scikit-survival
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of external datasets, creating a potential surface for indirect prompt injection where malicious instructions could be embedded in data files to influence the agent.
- Ingestion points: The skill and its reference files (specifically references/data-handling.md) provide code templates for loading data from external CSV and ARFF files using pandas and scikit-survival utilities.
- Boundary markers: The provided templates do not implement specific boundary markers or instructions for the agent to ignore potential instructions embedded within the datasets.
- Capability inventory: The skill enables the agent to train various machine learning models (Cox models, Random Survival Forests, SVMs), make predictions, and generate plots. It does not include high-risk capabilities like arbitrary shell execution or network exfiltration.
- Sanitization: The code snippets include logic for data validation related to survival analysis (e.g., checking for non-positive time values) but lack security-oriented sanitization to detect or neutralize potential prompt injection patterns.
Audit Metadata