The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/agent.py executes PowerShell commands to audit system security settings.\n
The run_ps function utilizes subprocess.check_output to interact with the Windows environment.\n
Commands targeted include net accounts for password policies, auditpol for auditing configuration, and Get-NetFirewallProfile for firewall verification.\n- [PERSISTENCE]: The skill provides instructions to establish a weekly scheduled task for automated security scanning.\n
Use of the Register-ScheduledTask PowerShell cmdlet allows the system to run compliance checks automatically as a service.\n
This mechanism is an intended feature for continuous security compliance monitoring and drift detection.\n- [INDIRECT_PROMPT_INJECTION]: The scripts/process.py script parses external XML files generated by assessment tools.\n
Ingestion points: Data enters the process via the parse_ciscat_xccdf_results function which reads XML files provided as command-line arguments.\n
Boundary markers: None identified in the script to distinguish between trusted data and potentially malicious embedded instructions.\n
Capability inventory: The script performs file read/write operations and generates JSON/CSV reports.\n
Sanitization: The script uses standard XML parsing libraries without explicit sanitization or validation of the input against a strict schema.\n- [EXTERNAL_DOWNLOADS]: The documentation references external security resources for downloading benchmarks and GPO templates.\n
The skill directs users to official sources such as cisecurity.org and microsoft.com for security baselines and tools like LGPO.exe.

hardening-windows-endpoint-with-cis-benchmark