paper-self-revise

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

• Third-party content exposure detected (high risk: 0.85). Outsider-authored free text can enter the LLM context when the skill reads review materials from the provided project folder (e.g., review*.txt/.md or extracted text from review.pdf/annotated*.tex/.pdf) and then displays each extracted review comment and the referenced manuscript passages to the user/LLM during Phase 1–3.