database
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes external database schemas and migration files, which serves as an entry point for potential indirect prompt injection attacks. \n- Ingestion points: Processes external schema or migration files as specified in the argument hint in SKILL.md. \n- Boundary markers: The skill does not define explicit delimiters or instructions to the agent to distinguish between data and embedded commands within the analyzed files. \n- Capability inventory: The skill limits the agent's environment to the Read, Grep, and Glob tools, reducing the risk of unauthorized actions. \n- Sanitization: No validation or sanitization logic is provided for the content of the ingested files.
Audit Metadata