Your Technician’s Laptop Is Part of Every Client’s Security Boundary

Share Article:

Table of Contents:

A technician joins a remote-support session, and within about nine seconds, an attacker-controlled remote-access client is running on the technician’s workstation, according to a November investigation by Huntress. For MSPs, that is a reason to reconsider which devices deserve the strictest protection.

This was not a technician being persuaded to open an email attachment: Huntress observed exploitation of CVE-2026-84869, a ScreenConnect client flaw that allowed a guest-pushed file to run without a confirmation prompt. The uncomfortable lesson for MSPs is that a normal support action can become the entry point.

That should change how you classify the machine doing the supporting. Treat a technician’s workstation as privileged infrastructure, not an ordinary employee laptop with extra software.

The customer’s endpoint isn’t the only device at risk

Huntress counted at least 688,000 guest “run file” requests across 42 MSP ScreenConnect servers between October 20 and November 16; those are request counts, not 688,000 successful compromises. The distinction matters, but so does the target: attackers were waiting for technicians to connect.

Start your own assessment with a question: if this workstation were compromised, what could an attacker reach through it? Map the technician’s RMM permissions, password-vault access, customer administrative accounts, cloud portals, and backup consoles rather than stopping at the laptop’s asset record.

Microsoft’s privileged-access guidance makes the underlying principle explicit: the security of an administrative session depends on the trustworthiness of the originating device, and identity protections cannot compensate for a compromised workstation (Microsoft Learn). For an MSP, apply that principle across customer boundaries.

Not every technician should have access to every client. Your baseline should make that limitation deliberate, enforceable, and easy to verify.

Verify the client, not just the console

For this ScreenConnect flaw, ConnectWise says to verify that every technician client runs version 26.6.5 or later, including in cloud-hosted deployments; updating the server or cloud console alone does not confirm that technician clients are updated (ConnectWise). Make the installed client version part of the evidence required to close the remediation ticket.

Do not let this become a server-only patching exercise. Inventory the machines technicians actually use, including spare laptops and after-hours devices, and verify their installed clients before returning them to support duty.

Keep the different ScreenConnect risks separate, too: ConnectWise’s guidance distinguishes updating technician clients from disabling identified third-party extensions, and says neither action removes unauthorized access already established on endpoints. Use the current vendor advisory for deployment-specific actions rather than treating one completed update as an all-clear.

Build a privileged technician workstation baseline

A stricter baseline should change how technicians work, not just add another security agent. Microsoft’s privileged-workstation model separates administrative activity from email, general browsing, and productivity workloads, while applying centralized management, monitoring, and application restrictions.

Translate that into a practical MSP standard:

  • Separate everyday work from administration: Use dedicated, hardened devices for high-impact administrative work. Keep email, unrestricted browsing, and routine document handling off those devices. Do not accept a second browser profile as equivalent to a separate trusted workstation.

  • Limit customer reach: Assign named administrative identities and scope permissions to the customers and tasks each technician needs. Require a defined elevation process for broader access, with an owner and a time limit.

  • Reduce stored secrets: Prohibit customer passwords in browser stores, local text files, and scripts. Use an approved vault with scoped access, and review whether technicians can export more credentials than their jobs require.

  • Constrain software and connections: Maintain an approved set of administrative tools, restrict unnecessary local administrator rights, and test application and network controls against real support workflows before enforcement.

  • Verify device health: Require patch compliance, functioning endpoint protection, encryption, and retained security telemetry before granting privileged access. Where supported, enforce those conditions through access policy rather than a checklist alone.

A jump host should not be your excuse to leave the originating laptop unprotected. Microsoft’s “clean source” principle requires the systems controlling a protected resource to be at least as trustworthy as that resource.

For a small MSP, prioritize devices with the broadest customer access first. Document temporary exceptions with an accountable owner, compensating controls, and an expiration date instead of quietly making convenience the permanent standard.

Monitor how support tools behave

In the observed attack, the payload was another ScreenConnect client, rebranded to resemble Windows Update and connected to an attacker’s relay. That makes “we allow ScreenConnect” an inadequate definition of approved activity.

Record which instances, relay destinations, installation paths, and administrative uses are authorized. Ask your detection provider to distinguish the approved deployment from a second, unauthorized copy rather than broadly suppressing alerts for familiar remote-support software.

Huntress recommends investigating technician clients launching scripts from temporary folders, unexpected Startup-folder content, and ScreenConnect clients connecting to relays the MSP does not own. Build those behaviors into your review, but investigate them in context instead of treating a filename alone as proof.

The same discipline should govern support sessions. Require technicians to tie an unfamiliar session to a verified customer request before joining it, and define who reviews suspicious guest activity.

Contain the access, not just the laptop

A workstation response plan is incomplete if it ends with deleting the payload. ConnectWise explicitly warns that software updates and credential resets do not replace investigation and remediation of an existing compromise.

Build the following decisions into your technician-compromise playbook:

  • Isolate and preserve: Isolate the suspected workstation and preserve relevant endpoint and platform evidence. Use trusted security tooling rather than opening another ScreenConnect session to investigate a suspected compromised device, consistent with ConnectWise’s response guidance.

  • Revoke from a clean device: Invalidate affected ScreenConnect and identity-provider sessions, then reset relevant credentials; password resets alone do not invalidate every existing session. Include other administrative platforms in the review and verify their session-revocation behavior.

  • Assess reachable secrets: Huntress recommends rotating credentials reachable from the compromised workstation, including RMM, PSA, password-vault, and customer administrative credentials. Also assess exposed API keys, certificates, and other non-password credentials for replacement.

  • Scope customer exposure: Build a timeline of the technician’s access and review reachable customer environments for suspicious sign-ins, commands, account changes, and remote-access installations. Prioritize observed activity, but do not exclude a client merely because it was not the last support session.

  • Restore deliberately: Establish a trusted workstation before restoring privileged access. Require documented remediation and monitoring checks, not just a clean scan or the absence of a known indicator.

Separate “potentially reachable” from “confirmed affected” in both the investigation and customer communications. Where logs are missing, record that uncertainty rather than treating it as evidence that nothing happened.

Make this week’s review about one technician

Choose a technician with broad customer access and walk through the full workflow. Verify the installed support client, identify where administrative sessions originate, list the credentials and customers reachable from that device, and test who can revoke access if it is compromised.

Use the findings to create a baseline you can apply across the team. The goal is not to stop technicians from supporting customers; it is to make sure the device doing that work deserves the trust you place in it.

Schedule a Demo of Blacksmith!

Check Out Our Compliance Podcast on Spotify!