automate-me

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

• Third-party content exposure detected (medium risk: 0.30). [Skill.md] requires gathering evidence from “user-provided examples… explicitly supplied… transcript references” and “the visible current conversation,” which are outsider-authored free text the workflow ingests before it selects specific, trusted items (it also mines first-class conversation history and repository conventions enforced by the user).