reflect

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md’s required workflow builds a session evidence package from the “first-class current-conversation or transcript resource” / “visible conversation context plus tool results” (and may inline user-provided goals/constraints into reviewer prompts), meaning outsider-authored text from a user’s conversation is ingested at runtime.