The Agent Skills Directory

[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it ingests and processes untrusted text from external knowledge bases.
Ingestion points: Document titles and content are retrieved via the yuque_get_toc and yuque_get_doc tools in the workflow (SKILL.md).
Boundary markers: The skill instructions do not include delimiters or specific safety instructions to isolate ingested content from the system logic.
Capability inventory: The skill utilizes yuque_create_doc to write new content and yuque_group_members to access team lists, providing a path for potential misuse if instructions are injected.
Sanitization: There is no evidence of sanitization or intent-validation for the data read from the Yuque repositories before it is summarized and published.

yuque-group-onboarding-guide