postgres-expert

Pass

Audited by Gen Agent Trust Hub on Jun 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides database development guidelines and SQL examples. No malicious patterns, obfuscation, or unauthorized data access were detected.
  • [PROMPT_INJECTION]: The skill handles user-provided SQL and schema data, creating a surface for indirect prompt injection. This is addressed by instructions to use parameterized queries.
  • Ingestion points: User-provided SQL queries and database schemas (SKILL.md).
  • Boundary markers: Absent.
  • Capability inventory: Generates SQL queries and transaction logic for database interaction (SKILL.md).
  • Sanitization: Instructions mandate 'Always use parameterized queries'.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 29, 2026, 07:17 AM
Security Audit — agent-trust-hub — postgres-expert