ui-visual-validator
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- [SAFE] (SAFE): No malicious code patterns, obfuscation, or unauthorized access attempts were detected in the skill instructions.
- [NO_CODE] (LOW): This skill consists entirely of markdown persona instructions and does not include scripts, binaries, or configuration files that execute code.
- [Indirect Prompt Injection] (LOW): The skill defines a surface for processing untrusted external data (screenshots and UI descriptions). 1. Ingestion points: Visual evidence provided for analysis. 2. Boundary markers: Absent. 3. Capability inventory: The skill is restricted to agent reasoning and does not provide file system, network, or subprocess capabilities. 4. Sanitization: Absent. The risk is limited to internal influence on agent reasoning without external side effects.
Audit Metadata