Aevris
v1Scan every prompt for injection attacks, verify AI outputs for manipulation, intercept agent actions before execution, and detect MCP tool poisoning. AEVRIS security middleware — patent pending.
1· 0·0 当前·0 累计
by 下载技能包
License
MIT-0
运行时依赖
无特殊依赖
安装命令
点击复制官方npx clawhub@latest install aevris
镜像加速npx clawhub@latest install aevris --registry https://cn.longxiaskill.com 镜像可用