GenAI-assisted coding (vibe coding), used by 99% of respondents. The use of vibe coding is generating insecure code faster than security teams can review it. The acceleration creates a massive risk gap: 52% of teams are shipping code weekly, but only 18% are able to fix vulnerabilities at that same pace. This confirms that traditional, human-led approaches to application security are inadequate, leaving security teams to fight threats with fragmented tools and slow, manual fix cycles.
The Speed of development across survey respondents.
As the pace of development increases, the disconnect between security assessment and remediation is becoming more apparent too. While teams are making progress by shifting away from outdated vulnerability prioritization methods, they still struggle to integrate security effectively into the development workflow. This introduces a large number of vulnerabilities into production, where 20% of organizations report that an average of 37% of their high or critical issues reach their production environments. Once in production those vulnerabilities linger, as 82% of organizations report it taking longer than a week to deploy code fixes. What is slowing teams down?
The traditional refrain toward implementing prevention that blocks risks from reaching production during rapid code development is still true today. The barriers are clear: 31% cite poor CI/CD integration and another 31% worry about slowing down development. On the positive note, only 17% rely on CVSS scores to prioritize their fixes as teams are now moving more toward context-rich decisions based on exploitability-based triage (32%) and business impact (33%).
The New Frontiers of Cloud Risk
Attackers are rapidly pivoting to exploit the foundational layers of the cloud, with a clear focus on ungoverned interfaces and overprivileged access. The volume and autonomy introduced by AI agents further accelerates this exploitation, turning minor gaps into major incidents.
Attacks on APIs Jump for 41%
APIs are the new primary entry point. Attacks on APIs increased for 41% of organizations in the last year, marking the sharpest rise of any threat category measured. As agentic AI relies heavily on APIs to operate, this explosion in usage has greatly expanded the attack surface. Furthermore, nearly every AI-related threat, including model supply chain tampering, token theft and prompt injection, involves an API boundary. This reinforces the role of ungoverned interfaces in scalable AI compromise, with 47% of AI system breaches involving data exfiltration through assistants or plugins.
Identity Still Remains the Weakest Link
Insufficient access controls remain a leading vector for credential theft and data exfiltration. 53% of organizations cite lenient identity and access management (IAM) practices as a top data security challenge. This problem is compounded by complexity. The number climbs to 57% among organizations running more than six AppSec tools, proving that the discipline required to maintain least privilege is failing to scale with tool sprawl. Data leaves through both legitimate business systems and breach events, making it fundamentally an identity problem.
The top three exfiltration vectors confirm this focus:
SaaS sync or export misuse: 63%
Overpermissive external sharing: 59%
Compromised credentials or tokens: 58%
Lateral Movement Risks Persist
Once an attacker gains a foothold, they can move freely. Twenty-eight percent point to unrestricted network access between cloud workloads as a growing threat, allowing attackers to pivot across environments and turn minor compromises into major incidents.
The Growing Imperative of Cloud & SOC Must Merge
The gap between detection and resolution is where breaches succeed. Today the cloud and SOC divide is proving too slow in the face of machine-speed threats. Structural fragmentation is clearly visible in response times, while 74% of organizations detect threats within 24 hours, 30% take more than a full day to resolve them. A delay like this is caused by disjointed workflows and isolated data sources between cloud and SOC teams, which stall incident response (IR) for 50% of organizations.
How SOC analysts spend their time after an incident.Respondents calling for cloud and security operations to merge.
The demand for consolidation shows up across the board:
89% of organizations believe cloud security and security operations must fully merge, not just integrate.
Organizations currently manage an average of 17 tools from five vendors, creating fragmented data and context gaps.
Consequently, 97% of respondents prioritized consolidating their security footprint to address the chaos of tool sprawl.
The model that worked for lift-and-shift can't contain threats that move at machine speed. Organizations are ready to collapse the distance between teams and tools.
About the Report
The State of Cloud Security Report 2025 draws from over 2,800 security leaders and practitioners across 10 countries and includes breakouts by region, industry and cloud maturity, along with the full incident data and strategic insights we’ve touched on here.
2,8000 survey respondents by country.
Learn More and Transform to an Agentic-First Platform
To stay ahead of adversaries who use AI to launch attacks at machine speed, human-led defense is no longer sufficient. The report emphasizes that organizations must counter with an equivalent evolution: Agentic security, leveraging autonomous agents to deliver cloud security from code to cloud to SOC.
For the latest discoveries in cyber research for the week of 15th December, please download our Threat Intelligence Bulletin.
TOP ATTACKS AND BREACHES
The Indian government confirmed cyber incidents involving GPS spoofing at seven major airports, including Delhi, Mumbai, Kolkata, and Bengaluru. The attack affected aircrafts using GPS-based landing procedures. Despite signal disruption to navigation data, authorities stated no flights were cancelled or diverted, with contingency measures and Air Traffic Control safeguards preventing operational impact.
