persistence-layer-change
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided schema changes to generate migration artifacts and update source code, creating a potential surface for indirect prompt injection.\n- Ingestion points: The skill takes 'Desired schema change' and 'Backward compatibility needs' as inputs in SKILL.md.\n- Boundary markers: There are no explicit instructions or delimiters defined to separate user input from agent instructions or to warn the agent against executing instructions contained within the schema descriptions.\n- Capability inventory: The skill is configured with 'write' mode and is tasked with writing migration files and updating existing data access paths across the repository.\n- Sanitization: The instructions do not specify any validation or sanitization logic for the inputs before they are used to generate code.
Audit Metadata