growing-nrr

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

• Third-party content exposure detected (high risk: 0.90). The workflow.md (Phase 3: Customer Feedback Synthesis) explicitly requires ingesting and AI-synthesizing public, user-generated sources — e.g., G2/Capterra review sites, social media mentions, community posts/Slack/Discord, and other external signals — and uses those signals to prioritize actions, routing, and interventions, so untrusted third‑party content can directly influence agent decisions.