check-originality

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow can fetch open-access sources at runtime via fetch-paper and then passes the fetched third-party paper text as readable files into scripts/overlap_check.py --against ..., which is outsider-authored free text ingested into the LLM context indirectly through the agent’s tool outputs/reports.