dual-coding-designer
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed to provide teachers with visual design strategies based on cognitive science principles. It does not use any tools or possess capabilities to interact with the underlying system or external networks.
- [SAFE]: Analysis of the prompt and metadata found no evidence of prompt injection, data exfiltration, or obfuscation. The skill operates by processing educational concepts into structured text recommendations.
- [SAFE]: No credentials, sensitive file paths, or remote code execution patterns were detected. All external references are limited to academic citations and theoretical frameworks.
- [SAFE]: Regarding indirect prompt injection (Category 8), the skill ingests untrusted text via the 'verbal_content' field; however, the agent lacks any capabilities (subprocess, file-write, network) to be exploited, and the input is bounded by labels in the prompt.
Audit Metadata