create-fto-screening-report-ip

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

• Third-party content exposure detected (medium risk: 0.30). In scripts/run_generic_fto_report.py Step 1, the agent parses outsider-supplied DOCX text from --input (Document(path) plus paragraph/table cell extraction) and later uses extracted strings to populate queries, claims-evidence fields, and the generated HTML/DOCX report.