US-based healthcare technology provider, TriZetto Provider Solutions, has notified healthcare clients of a long-running unauthorized access to a customer web portal. With this access a threat actor accessed historical eligibility transaction reports containing protected health information (PHI). Exposed data includes patient and insured PII.
700Credit, a US-based credit check and identity verification provider, suffered a data breach affecting at least 5.6 million people. The incident exposed private information after an unidentified attacker accessed dealer-collected data between May and October 2025. The company is notifying impacted individuals and offering credit monitoring, while Michigan’s attorney general urged affected users to enable credit freezes or monitoring to mitigate fraud risk.
Pierce County Library System in Washington has disclosed a cyberattack impacting over 340,000 individuals after threat actors accessed its systems, forcing a full shutdown. The breach exposed user data and extensive employee PII. The attack was claimed by the INC ransomware gang, which has targeted multiple US government entities in 2025.
Check Point Threat Emulation provides protection against this threat (Ransomware.Wins.INC)
The French Interior Ministry confirmed a cyberattack targeting its email servers, allowing an attacker to access a number of internal files. Authorities stated there is no evidence of serious data compromise at this stage. An investigation is ongoing, with no attribution yet identified.
Russian Government IT contractor Mikord was reportedly breached by an anonymous hacker group. The group claims to have maintained access for months, exfiltrated source code, internal communications, financial and technical records, and damaged infrastructure tied to a firm allegedly involved in Russia’s unified military draft database. While Mikord’s director confirmed a hack, Russia’s Ministry of Defense denied any breach or data leak.
An employee of Home Depot, the US home improvement retailer, had mistakenly exposed a private GitHub token, granting access to internal systems for nearly a year. The token enabled entry to hundreds of private code repositories and key cloud systems and was revoked upon discovery.
VULNERABILITIES AND PATCHES
Google released an urgent Chrome update on to address a high severity flaw (CVE-2025-14174) actively exploited in the wild and linked to the ANGLE graphics library used for WebGL. The bug likely enables memory corruption that could allow remote code execution.
Apple released emergency security updates to patch two actively exploited zero-day vulnerabilities, CVE-2025-43529 and CVE-2025-14174. The vulnerabilities were exploited in sophisticated targeted attacks against specific individuals. Both flaws affect WebKit and enable remote code execution or memory corruption via malicious web content, impacting iPhones, iPads, Macs, and other Apple platforms.
SAP released details and patches for three vulnerabilities, including CVE-2025-42880 (code injection in Solution Manager, CVSS 9.9), CVE-2025-55754 (Commerce Cloud Tomcat flaws, CVSS 9.6), and CVE-2025-42928 (jConnect deserialization, CVSS 9.1), alongside several high severity issues.
THREAT INTELLIGENCE REPORTS
Check Point Research reports a global rise in cyber attacks in November 2025, averaging 2,003 weekly attempts per organization, with education most targeted sector and rising exposure from generative AI. 727 ransomware incidents were recorded, a 22% increase YoY, with North America accounting for 55% of cases and industrial manufacturing being the top victim industry.
Check Point Research exposed ValleyRAT’s modular system, including a kernel-mode rootkit that can remain loadable on fully updated Windows 11 despite built-in protections. The research linked leaked builder artifacts to plugins and identified about 6,000 samples, with roughly 85 percent emerging in the last six months after the builder’s public release.
Check Point researchers revealed a phishing campaign where attackers impersonate file-sharing and e-signature services to deliver finance-themed lures that look like legitimate notifications. The attackers sent over 40,000 phishing emails targeting roughly 6,100 customers over the past two weeks, abusing Mimecast’s secure-link rewriting feature as a smokescreen to make their links appear safe and authenticated
Researchers have analyzed STAC6565 campaign, which with high confidence is associated with the GOLD BLADE threat group (aka RedCurl, RedWolf, and Earth Kapre). The campaign is mostly targeting Canadian organizations, blending data theft with selective QWCrypt ransomware. The threat actor uses multi-stage infection chains that include payloads downloaded via WebDAV, DLL side-loading using legitimate Adobe components, and BYOVD abuse to evade detection.
Researchers uncovered a new phishing technique called ConsentFix that tricks people into giving attackers access to their Microsoft accounts. The method uses a browser-native prompt that persuades victims to copy and paste a link. Once the link is submitted, attackers can get access without needing a password or multi-factor authentication.
At Amazon Web Services, we’re committed to deeply understanding the evolving needs of both our customers and regulators, and rapidly adapting and innovating to meet them. The upcoming AWS European Sovereign Cloud will be a new independent cloud for Europe, designed to give public sector organizations and customers in highly regulated industries further choice to meet their unique sovereignty requirements. The AWS European Sovereign Cloud expands on the same strong foundation of security, privacy, and compliance controls that apply to other AWS Regions around the globe with additional governance, technical, and operational measures to address stringent European customer and regulatory expectations. Sovereignty is the defining feature of the AWS European Sovereign Cloud and we’re using an independently validated framework to meet our customers’ requirements for sovereignty, while delivering the scalability and functionality you expect from the AWS Cloud.
