ml4t-feature-validation
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides standard Python code using established libraries (NumPy, SciPy, Scikit-Learn) for statistical analysis of financial features. No malicious patterns, obfuscation, or unauthorized data access were detected.
- [SAFE]: The referenced
ml4t-diagnosticlibrary aligns with the author's namespace (ml4t) and appears to be a legitimate vendor-owned resource for the skill's intended purpose. - [INDIRECT_PROMPT_INJECTION]: The skill defines functions that process external data arrays (features, targets, and dates).
- Ingestion points: Data is passed as parameters to Python functions like
validate_featureandic_decayinSKILL.md. - Boundary markers: None explicitly present for raw data ingestion.
- Capability inventory: The provided scripts are limited to mathematical operations and statistical calculations using NumPy and SciPy; they do not contain file-system writes, network operations, or command execution capabilities.
- Sanitization: The code includes basic data cleaning, such as
np.isnan()checks to handle missing values.
Audit Metadata