The ScreenConnect attack: when your IT provider is the way in
A November campaign abused a remote-support tool used by IT providers to reach nearly 50 client organisations in about 90 minutes. Here is what boards should take from it.

On the afternoon of 11 November, attackers began using ConnectWise ScreenConnect, one of the most widely used remote-support tools in the IT industry, to plant their own backdoors inside the networks of businesses that had never been targeted directly. Within roughly 90 minutes, security researchers say, at least 288 computers across 48 organisations had been compromised. None of those organisations made an obvious mistake. They were reached through the companies they trusted to manage their IT. For business leaders, the ScreenConnect attack is the clearest recent example of supply-chain risk in cyber security: the danger that your exposure is set not by your own defences, but by those of your suppliers.
This article explains what happened, why managed IT providers have become such attractive targets, what remains uncertain, and the questions every board and executive team should now be asking about the remote-access tools that sit inside their business.
What happened in the ScreenConnect attack
ScreenConnect is remote-support software. It allows a technician to see and control a computer from elsewhere, which is how most managed service providers, the outside firms that run IT for small and mid-sized companies, fix problems, install updates and support staff without visiting the office. A single provider may use it to manage thousands of machines across dozens of client businesses.
According to BleepingComputer, the security firm Huntress first detected the campaign on Wednesday 11 November, with activity beginning at around 17:47 UTC. Attackers used ScreenConnect's Run Command feature, a legitimate function technicians use to execute instructions on managed computers, to install rogue copies of the ScreenConnect client on endpoints. Those rogue clients connected not to the provider but to servers controlled by the attackers, giving them their own persistent route into each machine that would survive even after the provider's session ended.
Huntress's early count was six compromised managed service providers, more than 286 affected endpoints and 47 affected organisations, with the numbers still rising. An updated count reported by BleepingComputer the following day put the figure at a minimum of 288 endpoints across 48 organisations. All were targeted within 90 minutes, and about a third were hit within the first ten. Huntress also warned that the sequential tags attackers assigned to their victims suggested other providers may have been compromised earlier.
ConnectWise published an advisory on the night of 11 November confirming unauthorised access to ScreenConnect partner accounts and the installation of unauthorised remote-access software in customer environments. As a precaution, it disabled five third-party extensions across roughly 400 partner environments: Labtech Plus, RMM Plus, RMM+, RMM Plus Password Link and RMM+ Password Link. The company stressed that the presence of these extensions does not by itself mean an environment was compromised, and that it had no reason to believe partners without them were affected.
How the attackers got in: a third-party extension
The most important detail for business leaders is where the weakness appears to lie. It was not, on current evidence, in the core ScreenConnect product. Field Effect, summarising ConnectWise's preliminary assessment, reported that the company found no compromise of its own infrastructure and no vulnerability in ScreenConnect itself, and pointed instead to abuse of independent third-party extensions.
The Record reported that ConnectWise ultimately traced the activity to RMM+, an independently developed extension that provides an interface allowing external management tools to automate and control remote sessions. That interface is protected by a passphrase the partner sets during installation. BleepingComputer reported Huntress's assessment that versions 1.0.31 and 1.0.32 of RMM+ contained an authentication flaw: if no shared access key had been configured, a blank key passed the check, allowing commands to be issued without a ScreenConnect login or multi-factor authentication.
Investigators were careful to say that the initial access method was still being confirmed. Huntress told BleepingComputer that its telemetry pointed to compromised credentials or hijacked operator sessions, and that the activity appeared to abuse legitimate functionality rather than exploit a new software vulnerability. In short, the attackers did not need to break the lock. They found a side door that an add-on had left unlatched, and then used the tool exactly as it was designed to be used.
SCKit: a toolkit built to spread to technicians
Once inside, the attackers deployed a toolkit that researchers have named SCKit. According to BleepingComputer, Huntress and Blackpoint Cyber describe it as a set of 16 scripts and components designed to move from infected client computers back to the technicians who support them, and from there into other managed networks. Cybersecurity Dive reported that it includes PowerShell scripts, batch files and Windows command-line files deployed in several stages.
Its most unusual component, called RelayTrap, replaces the legitimate ScreenConnect process on an infected machine and waits for a technician to connect. Another, TrapWatch, monitors ScreenConnect services and prepares the replacement. Other parts attempt to steal ScreenConnect authentication cookies from web browsers, which would let an attacker hijack a session without a password or multi-factor code. Field Effect reported that commands ran with the highest system privileges, created exclusions in Microsoft Defender so that antivirus would not inspect the attackers' files, and stopped the legitimate ScreenConnect client so that a rogue one could take its place.
There is a further detail that boards should note. Huntress reported that RelayTrap appears to have been generated with the help of a large language model, with unusually detailed comments in its source code, and Field Effect said some SCKit code showed similar signs. The use of AI to write attack tooling is no longer theoretical; it is turning up in live incidents.
