gpt-slide-design
Pass
Audited by Gen Agent Trust Hub on Apr 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional and does not execute code, access the network, or manage sensitive credentials. It focuses on converting visual design elements from images and PDFs into a structured markdown specification.
- [SAFE]: The instructions include explicit 'anti-hallucination' and 'content-filtering' rules, directing the agent to ignore textual content and focus solely on design language. This naturally reduces the risk of indirect prompt injection from processed documents.
- [SAFE]: No obfuscation, persistence mechanisms, or privilege escalation patterns were detected.
Audit Metadata