profile-system

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill utilizes standard system diagnostic tools such as lspci, nvidia-smi, vainfo, and ffmpeg to detect hardware capabilities. All commands are used according to their intended diagnostic purposes.
  • [SAFE]: Data storage is restricted to a dedicated local directory for plugin data, following standard environment variable conventions (e.g., XDG_DATA_HOME). No evidence of sensitive file access or credential harvesting was detected.
  • [SAFE]: The skill does not perform any network requests, avoiding risks of data exfiltration.
  • [SAFE]: No obfuscation, prompt injection, or remote code execution patterns are present in the provided instructions or shell snippets.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 03:34 PM
Security Audit — agent-trust-hub — profile-system