hint-overlay-visual-verification
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted screenshots and UI text which could contain malicious instructions. 1. Ingestion points: The skill accepts Screenshot paths, Required callout text, and Scenario and overlay state as inputs (SKILL.md). 2. Boundary markers: The skill implements strict Hard Gates and a Validation Checklist to ensure output matches expected state (SKILL.md). 3. Capability inventory: The skill utilizes the agent's image, browser, or computer inspection capability to analyze external data (SKILL.md). 4. Sanitization: No explicit mention of text sanitization or filtering for the content discovered within visual assets.
Audit Metadata