Skills
skills/arist12/amd-skills/env-probe/Gen Agent Trust Hub

env-probe

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The references/env_probe.py script executes local system utilities to retrieve hardware and configuration metadata.
  • Evidence: It uses subprocess.run to execute rocm-smi and rocminfo. These calls use static arguments and do not involve shell invocation or unvalidated input.
  • [EXTERNAL_DOWNLOADS]: The documentation and diagnostic script reference the installation and verification of well-known machine learning libraries.
  • Evidence: The skill checks for and suggests the installation of standard packages such as torch, triton, and AMD-specific optimization libraries like aiter and ck4inductor.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 06:42 AM