response-letter

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

• Third-party content exposure detected (medium risk: 0.65). The required runtime workflow ingests the user-provided “Reviewer comments” and “Manuscript” text into the agent’s LLM context; if those comments/manuscript excerpts originate from outside parties (e.g., peer reviewers), that is outsider-authored free text entering the model context (indirect prompt-injection risk).