AI Agent 安全設計模式 — 基於 Claude Code 七層縱深防禦架構的逆向分析。Use when: 設計工具安全機制、實作命令過濾、建構權限模型、設計 sandbox。
$npx -y skills add cablate/ai-toolkit --skill cc-security-patternsInstalls into the current project.
Run `npx skills use "https://github.com/cablate/ai-toolkit" --skill "cablate/ai-toolkit/cc-security-patterns"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.
Use the skills in "https://github.com/cablate/ai-toolkit" that are relevant to the current task. Run `npx skills add "https://github.com/cablate/ai-toolkit"` and select the relevant skills, then follow their instructions.