skills/nvidia/skills/mcore-testing/Gen Agent Trust Hub

mcore-testing

Pass

Audited by Gen Agent Trust Hub on May 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides structured instructions for running tests locally using uv, pytest, and torch.distributed.run. These operations are consistent with the skill's primary purpose of maintaining a large-scale machine learning codebase.
  • [EXTERNAL_DOWNLOADS]: The skill describes processes for downloading test datasets and "golden values" from the project's official GitHub repository. These downloads are facilitated by internal scripts (download_golden_values.py) and standard test configurations (conftest.py). All identified URLs point to well-known, official service providers and vendor-owned repositories.
  • [DATA_EXPOSURE]: The instructions reference the use of environment variables and configurable file paths (DATA_PATH, DATA_CACHE_PATH) for managing datasets. No hardcoded credentials or sensitive configuration files were detected; the skill follows standard practices for secret management by advising the use of environment-specific paths.
Audit Metadata
Risk Level
SAFE
Analyzed
May 30, 2026, 01:38 AM
Security Audit — agent-trust-hub — mcore-testing