What did not work, and what remains unclear
It would be wrong to overstate the damage. SCKit included an attempt to exploit a known ScreenConnect client flaw, catalogued as CVE-2026-84869, which allows a file to be sent back to a technician's machine. That flaw was fixed in ScreenConnect version 26.6.5. Blackpoint researchers reproduced the attempt against a current client and found the existing patch blocked it, and told BleepingComputer they had no evidence of successful exploitation. Huntress reported that the attempt to take over technician consoles failed in every recovered host log it examined, and Field Effect said session-capture and propagation attempts had failed in every incident analysed so far.
The available reporting does not describe ransomware being deployed or data being stolen in this campaign. What it does describe is persistence: attackers establishing durable, hidden access inside dozens of organisations. That is often the first stage of something larger, and it is the reason investigators are urging thorough clean-up rather than a quick reset.
Context also matters. The Record noted that CISA, the US cyber-security agency, warned in September 2026 that CVE-2026-84869 was being exploited, and that Microsoft warned in October that a Chinese government-aligned group had used automated tools in a supply-chain attack on ConnectWise. Researchers have not attributed the November campaign to that group. But the pattern is clear: remote-management platforms are being targeted repeatedly and deliberately, by both criminal and state-aligned actors.
Why managed service providers are prime targets
Most businesses outsource at least part of their IT, and for good reason. A specialist provider brings skills, tooling and around-the-clock coverage that few mid-sized firms could build alone. But outsourcing concentrates access. A provider's remote-support console is, by design, a single place from which a technician can reach every computer of every client. To an attacker, that console is worth more than any individual company's network, because breaking into one provider can open the door to dozens of businesses at once.
The ScreenConnect campaign shows how efficiently that leverage can be used. Researchers counted nearly 50 organisations reached in about an hour and a half. No phishing emails were needed at those organisations, no employee clicked a bad link, and their own firewalls were not breached. The attackers simply arrived through a channel that was already trusted and already permitted.
It also shows that the weakest point may sit one layer further out still. RMM+ was not a ConnectWise product but an independent extension installed by partners to make their work easier. Every plug-in, integration and add-on extends the trust boundary of the core platform, and each one carries its own security quality. Many businesses do not know which remote tools their provider uses, let alone which extensions those tools run.
What business leaders should ask now
For executives, the practical response is not to abandon outsourcing but to govern it more closely. The guidance published by ConnectWise and the researchers, as reported by Cybersecurity Dive and Field Effect, translates into a short set of questions any chief executive or board can put to their IT provider this week.
First, do you use ScreenConnect or any comparable remote-access tool to manage our systems, and did any of your environments have the affected extensions installed? If so, have they been confirmed as disabled, including on any self-hosted, on-premises servers, which ConnectWise said partners must disable themselves?
Second, have you reviewed Run Command history, Shared Toolbox activity, extension installation records and authentication logs for any activity that cannot be attributed to an authorised technician? Huntress advised checking Run Command history even where no alerts had fired.
Third, if any of our devices are suspected of compromise, are you investigating them with separate security tools rather than opening interactive remote sessions? Both ConnectWise and Huntress warned that connecting to an infected machine through ScreenConnect could expose the technician, which is precisely what SCKit was built to exploit.
Fourth, have credentials been rotated and active sessions revoked from a known-clean workstation, with multi-factor authentication enforced for administrative access? And have ScreenConnect clients been updated to at least version 26.6.5?
Fifth, and most important, how do you confirm that systems are clean before returning them to service? ConnectWise warned that disabling extensions, updating software or resetting credentials does not remove access or persistence that attackers have already established. Field Effect reported that researchers saw reinfection after isolation was lifted. Clean-up must include searching for unauthorised clients, scheduled tasks and registry changes.
Building supplier risk into governance
Beyond the immediate incident, the ScreenConnect attack argues for a permanent change in how companies govern their technology suppliers. Contracts with managed providers should specify which remote-access tools are used, which extensions are permitted, how quickly the client will be notified of an incident, and what logs will be retained and shared. Remote access should be limited to the systems and hours genuinely required, rather than left open to every machine at all times. Access by third parties should be reviewed as regularly as access by employees.
Boards should also expect their security reporting to cover suppliers, not just internal controls. A dashboard that shows strong defences inside the company but says nothing about the dozen firms with administrative access to its systems is telling only half the story. This is the kind of independent review that B12 IT Solutions and other security advisers increasingly carry out for clients: mapping who can reach the business remotely, through which tools, and with what safeguards.
The B12 view
The November ScreenConnect campaign was contained faster than many supply-chain attacks, and on current evidence its most dangerous techniques failed. That is a credit to the researchers and to ConnectWise's swift decision to disable the extensions across hundreds of environments. But the lesson does not depend on the outcome. Attackers reached nearly 50 organisations in about 90 minutes, without targeting any of them directly, by abusing a trusted tool and a weak add-on.
For business leaders in Dubai and beyond, the takeaway is simple. Your cyber-security perimeter now includes every provider that can log in to your systems, and every piece of software they use to do it. The right response is not alarm, but visibility: know who has access, through what, and how you would know if it were misused. Companies that can answer those questions calmly are the ones that will treat the next incident of this kind as a routine check rather than a crisis.