Today, we’re pleased to share further details about the AWS European Sovereign Cloud: Sovereign Reference Framework (ESC-SRF). This reference framework aligns sovereignty criteria across multiple domains such as governance independence, operational control, data residency and technical isolation. Working backwards from our customers’ sovereign use cases, we aligned controls to each of the criteria and the AWS European Sovereign Cloud is undergoing an independent third-party audit to verify the design and operations of these controls conform to AWS sovereignty commitments. Customers and partners can also leverage the ESC-SRF as a foundation upon which they can build their own complementary sovereignty criteria and controls when using the AWS European Sovereign Cloud.
To clearly explain how the AWS European Sovereign Cloud meets sovereignty expectations, we’re publishing the ESC-SRF in AWS Artifact including the criteria and control mapping. In AWS Artifact, our self-service audit artifact retrieval portal, you have on-demand access to AWS security and compliance documents and AWS agreements. You can now use the ESC-SRF to define best practices for your own use case, map these to controls, and illustrate how you meet and even exceed sovereign needs of your customers.
A transparent and validated sovereignty model
The ESC-SRF has been built from customer feedback, regulatory requirements across the European Union (EU), industry frameworks, AWS contractual commitments, and partner input. ESC-SRF is industry and sector agnostic, as it’s written to address fundamental sovereignty needs and expectations at the foundational layer of our cloud offerings with additional sovereignty-specific requirements and controls that apply exclusively to the AWS European Sovereign Cloud. Each criterion is implemented through sovereign controls that will be independently validated by a third-party auditor.
The framework builds on core AWS security capabilities, including encryption, key management, access governance, AWS Nitro System-based isolation, and internationally recognized compliance certifications. The framework adds sovereign-specific governance, technical, and operational measures such as independent EU corporate structures, dedicated EU trust and certificate services, operations by AWS EU-resident personnel, strict residency for customer data and customer created metadata, separation from all other AWS Regions, and incident response operated within the EU.
These controls are the basis of a dedicated AWS European Sovereign Cloud System and Organization Controls (SOC) 2 attestation. The ESC-SRF establishes a solid foundation for sovereignty of the cloud, so that customers can focus on defining sovereignty measures in the cloud that are tailored to their goals, regulatory needs, and risk posture.
How you can use the ESC-SRF
The ESC-SRF describes how AWS implements and validates sovereignty controls in the AWS European Sovereign Cloud. AWS treats each criterion as binding and its implementation will be validated by an independent third-party auditor in 2026. While most customers don’t operate at the size and scale of AWS, you can use the ESC-SRF as both an assurance model and a reference framework you can adapt to your specific use cases.
From an assurance perspective, it provides end-to-end visibility for each sovereignty criterion through to its technical implementation. We will also provide third-party validation in the AWS European Sovereign Cloud SOC 2 report. Customers can use this report with internal auditors, external assessors, supervisory authorities, and regulators. This can reduce the need for ad-hoc evidence requests and supports customers by providing them with evidence to demonstrate clear and enforceable sovereignty assurances.
From a design perspective, you can refer to the framework when shaping your own sovereignty architecture, selecting configurations, and defining internal controls to meet regulatory, contractual, and mission-specific requirements. Because the ESC-SRF is industry and sector agnostic, you can apply criteria from the framework to suit your own unique needs. Depending on your sovereign use case, not all criteria may apply to your use case sovereign needs. The ESC-SRF can also be used in conjunction with AWS Well-Architected which can help you learn, measure, and build using architectural best practices. Where appropriate you can create your version of the ESC-SRF, map to controls, and have them tested by a third party. To download the ESC-SRF, visit AWS Artifact (login required).
A strong, clear foundation
The publication of the ESC-SRF is part of our ongoing commitment to delivering on the AWS Digital Sovereignty Pledge through transparency and assurances to help customers meet their evolving sovereignty needs with assurances designed, implemented, and validated entirely within the EU. Within the framework, customers can build solutions in the AWS European Sovereign Cloud with confidence and a strong understanding of how they are able to meet their sovereignty goals using AWS.
For more information about the AWS European Sovereign Cloud, visit aws.eu.
If you have feedback about this post, submit comments in the Comments section below.
We are proud to announce that Frost & Sullivan has recognized Palo Alto Networks Prisma® Browser as the best-positioned market leader in the Frost Radar: Zero Trust Browser Security (ZTBS), 2025 report, securing the premier position for innovation and a leadership position on growth.
This recognition comes at a pivotal moment. For the modern enterprise, the browser is no longer just an application; it is your new OS. With 85% of the work happening in browsers, it has become the focal point where revenue is generated and sensitive data is accessed. However, this shift has transformed your primary workspace into the primary attack vector, with 95% of organizations having reported a security incident originating in the browser, placing it on the frontline against sophisticated AI® threats and critical vulnerabilities. The risk of evasive, AI-driven phishing attempts is compounded by the widespread use of managed and unmanaged devices, creating blind spots that allow sensitive data to be exfiltrated faster than ever.
