Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Is COM Hijacking?

COM hijacking redirects a Component Object Model class identifier to attacker code by writing a registry key under the current user hive, giving quiet persistence without administrator rights.

Illustration for the cybersecurity term COM Hijacking
Home / Blog / Cybersecurity Terms / What Is COM Hijacking?
17 November 2026

What Is COM Hijacking?

What Is COM Hijacking?

COM hijacking is a Windows persistence and execution technique in which an attacker redirects a Component Object Model class identifier so that Windows loads the attacker’s code instead of the intended library. It requires no administrator rights, survives reboots, and leaves behind a registry key rather than a file in a startup folder. MITRE ATT&CK catalogues it as T1546.015.

The Registry Behaviour That Makes It Possible

COM is the object model that much of Windows is built on. Every COM class has a globally unique class identifier, a CLSID, and the registry maps that CLSID to the DLL or executable that implements it. The mapping lives in two places: HKEY_LOCAL_MACHINE\SOFTWARE\Classes\CLSID for the whole machine, and HKEY_CURRENT_USER\SOFTWARE\Classes\CLSID for the signed-in user.

When a process resolves a CLSID, the user hive is consulted first. This is by design — it allows per-user component registration without administrator rights — and it is the entire vulnerability. An attacker running as a standard user creates a key for an existing CLSID under HKCU, with an InprocServer32 value pointing at their DLL. The next time any process in that user’s session instantiates the object, their DLL is loaded into it.

Choosing a Target CLSID

The craft lies in picking the right identifier. Attackers look for:

Because a legitimate library is still needed for the host application to work, better implementations proxy the original: load the real DLL, forward the exported functions, and run the malicious code alongside it. The user notices nothing.

Who Uses It

COM hijacking has a long operational history. APT28 used it to persist through a scheduled task referencing a hijacked class. The Ramsay espionage toolkit, several Trickbot loaders, and various banking trojans have all used the technique, and it appears regularly in commodity loaders because the code is short and the registry write is unprivileged. It is also a route to AMSI bypass, by redirecting the CLSID of the AMSI provider itself.

Detection

The technique is quiet but not invisible, and the signal quality is unusually good:

Reducing the Opportunity

You cannot disable per-user COM registration without breaking Windows, so the controls are compensating. Application control, with Windows Defender Application Control or AppLocker in DLL enforcement mode, stops the hijacking DLL from loading even when the registry points at it. Keeping user profile directories out of the set of locations from which code may execute does the same job. Restricting local administrator rights does not prevent the technique, but it limits what the loaded code can do next, which is where least privilege earns its place.

Treat a confirmed COM hijack as evidence of an established foothold rather than an isolated artefact. Something put it there, and that something had code execution on the host. Our digital forensics and incident response services work backwards from exactly these artefacts to the initial access.

FAQ – Frequently Asked Questions

Does COM Hijacking Need Administrator Rights?

No, and that is its main attraction. The attacker writes to the current user’s own registry hive, which any standard account may do, and Windows consults that hive before the machine-wide one. Administrator rights would allow a machine-level hijack affecting all users, but the unprivileged, per-user variant is sufficient for persistence and is far more common.

How Is It Different From DLL Side-Loading?

DLL side-loading relies on an application’s library search order, placing a malicious DLL where a legitimate executable will find it on disk. COM hijacking changes a registry mapping so that the system resolves a class identifier to the attacker’s library instead. Both end with a signed process loading hostile code, but the artefact you hunt for is different: a file in one case, a registry value in the other.

What Registry Keys Should I Monitor?

Focus on key and value creation under HKEY_CURRENT_USER\Software\Classes\CLSID, especially InprocServer32 and LocalServer32 values, and the equivalent paths under HKEY_USERS for other logged-on profiles. Sysmon events 12 and 13 capture these. Also review scheduled tasks that use a COM handler action, since they provide the trigger that makes the persistence useful.

Will Removing the Registry Key Clean the Host?

It removes that persistence mechanism, not the compromise. Whatever created the key had code execution on the machine, and intrusions normally install more than one foothold. Preserve the key, the referenced DLL, and surrounding telemetry, then investigate how the attacker arrived and what else they did before deciding whether cleaning or rebuilding is appropriate.

Related Terms

You May Be Interested In These

What Is Forward Secrecy?

Cybersecurity Terms

17 November 2026

What Is Forward Secrecy?

Read More
What Is IPsec?

Cybersecurity Terms

17 November 2026

What Is IPsec?

Read More
What Is Symmetric Encryption?

Cybersecurity Terms

17 November 2026

What Is Symmetric Encryption?

Read More