agent-native-reviewer

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of markdown-based architectural guidelines and does not include any executable scripts or requests for privileged tools.
  • [SAFE]: No indicators of prompt injection, obfuscation, or data exfiltration attempts were found within the provided content.
  • [SAFE]: The instructions for reviewing source code and pull requests introduce an indirect prompt injection surface; however, the lack of active tools or capabilities within the skill renders this surface non-exploitable.
  • [SAFE]: The principles discussed, such as 'Dynamic Context Injection,' refer to recommended application design patterns for context management and are not related to shell command injection vulnerabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 01:20 AM