Skills
skills/ascend/agent-skills/triton-operator-performance-eval/Gen Agent Trust Hub

triton-operator-performance-eval

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill documentation in references/profiling-tools.md includes Python code snippets designed to automate performance profiling via the msprof system utility.
  • Evidence: The function profile_with_msprof uses subprocess.run with shell=True to execute profiling commands constructed via f-strings.
  • Note: While this pattern is often associated with command injection risks, in this context it is used for its primary intended purpose of interacting with NPU developer tools and occurs within vendor-provided reference code.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 12:38 PM