To combat this, enterprises need a browser that doesn't just display the web but actively defends it with its users, apps, data and devices. This is a necessity that drives our latest industry recognition.
Proven Leadership Validated by the Market
Prisma Browser’s recognition as the best-positioned leader, securing the premier position for innovation and a leadership position on growth, is a testament to our commitment to deliver best-in-class security that is both easy to deploy and that IT and users love to use. By integrating Palo Alto Networks Precision AI® technology, Cloud-Delivered Security Services (CDSS) and Enterprise DLP, we ensure our customers benefit from the power of our security engines. And because they are natively integrated in the browser, we are mitigating threats hiding in encrypted traffic, blind spot web channels, AI-powered spear phishing and other evasive web threats that legacy security tools simply cannot identify.
Prisma Browser’s Innovation Advantage
Our leadership is driven by continuous strategic innovation in the secure browser space. Prisma Browser delivers critical "last-mile" protection through the native integration of CDSS, including Advanced WildFire® for zero-day malware analysis and Advanced URL Filtering instantly at the point of user interaction. Building on this foundation, our latest innovations extend secure work to all applications, including those beyond SSO, providing full visibility and last-mile protection for unmanaged applications, such as GenAI apps, closing gaps left by incomplete identity coverage. We further solidify this best-in-class security through additional cutting-edge innovations: Advanced Web Protection for real-time evasive threat protection, Advanced Browser Protection for zero-day browser exploitation defense, and Advanced Extension Security for runtime extension security.
At the core of this defense is Precision AI, our proprietary engine that combines machine learning, deep learning and generative AI to automate detection, prevention and remediation with industry-leading accuracy. Unlike standard security tools that rely on static signatures, Prisma Browser, powered by Precision AI, inspects live, fully rendered content. It detects evasive phishing attempts (such as AI-generated cloaking) and malicious reassembly attacks that legacy tools miss, effectively fighting AI with AI. Fueled by intelligence from over 70 thousand customers, Prisma Browser delivers unmatched threat detection, identifying and blocking up to 8.95 million new and unique attacks every single day.
The Frost Report says this about Palo Alto Networks Innovation:
Key differentiating capabilities include last-mile data leakage protection with browser-level visibility; AI-powered web attack detection and prevention with full page runtime visibility; detection and disabling of malicious extensions using behavioral monitoring; an advanced AI-powered DLP engine; in-browser anti-exploit protection; and a rich library of AI applications and agents.
Crucially, Enterprise DLP capabilities are embedded directly into the rendering engine, granting granular control over sensitive data that traditional network-level tools effectively miss. This helps ensure that data on both managed and unmanaged devices remains secure against exfiltration via clipboard restrictions, screenshot blocking, real-time redaction and more, without disrupting the user experience.
Prisma Browser’s Growth Advantage
Central to the widespread adoption of Prisma Browser is our proven ability to secure the managed workforce at scale without disrupting daily workflows. One of our key differentiators is our 100% license portability, which allows organizations to deploy Prisma Browser across their entire fleet of devices, whether as full browsers, extensions, mobile solutions and firewall connectors with complete flexibility. This frictionless deployment model enables IT teams to instantly layer enterprise-grade security and unified policies onto the same native browser UX employees already know and use.
For CISOs and CIOs focused on streamlining operations, Prisma Browser is also offered as a fully integrated solution within the Prisma® SASE platform, enabling unified policies across all Palo Alto Networks solutions.
Looking Ahead
While we are proud of our position on the Frost Radar: Zero Trust Browser Security (ZTBS) report, we are just getting started. By accelerating initiatives in GenAI security, complete web protection, modern data protection and VDI reduction, we are redefining the browser. We don't just want the browser to be where you work; we are transforming it from the primary attack vector into one of the organization's most robust lines of defense and the single point where they can identify AI driven attacks and fight AI with AI.
For the latest discoveries in cyber research for the week of 8th December, please download our Threat Intelligence Bulletin.
TOP ATTACKS AND BREACHES
The University of Pennsylvania and the University of Phoenix were hit by data breaches after attackers exploited zero-day vulnerabilities in Oracle E-Business Suite servers. At least 1,488 people at UPenn and numerous students, alumni, donors, staff, faculty, employees, and suppliers at Phoenix were impacted. The Cl0p ransomware gang is likely responsible, as part of a broader campaign.
Check Point IPS, Threat Emulation and Harmony Endpoint provide protection against this threat (Oracle Multiple Products Remote Code Execution; Ransomware.Win.Clop; Ransomware.Wins.Clop; Ransomware.Wins.Clop.ta.*)
Financial software provider Marquis Software Solutions has disclosed a data breach that impacted over 74 banks and credit unions across the US and exposed sensitive data of more than 400,000 customers. The Akira ransomware gang is possibly responsible for the attack, which exploited vulnerabilities in SonicWall firewalls to gain network access.
