The Agent Skills Directory

[SAFE]: No malicious patterns or security risks detected. The skill instructions are focused on journalistic integrity, statistical accuracy (effect sizes, sample demographics), and adherence to the Protection of Children and Youths Welfare and Rights Act (兒少法 §69). All references point to internal local documentation.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external, untrusted materials such as research abstracts and interview transcripts. While this creates a potential attack surface for indirect prompt injection, the skill lacks any dangerous capabilities such as shell execution, filesystem modification, or network operations. Furthermore, it incorporates specific 'Risk Check' and 'Research Evidence Audit' steps that serve as content validation frameworks.

med-education