visual-testing-advanced

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements visual testing logic using well-known industry tools. No malicious prompt instructions, data exfiltration patterns, or unauthorized command executions were found.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it visits and processes content from external websites.
  • Ingestion points: Uses page.goto() to navigate to target URLs in SKILL.md.
  • Boundary markers: Not explicitly defined for external web content.
  • Capability inventory: Performs visual diffing and captures screenshots using Playwright and Percy.
  • Sanitization: Implements masking for dynamic elements like timestamps and advertisements to improve consistency and reduce noise.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 09:20 AM