skills/g1joshi/agent-skills/podman/Gen Agent Trust Hub

podman

Pass

Audited by Gen Agent Trust Hub on Oct 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill provides accurate and helpful instructions for managing containers with Podman, consistently promoting secure best practices such as rootless operation.
  • [COMMAND_EXECUTION]: The skill includes standard shell commands for managing the OCI container lifecycle (e.g., podman run, podman pod create) and Linux service management (systemctl). These are expected for the skill's purpose.
  • [PERSISTENCE]: The documentation explains how to configure container autostart using systemd user services and loginctl enable-linger. These are standard persistence mechanisms for server-side applications and are described transparently for their intended use case.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 11, 2026, 03:37 AM