Check Point Threat Emulation provides protection against this threat (Ransomware.Wins.Akira.ta.*; Ransomware.Wins.Akira)
American pharmaceutical firm Inotiv has reported on a ransomware attack that occurred in August 2025. The Qilin ransomware group claimed responsibility, leaking personal information from over 9,500 individuals, including current and former employees and their family members.
Check Point Threat Emulation and Harmony Endpoint provide protection against this threat
South Korean retail giant Coupang has confirmed a data breach that resulted in the exposure of personal information belonging to nearly 34 million clients, including full names, phone numbers, email addresses, and more. No payment details or account passwords were leaked in the incident.
YouTube app for Android TV, SmartTube, has been targeted in an attack that resulted in the compromise of its developer signing keys and the distribution of a malicious update containing hidden malware. The incident impacted Android TV, Fire TV Stick, and similar device users.
Belgian postal and package delivery service, Bpost, has suffered a data breach that resulted in the exfiltration of 5,140 files totaling about 30.46GB from a third-party exchange platform. The stolen data reportedly includes personal and business information of some customers of the affected department. The ransomware group TridentLocker has claimed responsibility for the attack.
Canadian wireless telecommunications provider, Freedom Mobile, has experienced a data breach that resulted in attackers gaining unauthorized access to its customer account management platform and stealing personal information, including names, addresses, dates of birth, phone numbers, and account numbers. The company has not disclosed the exact number of affected customers.
VULNERABILITIES AND PATCHES
Check Point has elaborated on the critical React2Shell vulnerability, CVE-2025-55182, that affects React 19.x and related server-side frameworks such as Next.js 15.x/16.x. The vulnerability enables unauthenticated remote code execution via malicious HTTP requests targeting the server’s decoding process. Exploitation allows attackers to gain full control over application servers, intercept sensitive data, inject false transactions, and potentially pivot deeper into enterprise environments.
Check Point IPS provides protection against this threat (React Server Components Remote Code Execution (CVE-2025-55182))
Check Point Research revealed a vulnerability in OpenAI Codex CLI that allowed attackers to achieve remote code execution via malicious project-local configuration files (MCP entries) executed without user prompts. OpenAI released a patch in version 0.23.0 to address the automatic execution risk.
Check Point Research shared details of a critical exploit in Yearn Finance’s yETH pool, where an attacker abused a smart contract flaw to mint trillions of tokens with a minuscule deposit, resulting in the theft of approximately $9 million in assets from the Ethereum-based DeFi protocol.
THREAT INTELLIGENCE REPORTS
Check Point summarizes a multiyear Salt Typhoon cyber-espionage campaign that compromised 80 telecom providers worldwide and a US state Army National Guard network, chaining SIM-based credential theft, network scans, Ivanti/PAN-OS/Cisco CVEs and GTP/GTPDOOR abuses to exfiltrate sensitive communications and configuration data.
US and Canadian cybersecurity agencies outlined BRICKSTORM, a stealthy backdoor used by Chinese affiliated hackers to infiltrate VMware vSphere environments and maintain long-term access. The campaign targeted government services and IT, stealing credentials via VM snapshots and creating hidden machines.
The ShadyPanda threat actor ran a seven-year campaign weaponizing verified Chrome and Edge extensions to infect over 4.3 million devices with spyware for remote code execution, payload delivery, traffic redirection, credential and cookie theft, browser fingerprinting, HTTPS credential interception, and behavioral biometrics exfiltration.
Researchers identified a campaign weaponizing Velociraptor, a digital forensics tool, to establish stealthy command channels and maintain persistence in enterprise environments. Attackers exploited SharePoint’s “ToolShell” chain using CVE-2025-49706 and CVE-2025-49704, linked to Storm-2603, and in confirmed cases delivered Warlock ransomware.
Albiriox, a new Android banking trojan sold as Malware-as-a-Service (MaaS), targets over 400 financial and crypto apps using VNC-style remote control, accessibility abuse, overlays, and black-screen masking for on-device fraud. The malware is spread via smishing, WhatsApp lures, and fake apps with droppers over unencrypted TCP C2 channels using structured JSON messages.
For the latest discoveries in cyber research for the week of 1st December, please download our Threat Intelligence Bulletin.
TOP ATTACKS AND BREACHES
OpenAI has experienced a data breach resulting from a compromise at third-party analytics provider Mixpanel, which exposed limited information of some ChatGPT API clients. The leaked data includes names, email addresses, approximate location, operating system, browser information, referring websites, and organization or user IDs. No sensitive credentials or API keys were exposed.
Dartmouth College, a private Ivy League research university in New Hampshire, has been a victim of a data breach that resulted in the theft of personal information, including names, Social Security numbers and financial details, from its Oracle E-Business Suite servers. The Cl0p extortion gang was responsible for exploiting zero-day vulnerability as part of a broader campaign. Other targets include Harvard University, Envoy Air, and others with sensitive data exposed via dark web and torrent sites.
