# OpenAI Agents JS — TypeScript Multi-Agent SDK

> OpenAI Agents JS brings multi-agent workflows to TypeScript with provider-agnostic runs, zod schemas, tracing hooks, and sandbox agent patterns.

## Install

Save the content below to `.claude/skills/` or append to your `CLAUDE.md`:

## Quick Use

1. Install the verified package or repo entrypoint:

```bash
npm install @openai/agents zod
```

2. Run the first local check:

```bash
node ./agent.js
```

3. Add a repeatable verification command:

```bash
npm test
```

---

## Intro

OpenAI Agents JS is a verified GitHub-backed tool for modern AI and developer workflows, sourced from `openai/openai-agents-js` with 3,000 stars and a MIT license snapshot. Best for: TypeScript teams that want agent orchestration, structured outputs, and tool calls without moving the app into Python. Works with: Node.js, Bun, Deno, zod schemas, OpenAI APIs, sandbox agents, tracing workflows. Setup time: 12 minutes. Use it when you need a concrete, repeatable path rather than another one-off shell snippet.

---

## Operating Pattern

### Fit check

| Question | Practical answer |
|---|---|
| What do you install? | `@openai/agents` from `openai/openai-agents-js` |
| What is the first command? | `node ./agent.js` |
| What proves it works? | `npm test` |
| How long should a pilot take? | 12 minutes for a small repo or sandbox |

### Adoption loop

1. Run the tool on a disposable branch or sandbox project.
2. Capture before/after output so reviewers can see the exact effect.
3. Add the smallest CI or local check that prevents regressions.
4. Document owner, upgrade command, and rollback command in the repo.

### Recommended use

Start with one agent and one typed tool. Add handoffs only after logging, schema validation, and retry behavior are visible in development.

### Guardrails

Keep side-effecting tools behind explicit approvals. A TypeScript SDK does not remove the need for permissions around file, deploy, or billing actions.

### Rollout checklist

- Pin the package or release version before using it in CI.
- Keep credentials in environment variables or the platform secret store.
- Add one owner who is responsible for upgrades and breaking-change triage.
- Re-check the GitHub repo before writing docs that mention APIs or install paths.


---

### FAQ

**Q: Is this production-ready?**
A: The repo exists at `https://github.com/openai/openai-agents-js` and has 3,000 GitHub stars. Treat the first rollout as a controlled pilot until your team has tested install, rollback, and CI behavior.

**Q: Why use it instead of a generic script?**
A: The value is repeatability: a named package, a documented command, a source repo, and a small verification path that can be reviewed by teammates.

**Q: What should I measure first?**
A: Measure setup time against the 12 minutes target, count how many files or tasks it changes, and record whether the CI command catches the same issue locally.

---

## Source & Thanks

> Built from [openai/openai-agents-js](https://github.com/openai/openai-agents-js). License: MIT.
>
> GitHub stars verified from `api.github.com/repos/openai/openai-agents-js`: 3,000.

---

<!-- ZH -->

## 快速使用

1. 安装已验证的包或仓库入口：

```bash
npm install @openai/agents zod
```

2. 跑第一次本地检查：

```bash
node ./agent.js
```

3. 加一个可重复验证命令：

```bash
npm test
```

---

## 简介

OpenAI Agents JS 是一个已通过 GitHub 仓库验证的现代 AI / 开发者工具，来源 `openai/openai-agents-js`，当前星标快照 3,000，许可证 MIT。适合：想要 agent orchestration、结构化输出和工具调用，但不想把应用迁到 Python 的 TypeScript 团队。兼容：Node.js、Bun、Deno、zod schema、OpenAI API、sandbox agent、tracing 工作流。装机时间：12 minutes。当你需要的是可重复落地路径，而不是一次性 shell 片段时，用它。

---

## 操作模式

### 适配检查

| 问题 | 实用答案 |
|---|---|
| 安装什么？ | 来自 `openai/openai-agents-js` 的 `@openai/agents` |
| 第一条命令？ | `node ./agent.js` |
| 如何证明可用？ | `npm test` |
| 小范围试点多久？ | 小仓库或 sandbox 约 12 minutes |

### 接入循环

1. 先在一次性分支或 sandbox 项目里跑。
2. 记录 before / after 输出，让 reviewer 看见实际影响。
3. 加最小的 CI 或本地检查，防止同类问题回归。
4. 在仓库文档里写清 owner、升级命令和回滚命令。

### 推荐用法

先从一个 agent 和一个 typed tool 开始。等日志、schema 校验和重试行为在开发环境可见后，再加入 handoff。

### 风险边界

有副作用的工具要放在明确审批后面。TypeScript SDK 不会替你解决文件、部署或计费动作的权限边界。

### 推广检查表

- 进入 CI 前固定包版本或 release 版本。
- 凭据放进环境变量或平台 secret store。
- 指定一个 owner 负责升级和 breaking change 分流。
- 文档里写 API 或安装路径前，重新核验 GitHub 仓库。


---

### FAQ

**Q: 能直接上生产吗？**
A: 仓库已验证存在：`https://github.com/openai/openai-agents-js`，GitHub 星标 3,000。第一次上线仍建议控制试点，先验证安装、回滚和 CI 行为。

**Q: 为什么不用普通脚本？**
A: 价值在可重复：有命名包、文档化命令、源仓库和可被团队 review 的最小验证路径。

**Q: 第一步应该量什么？**
A: 先量装机时间是否接近 12 minutes，它改了多少文件或任务，以及 CI 命令是否能在本地抓到同类问题。

---

## 来源与感谢

> 来源：[openai/openai-agents-js](https://github.com/openai/openai-agents-js)。许可证：MIT。
>
> GitHub stars 已通过 `api.github.com/repos/openai/openai-agents-js` 验证：3,000。


---
Source: https://tokrepo.com/en/workflows/openai-agents-js-typescript-multi-agent-sdk
Author: OpenAI