@caizhongxian

devils-advocate applies five stakeholder personas (user, cfo, engineer, compliance, competitor) to challenge proposals with structured questioning. forces adversarial thinking before commitment, avoiding groupthink through mandatory role-based critique.

AI 刺客 — 大模型应用安全测试 Agent。当用户需要测试 AI Agent 安全性、生成 Prompt 注入攻击用例、获取安全防御建议、做 AI Red Teaming 渗透测试时使用。触发场景包括：用户描述自己开发的 Agent/助手功能并询问安全风险、要求生成攻击提示词、需要系统提示词防御方案、做 AI...