メニュー
Proactively identifying failure modes, misuse, and unintended consequences.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | harm-anticipation |
| description | Proactively identifying failure modes, misuse, and unintended consequences. |
Harm anticipation is the practice of systematically thinking through how an AI product could cause harm — before it does. It's preventive design, not reactive crisis management.
Work through each harm category systematically:
Think like an adversary:
Think about second-order effects:
扮演设计师角色，完成从用户研究、UX策略、UI设计到交互设计的全流程设计任务。支持设计研究、设计系统、原型测试、设计运营等多维度职责。
Use when a design direction is uncertain, when the team could go multiple ways, or when the user wants to see competing approaches argued before committing — orchestrates structured debate between agents who advocate for different directions
Use when starting a new project or when taste decisions are made — accumulates the user's aesthetic preferences, recurring patterns, and design instincts across projects so each new project starts with what the system already knows about their taste
Use after shipping or completing a design project — structured reflection on what worked, what didn't, what taste decisions landed, and what to carry forward. Feeds learnings back into design-memory so the next project is sharper
Coordinating text, image, voice, and tool-use modalities in a single interaction.
Managing shared context, memory, and state across multiple agents.
