mvcc
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE] (SAFE): The skill is entirely descriptive, providing logic and best practices for database versioning. No executable scripts or command-line operations are included.
- [PROMPT_INJECTION] (SAFE): No attempts to override agent behavior or bypass safety constraints were detected in the instructions.
- [DATA_EXFILTRATION] (SAFE): The skill does not access local files, environment variables, or hardcoded credentials.
- [EXTERNAL_DOWNLOADS] (SAFE): No external URLs or third-party package dependencies are referenced.
Audit Metadata