review-design

Pass

Audited by Gen Agent Trust Hub on Jul 5, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: No malicious instruction overrides, safety bypass attempts, or DAN-style injections were detected. The skill uses standard instructional language focused on UI/UX review.
  • [DATA_EXFILTRATION]: No network operations, hardcoded credentials, or attempts to access sensitive system files (e.g., .ssh, .aws) were found. The skill's access is limited to project documentation and code diffs for analysis.
  • [REMOTE_CODE_EXECUTION]: No remote script downloads, package installations, or dynamic code execution patterns are present.
  • [COMMAND_EXECUTION]: The skill does not perform any shell command execution or system-level operations. It is explicitly restricted to providing findings only and is forbidden from editing or refactoring code.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect injection as it processes untrusted data from project files.
  • Ingestion points: Reads project documentation (e.g., docs/frontend/DESIGN.md) and code diffs provided by the caller.
  • Boundary markers: None identified.
  • Capability inventory: The skill has no capabilities to write to the file system, execute code, or perform network requests.
  • Sanitization: None identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 5, 2026, 10:00 PM
Security Audit — agent-trust-hub — review-design