legal-incident-response

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

• Third-party content exposure detected (high risk: 0.85). Outsider free text can enter the LLM context via the authority-letter path’s incoming_letter_text/incoming_letter_summary inputs (Step 3 EXTRACT and Step 4 DRAFT in protocols/authority-letter.md), where the pasted letter content is authored by an external regulator and is then read by the LLM.