os-scripting

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: Provides instructions for using standard system tools like ps, top, netstat, and lsof to diagnose system state. These commands are entirely consistent with the skill's purpose.
  • [PRIVILEGE_ESCALATION]: Includes examples of installing software using sudo apt or brew and managing system services. These actions are standard for administrative workflows and do not indicate unauthorized escalation.
  • [PERSISTENCE]: References cron and systemd timers as methods for legitimate task automation, which is an explicitly stated feature of the skill.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection through log analysis.
  • Ingestion points: Accesses system logs and diagnostic outputs using tools like journalctl, tail, and dmesg (SKILL.md).
  • Boundary markers: No specific delimiters are used to wrap log data or command outputs.
  • Capability inventory: Commands such as systemctl, crontab, and shell script execution are present across several phases (SKILL.md).
  • Sanitization: External data from logs is not explicitly sanitized before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 11:28 PM