visual-verify

Pass

Audited by Gen Agent Trust Hub on Sep 26, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill provides legitimate instructions for UI verification, emphasizing visual inspection and objective measurement. No malicious behaviors, data exfiltration, or unauthorized persistence mechanisms were identified.
  • [COMMAND_EXECUTION]: The script scripts/selftest.py uses subprocess.run to execute the local compare.py utility as part of its unit tests. This execution is confined to a controlled test environment using synthetic image fixtures and follows secure coding practices, posing no risk to the host system.
  • [SAFE]: The core functionality of the skill involves local image processing using the established and trusted Pillow library. It performs deterministic transformations and comparisons without incorporating external or untrusted code execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 26, 2026, 06:23 AM