AI Safety Policy Review

ПродвинутыйsecurityМинимальный контекст: 16K

Converts product safety requirements into practical agent and LLM policies with abuse cases, escalation paths, safe defaults, audit events, and acceptance criteria.

ai-safetygovernancepolicyrisk-management

Варианты использования

  • AI feature governance
  • Safety requirement reviews
  • Agent permission design

Пример промпта

Review this AI feature policy. Identify abuse cases, define safe defaults and escalation paths, list audit events, and write testable acceptance criteria.

Рекомендуемые модели

claude-opus-4-6gpt-5gemini-2-5-pro

Совместимые инструменты

claude-codecursorkiroany

Модальности

Вход: text, file
→
Выход: text

Похожие Skills

threat-modelingprivacy-compliance-auditor

Автор

OpenModels Community

@openmodelsrun
AI Safety Policy Review — AI Agent Skill | OpenModels