video-caption-creation

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The skill's Sub-Agent Prompt Template explicitly requires reading SOURCE.md (the "full transcript for context") and EDITOR_HANDOFF.md and then generating hooks for each clip, which means it ingests user/third-party transcript content (untrusted, user-generated) as part of its required workflow and that content directly drives decisions about which hooks to produce.