amendment-history

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

• Third-party content exposure detected (high risk: 0.85). The skill ingests the full text of user-provided contract/amendment documents (base agreement and amendments) via file upload/repository/CLM integration, and those documents are outsider-authored free text unless the operating user authored them; the LLM then reads and quotes them in Step 2 and the Mode 1/2 outputs.