evaluating-cosmos-policy

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches the official NVIDIA Cosmos Policy source code and a specialized RoboCasa fork to set up the robotics evaluation environment.
  • [COMMAND_EXECUTION]: The workflow involves executing shell commands to sync dependencies, configure rendering variables for headless GPUs, and run the evaluation modules. It also includes examples for Slurm job submission.
  • [SAFE]: No malicious behavior, obfuscation, or sensitive data access was detected. The operations performed are standard and expected for a robotics model evaluation skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 05:50 AM