chai

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS] (SAFE): The skill references the official chai_lab Python package and its repository on GitHub. These are standard and authoritative sources for the Chai-1 model and do not pose a security risk.
  • [COMMAND_EXECUTION] (SAFE): Examples for executing 'modal run' and 'chai-lab predict' are appropriate for the tool's scientific purpose and do not involve unauthorized system modifications or privilege escalation.
  • [DATA_EXFILTRATION] (SAFE): Analysis of the provided scripts and instructions shows no attempts to access sensitive local files or exfiltrate user data.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:08 PM