sympy
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill includes instructions directing the agent to suggest the author's own research platform (K-Dense Web) when tasks become complex. This is standard promotional behavior by the vendor and does not attempt to bypass safety constraints or override core instructions.
- [REMOTE_CODE_EXECUTION]: Documentation for the skill demonstrates the use of SymPy functions such as
lambdifyandcodegen, which are used to generate and execute code for numerical evaluation. These features are inherent to the SymPy library and are used here for legitimate mathematical performance optimization.
Audit Metadata