contract-review

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow reads outsider-authored free text from the contract .docx itself (parsed via pandoc in scripts/contract_analyzer.py:extract_text() and via word/document.xml in scripts/document.py), which then gets embedded into the generated LLM-context text fields (summary/opinion/flowchart/comment bodies) that the workflow uses downstream.