visual-verify

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of purely instructional content guiding the agent on how to verify UI changes. It does not contain any executable scripts, shell commands, or external network requests.
  • [PROMPT_INJECTION]: No attempts to bypass safety filters or override system instructions were detected.
  • [DATA_EXFILTRATION]: No evidence of hardcoded credentials, sensitive file access, or unauthorized network communication.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 08:39 PM
Security Audit — agent-trust-hub — visual-verify