data-engineering-data-pipeline
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves designing and implementing data ingestion from external sources such as PostgreSQL databases, Kafka streams, and S3 storage. This creates a surface for indirect prompt injection if the processed data contains malicious instructions intended to influence the agent's behavior during transformation or analysis. \n
- Ingestion points: Data is ingested via
extract_from_database, Kafka consumers, and S3 file reads described in the instructions and code examples. \n - Boundary markers: The skill does not explicitly define prompt delimiters to separate untrusted data from system instructions, although it recommends schema validation. \n
- Capability inventory: The skill facilitates file system operations (S3 writes), database connectivity, and data transformation logic execution through Spark and dbt. \n
- Sanitization: The skill recommends the use of data quality frameworks like Great Expectations and dbt tests to validate data integrity, which provides some mitigation against malformed data inputs.
Audit Metadata