The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the skill. The instructions focus on routing frontend development tasks to specific design style guides to avoid generic output.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns or unsafe scripts were found. The skill suggests using industry-standard packages such as Framer Motion and GSAP, but emphasizes checking existing dependencies and prints installation commands rather than auto-executing them.
[DATA_EXFILTRATION]: No data exposure or exfiltration risks detected. The skill uses standard placeholder image services (picsum.photos) and does not access sensitive local files or environment variables.
[PROMPT_INJECTION]: No attempts to override agent safety guidelines or extract system prompts were found. The skill uses strong directives to enforce design quality, which are benign in context.

taste-metaskill