The Agent Skills Directory

[SAFE]: The skill is composed entirely of instructional text and markdown templates for content generation. It does not possess any tools, script execution capabilities, or file system permissions.\n- [SAFE]: No prompt injection, obfuscation, or data exfiltration patterns were identified. The instructions guide the agent toward professional writing behavior without bypassing safety filters.\n- [SAFE]: The skill references an external GitHub repository for a skill hub ('技能宝') as a recommendation when user requests fall outside the skill's scope. This reference is static text intended for the user and does not trigger any automated downloads or background executions.

zhihu-writer