Check Point IPS, Threat Emulation and Harmony Endpoint provide protection against this threat (Oracle Concurrent Processing Remote Code Execution; Ransomware.Win.Clop; Ransomware.Wins.Clop; Ransomware.Wins.Clop.ta.*)
Crisis24, a leader in crisis and risk management, was hit by a cyberattack on its OnSolve CodeRED emergency alert platform that resulted in widespread disruption of notification systems nationwide and the theft of user data. Leaked information including names, addresses, email addresses, phone numbers, and clear-text passwords affecting state and local governments, public safety agencies, and residents across the US. The INC Ransomware gang has claimed responsibility for the attack and is offering stolen data for sale.
Check Point Threat Emulation provides protection against this threat (Ransomware.Wins.INC)
Major American investment advisory provider SitusAMC has confirmed a data breach that resulted in the compromise of corporate data associated with client relationships, including accounting records, legal agreements, and potentially customer data. The breach impacted an undisclosed number of clients and customers, likely including largest banks and financial institutions in the US, with no information yet provided on the amount or exact type of data leaked.
A Russian postal operator Donbas Post has encountered a cyber-attack that disrupted its corporate network, web platform, and email systems, destroying over 1,000 workstations, 100 virtual machines, and several dozen terabytes of data, and forcing the suspension of services at postal branches and the call center. The Ukrainian Cyber Alliance has claimed responsibility.
The French Football Federation (FFF) has suffered a data breach that resulted in unauthorized access to administrative management software and theft of personal and contact information from members of French football clubs. Exposed data includes names, email addresses, and more.
VULNERABILITIES AND PATCHES
A new Mirai-based botnet, ShadowV2, was observed exploiting multiple known vulnerabilities (including CVE-2024-10914, CVE-2024-10915, and CVE-2024-53375) in IoT devices to gain control and launch distributed denial-of-service (DDoS) attacks. The botnet leveraged command injection and other flaws in routers, NAS devices, and DVRs across global sectors.
Check Point IPS provides protection against this threat (D-Link DNS NAS Devices Command Injection (CVE-2024-10914); D-Link DNS Series Command Injection; TP-Link Archer AXE75 Command Injection (CVE-2024-53375))
Security researcher uncovered more than 17,000 exposed credentials during a scan of 5.6 million public GitLab repositories, including API keys, passwords, and access tokens associated with over 2,800 domains. Many of these credentials – primarily Google Cloud, MongoDB, Telegram, and OpenAI keys – remain active. While most were leaked after 2018, some valid keys date back to 2009.
A patch was released for a critical authentication bypass vulnerability (CVE-2025-59366) in ASUS routers with AiCloud enabled, which allows remote attackers to exploit chained path traversal and OS command injection flaws for unauthorized function execution. Successful exploitation does not require user interaction and could result in attackers gaining control over vulnerable devices.
THREAT INTELLIGENCE REPORTS
Check Point researchers analyzed the Shai-Hulud 2.0 npm supply chain campaign that compromised over 600 npm packages and 25,000 GitHub repositories. Malicious preinstall scripts stole developer and multi-cloud credentials, exfiltrated them to attacker GitHub repos, registered infected hosts as self-hosted runners, and used the stolen tokens for worm-like propagation across npm and GitHub.
Check Point Threat Emulation provides protection against this threat (Trojan.Wins.ShaiHulud.ta.*)
Check Point researchers uncovered GhostAd, a large-scale Android adware campaign where at least 15 Google Play applications with millions of installs abuse foreground services, blank notifications, JobScheduler, and ad SDKs to run persistent background ads and drain device resources. These applications also use background execution and storage permissions to persist, hide, and silently exfiltrate external-storage files, including corporate documents, to attacker infrastructure.
Check Point overviews expected cyber risks at 2026, including converging agentic AI, quantum computing, and Web 4.0. The blog outlines 12 trends: autonomous AI operations, digital-twin/XR environments, LLM-native attacks, deepfake fraud, quantum “harvest-now, decrypt-later” exposure, data-pressure ransomware, expanding supply-chain, SaaS, and identity threats.
Researchers detailed HashJack, an indirect prompt injection technique that embeds malicious instructions in elements like URL fragments or emails to manipulate AI browser assistants – including Comet, Copilot for Edge, and Gemini for Chrome. This method enables threat actors to trigger phishing, misinformation, data exfiltration, and credential theft, exploiting LLMs’ inability to distinguish instructions from legitimate data.
For the latest discoveries in cyber research for the week of 24th November, please download our Threat Intelligence Bulletin.
TOP ATTACKS AND BREACHES
The notorious “Scattered LAPSUS$ Hunters” group claimed responsibility for a supply-chain attack involving the Salesforce-integrated platform Gainsight. The group stated that data from 300 organizations was compromised, including Verizon, GitLab and Atlassian. Salesforce has confirmed unusual activity related to Gainsight integrations and has revoked all active access tokens as a precaution, emphasizing there is no vulnerability in the Salesforce’s core platform.
