microsoft-outlook

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

• Third-party content exposure detected (medium risk: 0.30). Outsider-authored free text is ingested when the agent calls the mail read/search tools (e.g., listMessages/getMessage) against the user’s mailbox via Microsoft Graph, where message subjects/bodies (including attacker-written content inside emails) are returned as LLM-visible body/content strings.