The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/agent.py executes PowerShell commands via subprocess.check_output to audit security configurations such as password policies, audit policies, and firewall states. The commands are constructed from hardcoded templates and internal constants, minimizing the risk of command injection.
[EXTERNAL_DOWNLOADS]: The skill recommends downloading documentation and GPO templates from trusted domains including cisecurity.org and microsoft.com for benchmark compliance.
[SAFE]: The skill's operations are limited to reading system configuration and generating compliance reports. No evidence of data exfiltration, credential harvesting, or unauthorized persistence mechanisms was found. The registry and system queries performed are strictly aligned with the stated purpose of endpoint hardening.

hardening-windows-endpoint-with-cis-benchmark