llm-council

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

• Third-party content exposure detected (high risk: 0.90). This skill queries an external AI Gateway (https://ai-gateway.happycapy.ai) via scripts/ai_gateway.py and scripts/server.py and ingests those untrusted model responses (displayed in the dashboard and passed into synthesize_responses, anonymous_model_vote, and chairman_review), meaning third‑party content can directly influence subsequent model prompts and decisions.