data-layer-handbook
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as an architectural guide for database management. All identified external resources (GitHub and npm links) belong to the skill's author (bh611627) and represent standard project documentation and package distribution.
- [INDIRECT_PROMPT_INJECTION]: The skill contains instructions to analyze user-provided code for SQL injection vulnerabilities.
- Ingestion points: Untrusted SQL queries or code provided by the user (SKILL.md).
- Boundary markers: None explicitly defined in instructions.
- Capability inventory: Limited to read-only tool access and natural language advice; no file-write or network-write capabilities.
- Sanitization: Not applicable for this advisory use case.
- Given the
read-onlytool restriction, this surface poses no risk to the agent or user environment.
Audit Metadata