The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/agent.py uses subprocess.check_output to execute local system commands, including powershell.exe for Active Directory enumeration and the Impacket toolkit's getST.py for Kerberos protocol testing. These operations are consistent with the skill's primary purpose of security auditing.
[INDIRECT_PROMPT_INJECTION]: The script scripts/agent.py processes data from Active Directory attributes (via PowerShell) and reflects it in console output or JSON reports. This constitutes a potential attack surface if an adversary controls AD object names or properties, though the impact is limited to the integrity of the tool's own reporting.
Ingestion points: Active Directory object attributes retrieved via Get-ADObject and Get-ADComputer in scripts/agent.py.
Boundary markers: None present in the data processing or prompt interpolation logic.
Capability inventory: Subprocess execution of local binaries and file-writing capabilities (via the --output flag).
Sanitization: The script does not perform explicit sanitization of the AD attribute data before processing or outputting it.

exploiting-constrained-delegation-abuse