skills/nvidia/skills/doca-sta/Gen Agent Trust Hub

doca-sta

Pass

Audited by Gen Agent Trust Hub on Oct 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is documented as having no runnable code (documentation_only: true) and acts solely as a guidance provider for the agent to assist users with the DOCA STA library. All suggested commands, such as pkg-config and dmesg, are intended for manual user execution or agent-assisted environment verification and are appropriate for the development context.\n- [SAFE]: No malicious patterns such as prompt injection, obfuscation, or unauthorized data exfiltration were detected. External links and repository references point to official NVIDIA and well-known documentation sources.\n- [SAFE]: Mentions of elevated privileges (sudo) are presented as informational requirements for system-level diagnostics and are consistent with the skill's primary purpose of low-level hardware interaction.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 16, 2026, 05:02 AM