using-glean-productivity

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow uses search/meeting_lookup/user_activity to assemble a personal view, and those tools can surface outsider-authored free text (e.g., other people’s mentions/comments/emails/meeting notes) into the agent’s LLM context for summarization, creating an indirect prompt-injection risk.