model-security-scanning
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of documentation and security guidelines for model loading pipelines. There are no executable scripts, command line operations, or network requests. The instructions do not attempt to override agent behavior or access sensitive data. It correctly identifies security risks such as Python pickle deserialization and recommends safer alternatives like SafeTensors.
Audit Metadata