engineering-data-engineer
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill handles external data ingestion through JSON files and Kafka streams. While this is standard for data engineering, it constitutes an attack surface where external content could potentially influence agent behavior.
- Ingestion points:
SKILL.mddefines data readers for files and network streams. - Boundary markers: The skill requires the use of explicit data contracts and schema validation rules.
- Capability inventory: Capabilities include file system writes via Delta Lake and network communication via Kafka servers.
- Sanitization: The instructions provide templates for validation using the Great Expectations library.
Audit Metadata