recoup-deal-ingest

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

• Third-party content exposure detected (medium risk: 0.65). Outsider free text can enter the LLM context via scripts/dataroom-hygiene-scan.py, which reads seller-provided textual files under deals/{deal-id}/source/ (e.g., .txt/.md/.csv/.json/.yaml) and extracts matching concealment-language lines into workpapers/dataroom-hygiene.json / findings/dataroom-hygiene-findings.json, which the agent would then summarize for the LLM.