Deny-by-default governance for every tool call — identity, scope, rate limiting, injection detection, audit logging, plus opt-in output DLP, escalation, and behavioral monitoring. Hooks into OpenClaw at the process level so the agent can't bypass it.
每项工具调用默认拒绝访问,涵盖身份认证、权限范围、速率限制、注入检测和审计日志,并支持可选的输出数据防泄漏、权限升级和行为监控。通过进程级集成OpenClaw,确保代理无法绕过该机制
直接复制以下提示词,发送给你的 AI 助手即可完成安装。
点击右上角 下载SKILL 按钮