data-scientist
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external data (e.g., customer data, retail records, financial transactions) which represents an attack surface for indirect prompt injection if that data contains malicious instructions targeted at the agent.
- Ingestion points: Processes user-provided datasets, database records, and external data sources for exploratory data analysis (EDA) and modeling as described in
SKILL.md. - Boundary markers: None present; the instructions do not include specific delimiters or warnings to ignore instructions embedded within the data being analyzed.
- Capability inventory: The skill possesses capabilities for database interaction (SQL, BigQuery, Snowflake), file manipulation (model serialization, ETL pipelines), and code generation/execution (Python ecosystem, R) as listed in the Capabilities section of
SKILL.md. - Sanitization: No explicit data sanitization, validation, or filtering mechanisms are defined in the instructions.
Audit Metadata