跳到正文
OpenRouter· @OpenRouter · X·· 1 天前AI 评分47
AI 导读

今天,我们推出 OpenRouter Armor 测试版。 工具调用让智能体能够与外部世界交互,而基于正则的过滤往往不足以检测不受欢迎的工具调用。 Armor 使用 @typesafeai 的 Jev 模型,在你的智能体执行越界工具调用之前进行拦截,性能优于 Claude Code 的自动分类器,而成本仅为后者的一小部分。

正文

Today, we are launching OpenRouter Armor in beta.

Tool calls allow agents to interact with the world, and regex-based filtering often isn't enough to detect unwanted tool calls.

Armor uses @typesafeai's Jev model to guard against out of policy tool calls before your agent executes them, with better performance than Claude Code's auto classifier at a fraction of the cost.

来源:OpenRouter · x.com