Requirements Engineering

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime workflow ingests the user-provided input text (e.g., “Informal brief, meeting notes, customer text, or system spec”) into the agent’s LLM context; if that input contains outsider-authored free text (e.g., customer messages, meeting transcripts from others), it becomes LLM-readable prose and can enable indirect prompt injection.