skills/mizchi/skills/sql-security/Gen Agent Trust Hub

sql-security

Pass

Audited by Gen Agent Trust Hub on Jun 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a local source code scanner (sql-injection-scan.mjs) using standard Node.js file system modules (node:fs). It performs read-only operations to identify patterns matching SQL injection risks.
  • [SAFE]: The tool includes a deliberate opt-out mechanism (// sql-security: ok) and clearly documents its limitations, such as its inability to detect multi-line template literals, ensuring users are aware of the scanner's scope.
  • [SAFE]: Recommendations for external tools like secretlint refer to well-known security utilities and follow standard package management practices for development environments.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 29, 2026, 06:04 AM
Security Audit — agent-trust-hub — sql-security