跳到正文
Diogo Almeida· @CompleteSkeptic · X·· 1 天前AI 评分45
AI 导读

太酷了 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想要的工具调用。 Armor 使用 @typesafeai 的 Jev 模型,在你的智能体执行之前拦截违反策略的工具调用,性能优于 Claude Code 的自动分类器,而成本仅为其一小部分。

正文

SO FKIN COOL

引用OpenRouter@OpenRouter
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.
在 X 查看被引用的帖子

来源:Diogo Almeida · x.com