document-approval-tracker-scott-margetts

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow ingests user-provided email content (including forwarded chains) in Mode 2/3 when the operating user pastes it into the chat; that pasted text is outsider-authored free text from other reviewers/clients and becomes LLM-readable context for classification and output.