mantis-calibrate
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Dynamic Script Generation: The skill creates and runs a local Python helper script (
append_calibrate.py) to update finding files. This dynamic execution pattern is implemented to optimize performance and manage data within the workspace environment. - Data Processing Considerations: The skill ingests data from security findings and threat models to calculate risk scores. This process involves handling external content that could theoretically contain instructions intended to influence agent output, although the skill utilizes structured scoring rules that provide a framework for these calculations.
Audit Metadata