Kastra

Launching today 今日发布

Runtime authorization for Claude, Cursor, Codex and OpenClaw 为 Claude、Cursor、Codex 和 OpenClaw 提供运行时授权

Kastra is the runtime authorization layer for AI agents. It decides what agents can and cannot do before actions execute, enforcing policies with sub-1 ms latency across tools, prompts, inputs, and outputs. Kastra 是 AI 智能体的运行时授权层。它在操作执行前决定智能体可以做什么、不可以做什么，并在工具、提示词、输入和输出层面以低于 1 毫秒的延迟强制执行策略。

Use one control plane to govern agents and policies across Claude Code, Cursor, Codex, OpenClaw, the Anthropic SDK, the OpenAI SDK, and more. 通过统一的控制面板，管理 Claude Code、Cursor、Codex、OpenClaw、Anthropic SDK、OpenAI SDK 等多种环境下的智能体与策略。

Prevent unauthorized tool use, prompt injection, and exposure of sensitive data before they become incidents. Trust the rules, not the agents. 在未经授权的工具使用、提示词注入和敏感数据泄露演变为安全事件之前，将其拦截。信任规则，而非智能体。