The Agent Skills Directory

[SAFE]: The skill provides structured design guidelines and steps for UI review without any executable malicious code or suspicious network operations.
[SAFE]: External references, such as the design inspiration site misostack.com and accessibility tools like Lighthouse, are used legitimately for design and auditing.
[SAFE]: The instruction to use a local tool (tokens.py) for contrast checking is a standard part of the automated design workflow.
[SAFE]: The skill incorporates security best practices by explicitly referencing an internal application security gate for any code produced.
[SAFE]: The skill processes untrusted frontend code or design descriptions for review (Indirect Prompt Injection surface); however, its focus on structural analysis and the inclusion of a mandatory security gate mitigate the risk of executing embedded malicious instructions.

frontend-design