skill-authoring
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a meta-instruction set for authoring other AI skills. It outlines a workflow for researching design problems, extracting observation axes, and citing primary references.
- [INDIRECT_PROMPT_INJECTION]: The skill encourages the agent to research and open external URLs to gather design information. While this technically introduces a surface for indirect prompt injection from the content of those websites, the skill provides mitigating instructions, such as 'Extract repeated rules; do not clone one reference' and 'Verify the delivered result,' which instruct the agent to interpret rather than blindly follow content. Given this is the primary purpose of the skill, it is considered safe.
- [EXTERNAL_DOWNLOADS]: The files contain several URLs to design resources (e.g., Stanford University, Design Council) and GitHub repositories. These are documented as references for design methodology and do not involve the execution of remote code or the installation of untrusted software.
Audit Metadata