T1134 - Access Token Manipulation

Adversaries may modify access tokens to operate under a different user or system security context to perform actions and bypass access controls. Windows uses access tokens to determine the ownership of a running process. A user can manipulate access tokens to make a running process appear as though it is the child of a different process or belongs to someone other than the user that started the process. When this occurs, the process also takes on the security context associated with the new token.

An adversary can use built-in Windows API functions to copy access tokens from existing processes; this is known as token stealing. These token can then be applied to an existing process (i.e. Token Impersonation/Theft) or used to spawn a new process (i.e. Create Process with Token). An adversary must already be in a privileged user context (i.e. administrator) to steal a token. However, adversaries commonly use token stealing to elevate their security context from the administrator level to the SYSTEM level. An adversary can then use a token to authenticate to a remote system as the account for that token if the account has appropriate permissions on the remote system.(Citation: Pentestlab Token Manipulation)

Any standard user can use the runas command, and the Windows API functions, to create impersonation tokens; it does not require access to an administrator account. There are also other mechanisms, such as Active Directory fields, that can be used to modify access tokens.

Atomic Tests:

Currently, no tests are available for this technique.

Detection

If an adversary is using a standard command-line shell, analysts can detect token manipulation by auditing command-line activity. Specifically, analysts should look for use of the runas command. Detailed command-line logging is not enabled by default in Windows.(Citation: Microsoft Command-line Logging)

If an adversary is using a payload that calls the Windows token APIs directly, analysts can detect token manipulation only through careful analysis of user network activity, examination of running processes, and correlation with other endpoint and network behavior.

There are many Windows API calls a payload can take advantage of to manipulate access tokens (e.g., LogonUser (Citation: Microsoft LogonUser), DuplicateTokenEx(Citation: Microsoft DuplicateTokenEx), and ImpersonateLoggedOnUser(Citation: Microsoft ImpersonateLoggedOnUser)). Please see the referenced Windows API pages for more information.

Query systems for process and thread token information and look for inconsistencies such as user owns processes impersonating the local SYSTEM account.(Citation: BlackHat Atkinson Winchester Token Manipulation)

Look for inconsistencies between the various fields that store PPID information, such as the EventHeader ProcessId from data collected via Event Tracing for Windows (ETW), Creator Process ID/Name from Windows event logs, and the ProcessID and ParentProcessID (which are also produced from ETW and other utilities such as Task Manager and Process Explorer). The ETW provided EventHeader ProcessId identifies the actual parent process.

Shield Active Defense

Software Manipulation

Make changes to a system’s software properties and functions to achieve a desired effect.

Software Manipulation allows a defender to alter or replace elements of the operating system, file system, or any other software installed and executed on a system.

Opportunity

There is an opportunity for the defender to observe the adversary and control what they can see, what effects they can have, and/or what data they can access.

Use Case

A defender could feed or redirect requests for credentials with false data that can be used to direct an adversary into a decoy network or system.

Procedures

Hook the Win32 Sleep() function so that it always performs a Sleep(1) instead of the intended duration. This can increase the speed at which dynamic analysis can be performed when a normal malicious file sleeps for long periods before attempting additional capabilities. Hook the Win32 NetUserChangePassword() and modify it such that the new password is different from the one provided. The data passed into the function is encrypted along with the modified new password, then logged so a defender can get alerted about the change as well as decrypt the new password for use. Alter the output of an adversary’s profiling commands to make newly-built systems look like the operating system was installed months earlier. Alter the output of adversary recon commands to not show important assets, such as a file server containing sensitive data.