trim-loss-minimization
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data which could potentially influence downstream AI behavior if malicious instructions are embedded in the cutting operation records.
- Ingestion points: Data is ingested through
material_id,waste_breakdownkeys, and material specifications in theTrimLossAnalyzerclass withinSKILL.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the processing logic.
- Capability inventory: The skill performs mathematical calculations and generates visualizations using
matplotlib, including the ability to save files to disk viaplt.savefig. - Sanitization: There is no evidence of validation or sanitization for the input fields provided by the user.
- [DYNAMIC_EXECUTION]: The skill performs local dynamic imports within its optimization functions.
- Evidence: Found
from skills.one_d_cutting_stock import ColumnGenerationCuttingStockandfrom pulp import *inside function bodies inSKILL.md. While these target local or standard libraries, they represent dynamic loading patterns.
Audit Metadata