create-patent-based-rd-briefing-rd

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow reads user-supplied text from the author-provided authorized Excel workbook (via scripts/tag_relevant.py and scripts/generate_report.py using pandas.read_excel and openpyxl), and those cell values are used to construct the HTML (with escaping), so outsider-authored free text in workbook fields is ingested at runtime without any “select a specific item” fetch step.