phronesis

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 the M7 extraction pipeline’s Nous gate, which may call the external bookkeeping module/LLM on candidate.content and candidate.quote derived from journal payloads that originate from sponsor/interview transcripts (e.g., INTERVIEW_LOGGED.key_findings / interviewee), i.e., outsider-authored text.