OpenRouter 推出大规模检查智能体行为的分类功能
OpenRouter 出了个给智能体行为做把关的工具，能在危险 tool call 执行前拦住，跑评测和客服场景的可以看看。
OpenRouter 发布了一个用于检查智能体（Agent）行为的分类功能，可在有风险的 tool call 运行前进行拦截（gate）。
它还能标记需要人工接手的客服对话，并批量标注 eval traces。输入支持纯文本，也支持 system、user、assistant 和 tool 消息组成的完整 span，且 span 大小没有上限。
Built for checking agent behavior at scale: - Gate a risky tool call before it runs - Flag support conversations that need a human - Label eval traces in bulk Send plain text or the full span of system, user, assistant, and tool messages. There's no cap on span size.📊 1 ⚡
This report is categorized under AITOP's "Products" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include OpenRouter, agents, MCP and tools, coding assistants.
Source attribution: curated from OpenRouter (@OpenRouterAI).
