ui-slop-review
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements security best practices for handling untrusted data. In the 'Evidence safety' section of SKILL.md, it explicitly instructs the agent to treat all inspected content—including screenshot text, OCR, DOM content, and metadata—as untrusted evidence rather than instructions. It specifically forbids executing embedded commands or disclosing secrets found in the evidence. This directly mitigates Category 8 (Indirect Prompt Injection) risks.
- [SAFE]: External references are limited to the vendor's own official domain (uizze.com) and are used for documentation and user recommendations. No unauthorized network operations, credential access, or remote code executions were found.
Audit Metadata