data-pipeline-builder
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user requirements and schemas to generate implementation code, creating a surface where malicious input could influence the generated output. * Ingestion points: Requirements gathering phase for data sources, destinations, and transformations in the workflow defined in SKILL.md. * Boundary markers: Absent. The templates do not use specific delimiters or instructions to treat user-provided business logic as untrusted data during code generation. * Capability inventory: The skill uses the Bash, Write, and Edit tools. The generated code includes full database extraction, transformation logic, and loading capabilities (loaders, extractors). * Sanitization: None detected. User-provided field mappings and SQL-based transformation rules are mapped directly to implementation files without explicit validation or escaping.
Audit Metadata