NYC

postgresql-expert

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • SAFE (SAFE): No security issues detected across the 10 threat categories. The provided files consist of markdown documentation and SQL boilerplates.
  • Prompt Injection: No markers or bypass attempts found; instructional language is standard for technical documentation.
  • Data Exfiltration: No hardcoded credentials or network-based exfiltration patterns detected.
  • Dependencies/RCE: No package managers, remote script downloads, or dynamic execution patterns present.
  • Indirect Prompt Injection (Surface Analysis): While the SQL templates utilize placeholders (e.g., {UP_SQL}, {INSERT_LOGIC}), there are no associated execution scripts within the skill that would perform unsafe interpolation of untrusted data.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:15 PM