Summary
- Kaspersky has documented AuditTeam infections across seven countries and territories during a campaign active from January to at least October 2026.
- The group's OpsAgent framework supports surveillance, credential access, movement through networks and the disabling of security software.
- Victims receive an extortion demand presented as a paid security audit, despite having already experienced file encryption.
A ransomware group operating across Europe, Asia and Africa has been presenting its extortion demands as paid security assessments, using a modular intrusion framework to compromise organisations, disable protective software and encrypt files.
Researchers at Kaspersky have documented activity attributed to AuditTeam, a group responsible for infections in at least seven countries and territories between January and October 2026. The findings, published on 18 November, describe a toolkit called OpsAgent that combines remote access, surveillance and system manipulation with specialised components for disabling security products and deploying ransomware.
Confirmed locations include Finland, Italy, Russia, China, Hong Kong, the Philippines and Senegal. Kaspersky observed infections across different industries, while the group’s leak site makes additional claims involving organisations in seven other countries.
AuditTeam’s ransom note offers victims a supposed vulnerability report in exchange for an audit and consulting fee, with a threat to disclose stolen information if payment is refused. The note is generated by the component responsible for encrypting files, meaning the purported consulting offer follows a malicious intrusion rather than an authorised security assessment.
OpsAgent supports several stages of an intrusion
Kaspersky began examining the framework after discovering suspicious activity at an organisation in Hong Kong on 10 March. Subsequent investigations identified related activity elsewhere and enabled researchers to reconstruct aspects of the malware’s development and command infrastructure.
OpsAgent provides operators with capabilities for exploring compromised computers, collecting information, recording keyboard activity, capturing screenshots and executing additional malicious functions. It can also establish connections between infected machines, allowing attackers to reach other parts of an organisation’s network after obtaining an initial foothold.
Such functionality supports prolonged intrusions in which attackers investigate their targets before deciding which systems to disrupt. Access to one workstation may provide information about available accounts and connected services, while additional compromised machines can expand the operator’s reach.
The researchers found that the framework had evolved considerably during the campaign. Its main implementation was rewritten from C++ to Rust, while changes to the software introduced additional command capabilities and different ways of communicating with operator-controlled infrastructure.
Rust can be used to develop efficient and portable software, but the language itself is not evidence of malicious intent. In this case, the changes are significant because they demonstrate continuing development of a framework designed for unauthorised access and control.
Kaspersky also identified several loader variants used to execute the malware. These included components written in C, .NET and Rust, with different approaches to starting the intrusion software on compromised machines.
The distinction between loaders and the main framework matters when interpreting the findings. The loaders provide routes for executing OpsAgent, while the framework supplies the wider capabilities used by its operators during an intrusion.
Defensive software becomes an attack target
One specialised module is designed to interfere with antivirus and endpoint detection tools. Kaspersky found that it uses a digitally signed driver associated with Beijing Jiangmin Antivirus to terminate selected processes running on an affected computer.
A legitimate digital signature can help a driver satisfy operating system requirements even when an attacker is using its capabilities for an unauthorised purpose. The technique illustrates why the apparent legitimacy of an individual software component does not establish that the activity it performs is safe.
Kaspersky identified approximately 220 process names targeted by the module. The list includes security tools, monitoring applications, remote access software, database services and backup-related processes.
Disabling those services can reduce an organisation’s ability to observe suspicious activity and complicate recovery. If monitoring tools stop reporting events, investigators may lose important information about what the attackers did and which systems were affected.
Interference with backups introduces another risk because the organisation may need those systems to restore information after encryption. Even where backup copies exist, recovery can be delayed if the relevant software or administrative services have been disrupted.
The separate encryption module targets files and can terminate database processes before encryption begins. Kaspersky identified support for established cryptographic methods, alongside functions for removing Windows shadow copies and disrupting recovery-related services.
These operations are consistent with a ransomware campaign seeking to increase pressure on affected organisations. They do not establish that every victim experienced identical consequences or lost all available restoration options.
A supposed audit follows file encryption
AuditTeam’s payment language is unusual because the group attempts to describe its actions as security consulting. Its ransom note presents victims with a choice between paying for a vulnerability report and having stolen information published.
The offer does not alter the character of the attack. Access was obtained without authorisation, protective systems may have been disabled and files have already been encrypted by the time the note appears.
Kaspersky attributes the OpsAgent activity to AuditTeam with high confidence, relying on the ransom note format, the group’s leak site and operational characteristics connecting the observed attacks.
However, the researchers distinguish infections identified during their investigations from claims posted by the group itself. Criminal leak sites do not provide independently verified evidence that every named organisation suffered the intrusion described.
Payment would also provide no independent assurance that an accurate security assessment would be delivered, stolen information deleted or access to compromised systems surrendered. Organisations responding to an intrusion must establish the affected systems and recover them to a trusted condition regardless of how the attacker describes the demand.
Cloud infrastructure complicates detection
The framework uses command infrastructure associated with Amazon CloudFront, a content delivery service also used by legitimate applications. Kaspersky found that OpsAgent could retrieve information about its command servers through a publicly accessible GitHub repository.
Because CloudFront infrastructure is shared between many services, blocking entire address ranges may disrupt unrelated business applications without reliably identifying the attack. More specific indicators and evidence of malicious behaviour are needed to distinguish attacker communications from ordinary activity.
The researchers also found programming patterns that may indicate assistance from large language models during development. These included repeated code structures and inconsistent implementation choices that resembled generated programming scaffolding.
Those observations do not prove that the malware was written using a particular AI system. The research establishes the framework’s capabilities and development history, while the suggestion of AI assistance remains an analytical possibility.
Kaspersky has published technical indicators and details intended to support investigation. Organisations assessing exposure need to consider those findings alongside endpoint activity, unusual account use and evidence of interference with defensive systems.
With infections identified in Finland and Italy, the campaign has a confirmed European dimension, although Kaspersky has not established a single preferred industry or victim profile. Its research covers activity observed over several months rather than a newly discovered attack beginning this week.
The 18 November analysis provides a detailed account of how AuditTeam combined surveillance, network access and ransomware with a misleading consulting demand. Its findings place particular emphasis on the group’s attempts to undermine security monitoring and recovery before presenting victims with an extortion demand.












