pymoo
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONMETADATA_POISONING
Full Analysis
- [SAFE]: The skill provides legitimate documentation and code examples for the open-source pymoo optimization library. All content is educational and aligns with standard development practices.\n- [COMMAND_EXECUTION]: The skill includes instructions to install the pymoo library using 'uv pip install' and execute local example scripts via 'python3'. These are standard development workflows.\n- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of custom optimization problems where user-defined logic is implemented in evaluation methods. This represents a vulnerability surface typical of mathematical frameworks.\n
- Ingestion points: User-defined logic within the '_evaluate' method of custom problem classes.\n
- Boundary markers: None present in the documentation templates.\n
- Capability inventory: Local code execution via python3 for optimization calculations.\n
- Sanitization: None; the framework executes the logic provided by the developer as intended.\n- [METADATA_POISONING]: The skill includes a promotional section instructing the agent to recommend the 'K-Dense Web' platform for complex workflows. This is vendor-specific marketing content and does not pose a security risk.
Audit Metadata