dbt-transformation-patterns
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided source material and external data, creating a potential surface for indirect prompt injection.
- Ingestion points: The skill instructions mention inspecting "source material", "available files", and "external data source".
- Boundary markers: There are no specific instructions for using delimiters or protective markers to isolate untrusted input within the agent's context.
- Capability inventory: The agent is authorized to modify local files (dbt transformations) and perform network interactions via the sandbase_run tool.
- Sanitization: While quality gates and validation steps are required, there is no explicit instruction to sanitize or filter untrusted data for embedded commands.
Audit Metadata