synthetic-data-generation
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- Prompt Injection (SAFE): No patterns detected that attempt to override system instructions or bypass safety guardrails.
- Data Exposure & Exfiltration (SAFE): No hardcoded credentials or sensitive file paths were found. The use of Databricks Unity Catalog and Volumes follows platform best practices.
- Unverifiable Dependencies & Remote Code Execution (SAFE): The skill utilizes well-known, trusted libraries (Faker, NumPy, Pandas, PySpark). While it describes a workflow for executing code on Databricks, it does not involve suspicious remote script downloads or piped execution.
- Indirect Prompt Injection (LOW):
- Ingestion points: User-defined configuration variables (seeds, volumes, distributions) used for data generation.
- Boundary markers: None explicitly defined in the README documentation.
- Capability inventory: Ability to execute Python and Spark scripts on Databricks clusters via MCP tools.
- Sanitization: Not specified in the high-level documentation, though the intended use case is for generating test/demo data in a controlled environment.
Audit Metadata