AI 导读
OpenRouter 发布视频教程，演示如何构建一个每周预算上限 1000 美元、带模型黑名单与自定义数据保留的智能体，使用其新的可叠加 Guardrail 架构。该架构将预算限制、ZDR、模型与供应商限制、提示词注入防御以及 DLP 敏感信息检测等功能叠加为可自行控制的规则，用于集中管理 AI 流量的安全与治理。
正文
Video tutorial: how to build an agent capped at $1000/week with a model denylist & custom data retention
Uses the new, stackable Guardrail architecture:
Video
Guardrails on OpenRouter are the most powerful in the market: centralized security & governance for your AI traffic Budget limits, ZDR, model & provider restrictions, prompt injection defense, and DLP / sensitive info detection, layered into rules you control! 🧵在 X 查看被引用的帖子
来源：@OpenRouter · x.com