HN 提问：代理沙箱还缺少什么？
1 分•作者: oryx1729•3 个月前
我正在使用 OpenCode SDK 构建一个代理沙盒平台，该平台提供具有特定配置的可定制模板，用于通用代理。
我的想法是提供出色的命令行体验，以便轻松使用 Claude Code 或 Codex 等 AI 客户端来启动沙盒、获取/检索数据和提示、审查生成的输出以及管理用于提问用户问题的队列。
我还计划添加一个 AI 网关层，类似于 Tailscale Aperture（非官方合作，但其中一位工程师在 AI Engineer 的演讲非常棒，可以在 YouTube 上找到），用于监控网络访问、分发密钥以及控制沙盒可以访问的服务/系统。
对于正在构建代理工作流的人们，如果您尝试过其他沙盒解决方案，我很想听听您的反馈。
缺少了什么？最困难的部分是什么？
查看原文
I'm building an agent sandbox platform with opinionated customizable templates for general purpose agents using the OpenCode SDK.<p>The idea is to provide a great CLI experience so it's easy to use an AI client like Claude Code or Codex to launch sandboxes, fetch/retrieve data and prompts, review generated outputs, and manage queues for ask-user questions.<p>I also plan to add an AI gateway layer, similar to Tailscale Aperture(not affiliated but one of their engineers gave a great talk at AI Engineer it's on YT) for monitoring network access, distributing secrets, and controlling what services/systems a sandbox has access to.<p>For people building agent workflows, I'd love to hear feedback if you've tried other sandbox solutions.<p>What was missing and what was the hard part?