governance
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides instructions for security auditing, specifically targeting PII detection, data lineage extraction, and the generation of least-privilege IAM policies. No malicious patterns, obfuscation, or unauthorized access attempts were detected.\n- [INDIRECT_PROMPT_INJECTION]: The skill identifies a data ingestion surface where the agent processes external ETL code and table schemas.\n
- Ingestion points: Analyzes user-provided table schemas and ETL code (Spark, dbt, SQL) to detect PII and lineage.\n
- Boundary markers: None explicitly defined in the instructions for separating untrusted data from processing logic.\n
- Capability inventory: The skill is limited to generating text-based recommendations, SQL masking functions, and IAM policy JSON; it lacks direct system execution or network capabilities.\n
- Sanitization: No specific sanitization or filtering of the input code/schemas is described.
Audit Metadata