it-operations

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONPERSISTENCE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides numerous Bash and PowerShell scripts for server provisioning, firewall configuration (using ufw and iptables), and service management (systemctl). These commands are intended for administrative use and are consistent with the IT Operations domain.\n- [EXTERNAL_DOWNLOADS]: The skill references and provides scripts to download infrastructure components, such as the Prometheus node_exporter, directly from official GitHub releases. These downloads target a trusted source and are documented as part of standard setup procedures.\n- [INDIRECT_PROMPT_INJECTION]: The skill documents workflows for handling external data from user reports and ChatOps integrations (e.g., Slack bot example in reference/automation.md). While this introduces a potential vulnerability surface, the provided scripts mitigate risk through input validation and secure execution patterns.\n
  • Ingestion points: Slack commands and user alerts in reference/automation.md and reference/incident-management.md.\n
  • Boundary markers: None explicitly defined in the reference code.\n
  • Capability inventory: Subprocess execution for user and service management in reference/automation.md.\n
  • Sanitization: Use of service whitelists and list-based subprocess.run calls.\n- [DYNAMIC_EXECUTION]: Several automation scripts in reference/automation.md and reference/infrastructure.md use the Python subprocess module to dynamically execute system commands based on inputs like usernames or service names. These scripts follow security best practices by avoiding shell execution and using list-based arguments.\n- [PERSISTENCE]: The skill includes instructions for managing system persistence, such as configuring SSH keys, creating systemd services, and setting up cron jobs. These are standard and necessary components for automated infrastructure management.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:00 PM