decide

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow (see SKILL.md Phase 1 / Round 1 and the templates in references/debate-prompts.md — e.g., Advocate A/B and Moderator "RESEARCH REQUIREMENT" and "COUNTER-EVIDENCE RESEARCH MANDATE") explicitly instructs agents to research externally and cite at least two sources (URLs), meaning the agents will fetch and interpret open/public third‑party content that can materially change their arguments and actions.