postgres-agents-query-reviewer

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a rigorous, manual-style workflow for code review that enhances security by specifically checking for data-loss risks (C-2: missing WHERE clause) and silent correctness bugs (C-10: NOT IN with NULLs).
  • [SAFE]: All external references point to official PostgreSQL documentation and established technical standards for database auditing.
  • [SAFE]: The execution logic is entirely instructional, directing the AI agent to generate analytical reports rather than executing arbitrary shell commands or making unauthorized network requests.
  • [SAFE]: The identified 'sibling skills' (e.g., postgres-syntax-dml-ddl) represent a modular organizational structure common in development environments and do not involve remote code downloads or untrusted dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 01:10 AM
Security Audit — agent-trust-hub — postgres-agents-query-reviewer