Endpoint Craft

Six Questions to Ask Your IT Provider About Remote Access

Back to Insights

The tool that lets your provider in can also let others in

Almost every business that outsources IT support has given someone else a key to its devices. Remote monitoring and management tools are how a provider patches laptops, fixes a broken printer driver at 8am, or helps a new starter on their first day. That convenience is genuine, and for most organisations it is the right way to run IT. It also means your security now depends partly on how carefully your provider looks after that key.

In November 2026, Microsoft Threat Intelligence published research on SCKit, a malicious toolkit that abuses ScreenConnect, a widely used remote support platform. It installs unauthorised remote access clients on managed endpoints, then waits for a support technician to connect. When they do, it tries to compromise the technician's own workstation, steal authenticated sessions, and use the provider's management console to reach other customers' devices. Microsoft noted that the original way in was still under investigation, but the lesson for buyers is already clear.

ONE ENDPOINT, MANY CUSTOMERSCustomerendpointTechnicianconnectsSessionhijackedOthercustomersReach is set by the permissions of the hijacked account

The supply chain risk in outsourced IT: a single trusted connection can become a route to many organisations.

This is not an argument against outsourcing. It is an argument for choosing a provider whose own house is in order, and for knowing what questions to ask before you sign.

Why provider hygiene is now a buying criterion

Most procurement conversations about IT support focus on response times, price per user and which hours are covered. Those matter. What is often missing is any discussion of how the provider secures the tools that connect into your estate. Yet a provider's remote access platform sits in a uniquely privileged position: it typically runs with high privileges on every managed device, and a single technician account may be able to reach dozens of customers.

Attackers understand this. Compromising one provider can be far more efficient than attacking a hundred small businesses individually. That makes your provider's operational security part of your own risk profile, whether or not it appears in the contract.

Six questions to ask your IT provider

You do not need to be a security specialist to have this conversation. A good provider will welcome these questions and answer them plainly.

1. Which remote tools can reach our devices?

Ask for a list. A disciplined provider maintains an inventory of approved remote management tools and can tell you exactly what is installed on your endpoints. Microsoft's own guidance after SCKit includes keeping an inventory of authorised remote tools and using application control to block anything else. If the answer is vague, that is a signal.

2. How are technician accounts protected?

Look for phishing resistant multifactor authentication, named accounts rather than shared logins, and permissions scoped so that one technician cannot reach every customer by default. The SCKit research makes the point that how far an attack spreads is set by the permissions of the account it hijacks. Least privilege is not a theoretical nicety here.

3. How quickly are agents and clients patched?

Remote support platforms receive security updates like any other software, and those updates do not always reach every installed client automatically. In the SCKit case, ConnectWise noted that updating the cloud console did not update technician clients. Ask how your provider tracks versions across their tooling, not just across your laptops.

4. Are the technicians' own workstations hardened?

SCKit was built to jump from a customer device to the technician who connects to it. The machines your provider's staff work from deserve at least the protection you would expect on your own devices: endpoint detection and response, tamper protection, and controlled admin rights.

5. What happens if something goes wrong?

A mature provider has an incident plan that includes their own platform. Microsoft recommends investigating suspected devices through an out of band channel rather than the potentially compromised remote tool, invalidating sessions rather than relying on password changes alone, and rebuilding devices when their integrity cannot be confirmed. Ask whether your provider would recognise and follow that kind of process.

6. Can we see the evidence?

Policies are easy to write. Ask what you can actually see: audit logs of remote sessions against your devices, reports on agent versions, and a named contact for security concerns.

WHAT GOOD LOOKS LIKEVisible audit and reportingTested incident responseLeast privilege technician accessApproved, patched remote tools

Each layer builds on the one below: inventory and patching first, visibility last.

Reducing your dependence on any single tool

There is a broader design point too. Organisations built on modern Microsoft endpoint management rely less on third party remote agents in the first place. Intune can deploy applications, enforce configuration and run remediation scripts without a separate always on remote access client. Defender for Endpoint provides live response for investigation. Remote Help offers Microsoft-native remote assistance governed by Entra ID roles and Conditional Access.

That does not mean every third party tool should go. It means each one should earn its place, and the fewer privileged agents sit on your devices, the smaller the surface an attacker can borrow.

Practical steps for this quarter

  • Request a list of every remote management tool your provider uses on your estate, and confirm each one is still needed.
  • Ask for written answers to the six questions above and keep them with your supplier records.
  • Add remote access security to your next contract review alongside response times and cost.
  • Review whether Intune and Defender capabilities you already licence could replace some third party agents.

Conclusion

Outsourcing IT remains the right decision for many organisations, but trust should be informed rather than assumed. The SCKit research is a timely reminder that a provider's remote tooling is part of your attack surface, and that the best partners treat their own security with the same seriousness they apply to yours.

If you would like an independent view of how your endpoints are managed today, or help moving towards a leaner, Microsoft-native approach with Intune and Defender, Endpoint Craft is happy to talk it through.

Ready to bring your idea to life?

Get in touch to talk through your goals and find out how Endpoint Craft can help you work smarter, stay secure, and embrace the AI era.

Contact Now →