My agents finally have a shared mailbox. I still have to wake them up myself.
Summary
作者为多个 AI 智能体搭建了一个基于 git 仓库的共享邮箱系统,用于智能体之间的任务交接,但发现唤醒机制仍需人工介入。文章探讨了送达、已读、开始、完成四种状态的追踪,并向社区征求唤醒方案的经验。
Similar Articles
I collected 51 AI agent setups people actually use - what I found.
作者调研了 Reddit、X、博客、Substack 和 GitHub 上 51 个真实在用的 AI agent 工作流,总结出三条可复用的设置(计划先行、双 agent 共享日志、邮件收件箱自动分拣),并归纳出共同模式:任务窄、只起草不执行、交接写入文件、由第二个 agent 审查、监控静默告警。
I've had four long running agents running my newsletter for months. The hard part wasn't memory, it was the context window
作者分享用四个长期运行的 Claude Code 智能体运营自己的 AI 通讯的经验,指出最大瓶颈不是记忆或工具,而是上下文窗口,并介绍了智能体自写检查点、自清空、自我恢复的机制,以及相关的失败案例和开源工具。
Separate inbox for every agent?
作者为每个 AI agent(intake、audit、broker desk、signup 等)分配独立邮箱,甚至为每个任务创建短期的 run-xxxx 邮箱来接收验证码,认为比共用一个 Gmail 更整洁,并询问他人是否有类似的按任务建邮箱的做法。
I gave my AI agent a voice and now I can't live without it!
一位拥有10多年经验的系统架构师分享了自己搭建AI个人助理的体验:通过WhatsApp发送语音笔记即可自动创建日历邀请、发送邮件、执行定时监控任务和自定义技能,极大提升了日程与任务管理效率。
the part of AI agents nobody talks about: what happens when two agents try to use the same email inbox
When multiple AI agents share an email inbox, they can collide on messages like OTPs, causing silent failures. The solution is dedicated per-agent inboxes with isolated read locks and long-polling instead of scheduled polling.