菜单
Budget how much autonomous preparation is acceptable based on risk level, missing evidence, human-only actions and approval needs.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Govern AI use cases as a portfolio with business owner, data owner, model risk tier, evaluation baseline, human review and rollback path.
Map which CIO actions can be advised, drafted, prepared for approval or must remain human-only.
Protect architecture runway by detecting capacity, integration, platform, technical-debt and exception risks before approval.
Predict likely audit findings from missing evidence, weak controls, unclear ownership, exceptions and compliance signals.
Prepare action plans, task drafts, escalations, approval requests, notifications, workflow steps and decision logs from The Autonomous CIO findings. Does not execute external actions.
Run the flagship governed CIO autopilot workflow. Use for broad CIO reviews, Board in 48h scenarios, "run the CIO", mixed enterprise context, monthly operating reviews, crisis-to-control reviews and executive decision-readiness work.
| name | autonomy-risk-budget |
| description | Budget how much autonomous preparation is acceptable based on risk level, missing evidence, human-only actions and approval needs. |
Keep governed autonomy bounded by risk, evidence and human-control requirements.
Accept action ledgers, autonomy gates, risk levels, missing evidence, decision packets and operating reviews.
This limits autonomy; it does not authorize external execution.
