code-review

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a standard methodology for reviewing .NET code. It utilizes established Roslyn-based MCP tools (detect_antipatterns, get_diagnostics, find_references) to identify common coding errors like async pitfalls and dependency violations. All identified patterns reflect industry best practices for security (checking for [Authorize] tags) and performance (detecting N+1 queries). No evidence of prompt injection, data exfiltration, or unauthorized code execution was found. The instructions prioritize security-critical areas and provide clear remediation guidance for developers.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 03:50 PM
Security Audit — agent-trust-hub — code-review