postgresql-admin
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: LOW
Full Analysis
- SAFE (SAFE): Comprehensive analysis of the instructions, documentation, and validation scripts revealed no security issues.
- PROMPT_INJECTION (SAFE): Instructions are strictly focused on database administration. No override/bypass markers or jailbreak patterns were detected.
- DATA_EXFILTRATION (SAFE): No unauthorized network calls or sensitive file access patterns. SQL examples provided are restricted to administrative tasks and use placeholders.
- INDIRECT_PROMPT_INJECTION (LOW): The skill has an attack surface for handling database identifiers; however, this is mitigated by strict parameter validation (
^[a-z][a-z0-9_]*$) and enum-based operation selection. - DYNAMIC_EXECUTION (SAFE): The provided Python script (
scripts/validate.py) usesyaml.safe_load()and performs standard file existence checks without executing arbitrary strings or untrusted code.
Audit Metadata