The Agent Skills Directory

[SAFE]: The skill consists exclusively of Markdown files providing educational content and operational guidelines for the AI agent. There are no executable scripts (Python, JavaScript, shell) or binaries included.
[SAFE]: The skill instructions include a routine practice of reading local project metadata (e.g., product-marketing.md) to provide better context. This does not involve accessing sensitive system files or making outbound network connections.
[SAFE]: Analysis of all 10 threat categories, including prompt injection, obfuscation, and data exfiltration, yielded no positive detections. The skill follows safe instructional practices.

offers