The Agent Skills Directory

[SAFE]: The skill implements a strong security posture by explicitly instructing the agent to strip credentials, API keys, tokens, and secrets from all outputs and processed data across multiple instruction files.
[SAFE]: A mandatory confirmation policy is established, requiring the agent to obtain explicit user consent before proceeding with image generation, preventing unintended or automated resource usage.
[NO_CODE]: The skill consists exclusively of markdown instruction sets, layout definitions, and style galleries. No executable scripts (e.g., Python, JavaScript, or shell) are included in the skill package, significantly minimizing the risk of direct malicious code execution.
[SAFE]: Preference management is restricted to the skill's own configuration directories (project-local or home-based config), following standard practices for non-privileged configuration storage.

baoyu-infographic