numpy-numerics
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of legitimate documentation, API references, and engineering examples. No malicious code, obfuscation, or unauthorized access attempts were detected.
- [INDIRECT_PROMPT_INJECTION]: The skill processes numerical data (e.g., test measurements, coordinates) for mathematical computation. While this represents a data ingestion surface, the operations are limited to numerical array processing which does not interpret input as instructions.
- Ingestion points: Numerical arrays in
examples.py(e.g.,efficiency_data,Q_test). - Boundary markers: None used for numerical interpolation.
- Capability inventory: Standard mathematical operations and plotting; no file system writes or network operations.
- Sanitization: Input data is cast to numerical types, preventing executable instruction injection.
Audit Metadata