Eurofiber France SAS, the French unit of Dutch telecommunications provider Eurofiber Group N.V., has been a victim of a data breach. The attack resulted in an unauthorized access to its French ticket management system and exfiltration of customer information from its cloud division and regional sub-brands. A threat actor “ByteToBreach” claimed responsibility for the attack.
Italian IT provider Almaviva has confirmed a cyberattack, with stolen data including information from Ferrovie dello Stato Italiane, Italy’s national railway operator. Nearly 2.3 TB of sensitive files were leaked, including passenger passport data, employee records across FS subsidiaries, defense-related contracts, and financial documents. Almaviva says critical services remain operational.
South Korean giant battery maker LG Energy Solution has experienced a ransomware attack at a single overseas facility, which the company says has been restored, with headquarters unaffected. The Akira gang claimed to have stolen 1.7 terabytes of data.
Check Point Threat Emulation and Harmony Endpoint provide protection against this threat (Ransomware.Wins.Akira.ta.*; Ransomware.Wins.Akira; Trojan.Win.Akira)
Microsoft’s Azure cloud was hit by a massive 15.72 Tbps distributed denial-of-service (DDoS) attack (3.64 billion packets per second) against a public IP address in Australia, sourced from over 500,000 IPs. The high-rate UDP flood is attributed to the Aisuru Turbo Mirai-class IoT botnet, which abuses compromised home routers, cameras, and other internet-connected devices.
Check Point Threat Emulation and Harmony Endpoint provide protection against this threat (Trojan.Wins.Mirai)
French social security service provider, Pajemploi, has suffered a data breach that resulted in the theft of personal data linked to up to 1.2 million of private employers using its childcare services. Exposed information reportedly includes full names, places of birth, postal addresses, Social Security numbers, Pajemploi and accreditation numbers, and banking institution names.
AIPAC, a US political advocacy organization, has encountered a data breach tied to an external third-party system, with notification filed to the Maine attorney general on November 14th. Unauthorized access occurred between October 2024 and February 2025, impacting 810 individuals and exposing personal identifiers. No threat actor claimed responsibility.
VULNERABILITIES AND PATCHES
Fortinet warned of CVE-2025-58034, a FortiWeb command injection flaw actively exploited in the wild. The bug lets authenticated attackers run unauthorized code via crafted requests, with updates available for multiple 7.x and 8.x releases.
Check Point IPS provides protection against this threat (Fortinet FortiWeb Command Injection (CVE-2025-58034))
Google fixed CVE-2025-13223, a high-severity type confusion flaw in Chrome’s V8 engine. The bug is being actively exploited to run malicious code via crafted web pages. Google has issued fixes in Chrome 142.0.7444.175 and later.
Researchers warns of active exploitation and a public proof of concept of CVE-2025-11001, a 7-Zip Windows vulnerability that lets attackers run code by abusing ZIP symbolic link handling. The flaw carries a CVSS 7.0 score and was fixed in 7-Zip version 25.00.
THREAT INTELLIGENCE REPORTS
Check Point Research uncovered a surge in fraudulent Black Friday domains and brand impersonation. Roughly 1 in 11 new Black Friday domains are malicious, and 1 in 25 domains referencing Amazon, AliExpress, or Alibaba pose active threats, with fake storefronts stealing credentials and payment data. Recent examples also mimic HOKA and AliExpress.
Check Point researchers detailed a Europe-wide scam in which criminal networks use generative AI to impersonate health regulators and sell fake GLP-1 weight-loss products. The criminals clone logos and endorsements from the official health services, then localize persuasive ads to exploit drug shortages and public trust.
Akamai discovered a RAT that disguises its C2 traffic as LLM chat completions API requests, sending Base64- and XOR-encoded payloads without standard headers. The malware steals data from remote access tools and browsers and deploys a .NET proxy toolkit with persistence.
Researchers analyzed a Howling Scorpius campaign that used fake CAPTCHA prompts to install SectopRAT on a global data storage and infrastructure company, enabling remote control and lateral movement. Over 42 days, the attackers stole nearly 1 TB of data, deleted cloud backups, and deployed Akira ransomware across three networks, halting operations.
Google analyzed a nearly three-year APT24 cyber-espionage campaign centered on the BadAudio C++ downloader, which uses AES-encrypted C2 traffic, cookie-embedded host profiling, and control-flow flattening to deploy payloads such as Cobalt Strike Beacon in memory. The research details how APT24 shifted from strategic web compromises to large-scale supply-chain and spear-phishing operations that weaponize FingerprintJS-based browser fingerprinting, DLL search-order hijacking, and repeatedly re-compromised Taiwanese marketing infrastructure to deliver BADAUDIO across more than 1,000 domains.
For the latest discoveries in cyber research for the week of 17th November, please download our Threat Intelligence Bulletin.
