account-tier-scoring

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow ingests CRM/contact data, meeting/chat/email-like interactions, and LinkedIn engagement feed event context (from third-party/outsider-authored sources) into the signal stack for scoring and alert text, enabling indirect prompt-injection via fields like chat/support questions, outreach replies, and meeting transcripts.