systemd
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes standard system commands such as
systemctl,journalctl,systemd-run, andloginctlto interact with the systemd manager. These operations are fundamental to the skill's purpose of managing system services and logging. - [PRIVILEGE_ESCALATION]: The instructions correctly identify the need for
sudowhen managing system-level units, while using standard user permissions for user-level units. This follows the principle of least privilege for the respective scopes. - [PERSISTENCE_MECHANISMS]: The skill is designed to create and manage systemd units (services, timers, mounts), which are the standard and intended way to establish persistence and scheduled tasks on modern Linux distributions.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from unit files and system logs (e.g., via
systemctl catorjournalctl). \n - Ingestion points: Unit files passed to the verification script and log entries retrieved via
journalctl.\n - Boundary markers: None explicitly defined in the prompt instructions for interpolating content into the LLM context, but the verification script escapes its output.\n
- Capability inventory: The skill can create unit files, enable/start services, and run transient jobs via
systemd-run.\n - Sanitization: The
scripts/verify-unit.shscript performs robust JSON escaping of command output to ensure the agent receives well-formatted data.
Audit Metadata