regression
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill utilizes standard, well-known libraries such as xgboost, scikit-learn, shap, pandas, and ibis for regression modeling. The instructions follow established data science patterns for model training and diagnostics without any malicious intent or hidden behaviors detected.
- [INDIRECT_PROMPT_INJECTION]: The skill establishes a data ingestion pipeline from external files (e.g., Parquet) which constitutes a surface for indirect prompt injection if the source data is controlled by an adversary.
- Ingestion points: The skill instructs the agent to read data from local files using
ibis.duckdb.connect().read_parquet("data/train.parquet")in SKILL.md. - Boundary markers: No specific boundary markers or instructions are provided to mitigate risks from processing untrusted data.
- Capability inventory: The skill enables fitting models, generating visualizations with SHAP and matplotlib, and logging results via MLflow.
- Sanitization: The provided patterns do not include explicit validation or sanitization steps for input data features.
Audit Metadata