TOP ATTACKS AND BREACHES
Cl0p’s Oracle E-Business Suite (CVE-2025-61882) zero-day campaign continues to expand. There are new confirmed breaches at The Washington Post, Logitech, Allianz UK, and GlobalLogic, as well as a newly listed but unconfirmed breach involving the British National Health Service (NHS). The group has leaked data sets ranging from gigabytes to terabytes and is sending extortion emails to Oracle EBS customers. Oracle has issued emergency patches, but investigations indicate exploitation began months before disclosure.
Check Point IPS provides protection against this threat (Oracle Concurrent Processing Remote Code Execution)
Payment processor Checkout.com has discloseda data breach by the ShinyHunters threat group. Attackers accessed documents from a legacy cloud storage system that wasn’t properly decommissioned, potentially affecting about 25% of current merchants. That being said, no payment card numbers or funds were compromised. The company is notifying impacted parties and regulators.
DoorDash, a food delivery company, has confirmeda data breach after an employee fell victim to a social engineering scam. Contact details including names, physical addresses, email addresses, and phone numbers were accessed across the US, Canada, Australia, and New Zealand.
Ransomware group dubbed “J Group” claims to have breached Australian engineering firm IKAD. The group has reportedly exfiltrated 800GB of data by exploiting a VPN flaw and maintaining undetected access for five months. IKAD confirmed a cyber incident and the theft of non-sensitive contract and HR information, while denying exposure of classified defence data.
Pro-Russian group NoName057(16) launched DDoS attacks disrupting Danish government, municipal, and defense-related websites, including the Ministry of Transport, Borger.dk, and Terma. The outages were brief with no data loss, and the activity aligns with wider pro-Russia targeting of European institutions.
Port Alliance, a Russian port operator handling coal and fertilizer exports, has reportedthree days of cyberattacks combining DDoS and attempted network intrusions. Terminals remain operational, but digital services were disrupted by a botnet of more than 15,000 rotated IP addresses. The goal of the attack was to destabilize operations and disrupt business processes.
Princeton University disclosed a breach of its Advancement database on November 10, lasting less than 24 hours before attackers were removed. The compromised database contained names, contact information, and fundraising records for alumni, donors, faculty, students, and parents, but did not include Social Security numbers, passwords, or financial information.
VULNERABILITIES AND PATCHES
Microsoft’s October Patch Tuesday Microsoft addressed63 vulnerabilities, including an actively exploited Windows zero-day, CVE-2025-62215, a kernel privilege escalation flaw used to gain admin access. It also addressed CVE-2025-60724, a critical GDI+ vulnerability rated 9.8 enabling remote code execution via malicious documents or uploaded files, impacting Windows and Office.
Check Point IPS provides protection against this threat (Microsoft Windows Kernel Privilege Escalation (CVE-2025-62215))
Researchers uncoveredCVE-2025-20337 and CVE-2025-5777, critical zero-day flaws in Cisco Identity Service Engine and Citrix products actively exploited against internet-facing systems. The flaws enable remote code execution without login, administrator access, and deployment of custom in-memory webshells. The exploitation began before disclosure or complete patches.
Check Point IPS provides protection against these threats (Cisco Identity Services Engine Remote Code Execution (CVE-2025-20337), Citrix NetScaler Out-of-Bounds Read (CVE-2025-5777))
Researchers analyzedCVE-2025-12480, a critical authentication bypass in the Triofox enterprise file sharing platform (CVSS 9.1). Attackers are actively exploiting it to create admin accounts and run code via the built-in antivirus feature, installing remote access tools and tunneling RDP.
Check Point IPS provides protection against this threat (Gladinet Triofox Authentication Bypass (CVE-2025-12480))
THREAT INTELLIGENCE REPORTS
Check Point Research reports on a fragmented ransomware landscape in Q3 2025, with 85 active groups and 1,592 victims listed across leak sites, averaging 535 victims per month. Qilin led activity while LockBit 5.0 returned, signaling potential recentralization. Manufacturing and business services remained the most affected sectors.
Check Point Research published its October 2025 global threat report, highlighting a continued rise in cyberattacks, with organizations averaging 1,938 weekly attacks (+5% YoY) and ransomware incidents surging 48% YoY. The report also notes escalating GenAI-related data leakage risks, with 1 in 44 enterprise prompts exposing sensitive information.
Check Point researchers analyzed a phishing campaign abusing Meta’s Facebook Business Suite and the facebookmail.com domain to deliver convincing fake notifications. More than 40,000 emails targeted over 5,000 organizations across the US, Europe, Canada, and Australia, targeting SMBs in advertising-reliant sectors, bypassing filters, and directing victims to credential-harvesting sites.
Check Point researchers profiledthe Payroll Pirates, a malvertising network impersonating payroll systems, credit unions, and trading platforms in the US. Using Google and Microsoft ads, cloaking, and Telegram bots to bypass authentication codes, it has targeted over 200 interfaces and lured more than 500,000 users, with activity spiking in September 2025.
Kent Ickler // It’s been over two years since Jordan and I talked about a Blue Team’s perspective on Red Team tools. A Blue Team’s Perspective on Red Team Hack […]
Login
as the best-positioned market leader in the 
