senior-data-scientist
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and profile external datasets as its primary function. This creates a surface where malicious instructions could be embedded within data fields, potentially influencing the agent's behavior during analysis or visualization steps.
- Ingestion points: The skill loads and profiles datasets (Phase 1: Data Understanding).
- Boundary markers: No explicit instruction delimiters or 'ignore embedded instructions' warnings are defined for the agent to use when processing data values.
- Capability inventory: The skill uses tools for data transformation, statistical testing, visualization generation, and experiment tracking.
- Sanitization: The instructions focus on data quality (handling missing values, outliers) but do not include checks for natural language instructions embedded within data strings.
Audit Metadata