The Agent Skills Directory

[SAFE]: The skill is composed of natural language instructions that define aesthetic preferences and design constraints for image generation tasks. No code execution or sensitive data access patterns were identified. The instructions are focused on visual hierarchy, platform-specific design patterns, and maintaining visual consistency across generated assets.
[EXTERNAL_DOWNLOADS]: The metadata in MIRROR.md identifies https://github.com/Leonxlnx/taste-skill.git as the origin of the skill. This is a static reference for documentation purposes and does not represent a runtime security risk.
[NO_CODE]: The skill instructions explicitly direct the agent to generate images only and avoid generating code, minimizing the attack surface for code-based injections or malicious script generation.

taste-imagegen-frontend-mobile