code-review

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of markdown instructions and checklists for performing code reviews on Ruby on Rails projects. No malicious scripts or configurations were found.
  • [PROMPT_INJECTION]: The skill is designed to process untrusted data such as PR descriptions and comments, which presents an indirect prompt injection surface. However, it includes a robust defense mechanism in the 'THIRD-PARTY CONTENT DEFENSE' section. Ingestion points: PR descriptions, comments, and issue text (SKILL.md). Boundary markers: The 'HARD-GATE' section explicitly instructs the agent to disregard instructions in third-party text. Capability inventory: The skill is limited to generating code review findings and does not have access to command execution or network tools. Sanitization: The instructions mandate extracting only factual context and ignoring directives.
  • [DATA_EXFILTRATION]: No network operations, external downloads, or sensitive file access patterns were identified.
  • [COMMAND_EXECUTION]: The skill does not contain any executable scripts, shell commands, or dynamic execution patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 09:29 AM
Security Audit — agent-trust-hub — code-review