ConnectWise has taken the precautionary step of disabling five third-party extensions within partner environments following alarming reports of unauthorized access and active malware campaigns abusing the ScreenConnect platform.
In a formal security advisory published on its official site, the company stated that there is currently no evidence indicating that the incident involves a direct compromise of its own corporate infrastructure or the exploitation of an underlying zero-day vulnerability within the core ScreenConnect software itself. Instead, the unfolding situation points toward external abuse targeting connected administrative setups and administrative workflows.
According to ConnectWise, initial assessments suggest that the malicious activity involves threat actors abusing independent third-party extensions previously installed inside partner environments. The company noted that it is continuing to validate this preliminary assessment, while the exact initial access vector and the full scope of the security incident remain under active investigation by internal security teams and external partners.
As part of its containment strategy, ConnectWise has identified approximately 400 partner environments where at least one of the flagged third-party extensions was installed. However, the company emphasized that the mere presence of these extensions within an environment does not automatically indicate or establish a compromise. While these specific extensions provide an application programming interface capable of executing remote commands, access to that API is designed to be secured by a unique passphrase configured directly by the partner during the initial setup process.
To mitigate potential risks, ConnectWise has strongly urged on-premises partners to manually disable the affected extensions. Furthermore, the company is recommending that all partners ensure comprehensive security controls are actively running across their networks, preserve suspicious activity logs for forensic analysis, and maintain heightened vigilance for anomalous operational behaviors.

The emerging threat landscape surrounding this activity has also drawn the attention of prominent cybersecurity firms. Huntress, a managed detection and response company actively tracking the campaign, reported that it has observed threat actors successfully leveraging access to cloud-hosted ScreenConnect panels to push remote run commands. The primary objective of these commands is to silently install rogue, attacker-controlled ScreenConnect clients across at least 288 managed endpoints spread across 48 distinct organizations. This malicious activity was first detected on November 11, 2026, at approximately 5:47 p.m. UTC.
Huntress confirmed to industry researchers that it has found no evidence of a novel software vulnerability, noting instead that the exact methodology utilized by attackers to gain initial panel access remains unclear. Current technical telemetry points heavily toward the exploitation of compromised administrator credentials or the hijacking of active operator sessions to establish initial footholds within targeted environments.
Security analysts describe the overall attack as unfolding in a sophisticated, worm-like fashion utilizing a multi-step propagation process. The rogue clients are leveraged to deliver SCKit, a custom malicious toolkit designed specifically to establish persistent remote access and create a specialized relay trap capable of intercepting session cookies.
According to Huntress researchers Ben Nahorney and Susannah Matt, the infection path operates in distinct stages, moving from an initially infected endpoint to a technician who connects to it, and potentially from a technician’s workstation outward to every endpoint that their management panel can reach. However, researchers noted that while the full breadth of capability has been uncovered through deep payload analysis, not every mechanism has been actively witnessed in every live incident.
The SCKit toolkit is typically delivered as an encrypted payload alongside a Base64-encoded dropper script known as sckit-run.bat. Once executed, this initial script contacts a remote server to fetch a secondary PowerShell payload, designated as sckit-decoder.ps1, which is ultimately responsible for launching the malicious toolkit on the target system. Additional components of the malware are carefully engineered to add exclusions to Microsoft Defender, install a secondary attacker-controlled ScreenConnect client instance, establish a persistent scheduled task named TrapWatch, and replace legitimate ScreenConnect installations with rogue, attacker-controlled copies.
Intriguingly, forensic analysis of the scripts has revealed strong indicators that at least one of the malicious scripts was generated using a large language model. This is evidenced by unusually detailed source code references and embedded commentary explaining the internal functionality of the code.

While widespread lateral movement across technician workstations has not yet been widely observed in the wild, researchers have reconstructed additional malicious capabilities hidden within the scripts and configuration tools. This hidden functionality is designed to fetch an advanced dropper script known as Run-Lateral-Full.bat, which subsequently connects to an external server hosted at ns.mo7sen[.]dev to download a specialized loader. This loader gathers the workstation’s unique hardware identifier, transmits it back to the attacker infrastructure, and receives a tailored, encrypted kit specifically compiled for that particular workstation.
Once the payload executes across reachable sessions as the SYSTEM user on targeted endpoints, it installs the trap and perpetuates the infection cycle. Approximately sixty seconds after the initial broadcast, a utility component designated as ToolboxRelayUploader.exe automatically deletes the shared toolbox item to cover its tracks and hinder immediate discovery.
Adding to the urgency of the situation, security observers noted that threat actors have been systematically tagging each compromised managed service provider with a sequential number. Because the tags observed by researchers began partway through an ongoing numerical sequence, industry experts believe that numerous other MSPs may have been compromised earlier in the campaign. ScreenConnect users and administrators have been strongly advised to thoroughly check their historical Run Command logs, even if their monitoring systems have not yet triggered formal security alerts.
In a parallel security advisory, cybersecurity firm Sophos reported that it has similarly observed unauthorized ScreenConnect client deployments following the exact same operational modus operandi. This malicious activity has been traced back to specific ScreenConnect instance IDs acting as command sources, which are subsequently used to execute arbitrary commands and deploy additional malicious instances across victim networks.
Sophos detailed that administrative tools like curl.exe were abused to download a malicious installer package from external IP addresses into temporary system directories, where scripted command files then silently executed the installation of unauthorized remote access clients. Researchers also observed sophisticated batch file staging and PowerShell execution utilizing execution-policy bypass techniques to retrieve secondary scripts from external command-and-control infrastructure.
As investigations continue, affected organizations and security teams are advised to review administrative access logs, verify the integrity of installed extensions, and monitor network traffic for indicators associated with the ongoing campaign.
