如何构建 Claude Agent 团队:一个编写代码,一个审查代码,一个负责部署(内含详细配置)
One Claude session doing everything is like one employee who writes the code, reviews their own pull request, and pushes to production at 3am. Every engineering manager knows why that fails.
一个 Claude 会话完成所有工作,就像一个员工自己编写代码、审查自己的拉取请求,并在凌晨3点推送到生产环境。每个工程经理都知道这为什么会失败。
Boris Cherny does it differently. The creator of Claude Code runs hundreds of agents in parallel from his phone. Each one has a specific job. Each one feeds results to the next. He hasn't written a single line of code in 2026.
Boris Cherny 的做法与众不同。Claude Code 的创造者通过手机并行运行数百个 agent。每个 agent 都有特定的任务。每个 agent 将结果传递给下一个。他在 2026 年还没写过一行代码。
The architecture is not complicated. It is three agents with three separate claude-md files and one pipeline connecting them.
架构并不复杂。它是三个智能体,拥有三个独立的 claude-md 文件,以及一个连接它们的流水线。
This article shows you the exact setup: one agent writes code, one reviews it, one deploys it. Code goes from idea to production with you as the oversight, not the bottleneck.
本文向您展示了具体的设置:一个 agent 编写代码,一个审查代码,一个部署代码。代码从构思到生产环境,由您负责监督,而不是成为瓶颈。
Why One Agent Is Not Enough
为什么一个 Agent 是不够的
When one agent writes and reviews its own code, it misses the same things a human would miss reviewing their own pull request. No second pair of eyes. No separation of concerns.
当单个智能体编写并审查自己的代码时,它会遗漏人类在审查自己的 pull request 时同样会遗漏的问题。没有第二双眼睛把关。没有关注点分离。
When you split across multiple agents, each one operates with a different system prompt and a different job. The writer optimizes for speed. The reviewer optimizes for correctness. The deployer optimizes for safety.
当你在多个智能体之间进行拆分时,每个智能体都使用不同的系统提示词和不同的任务运行。编写者优化速度。审查者优化正确性。部署者优化安全性。
That separation is what turns a chatbot into a team.
正是这种职责分离将聊天机器人变成了一个团队。

The Architecture: Three Agents, Three Jobs
架构:三个智能体,三项任务
Each agent gets its own claude-md file with its own role, constraints, and workflow.
每个智能体都有自己的 claude-md 文件,包含其各自的角色、约束和工作流。
Agent 1 - The Writer
Agent 1 - 编写者
You are a code writer. Your only job is to implement features and fixes.
你是一名代码编写者。你唯一的工作是实现新功能和修复问题。
Write clean, readable code. Follow existing patterns in th...