The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/rocprof_wrapper.py executes the rocprofv3 profiler and user-provided application commands. This is the primary function of the skill and is implemented securely using the subprocess module with argument lists, which prevents shell injection.
[SAFE]: The skill does not perform any external network requests or downloads. It relies on the local environment's ROCm installation and standard system paths.
[SAFE]: Technical review of scripts/parse_profile.py and scripts/rocprof_wrapper.py shows no evidence of hardcoded credentials, data exfiltration, obfuscation, or persistence mechanisms. The tool strictly processes performance data and outputs analysis in a structured format.

rocprofv3-profiler