#7171st of 153 in Agent frameworks and SDKs
openai-agents-js
TypeScript SDK for multi-agent workflows with handoffs, guardrails and tracing
- Stars
- 4k
- License
- MIT
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- TypeScript
2,799 on 25 Apr 20264,029 on 14 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
OpenAI is hiring: 836 open roles
Overview
OpenAI's TypeScript SDK for building agents in code. It covers text agents, sandbox agents with a filesystem workspace, and realtime voice agents, with tools (functions, MCP, hosted), handoffs, guardrails, human-in-the-loop, sessions and built-in tracing. It is described as provider-agnostic and runs on Node.js 22+, Deno and Bun.
Who it is for: TypeScript developers building multi-agent or voice agent apps
Strengths
- Built-in tracing, guardrails, sessions and human-in-the-loop without extra libraries
- Realtime agents connect browser microphone and audio output via WebRTC
- Sandbox agents can use local Unix, Docker or hosted sandbox clients
- Runs on Node.js 22+, Deno and Bun; MIT licensed
Weaknesses
- Cloudflare Workers support is experimental and needs nodejs_compat
- Pull requests accepted only from repository collaborators
- UnixLocalSandboxClient works only on macOS and Linux; Windows needs Docker or hosted
- Examples and defaults center on OpenAI APIs and an OPENAI_API_KEY
What it needs
- no GPU
- Needs Node.js 22+, zod, OPENAI_API_KEY
- Models: OpenAI APIs, other providers (unspecified)
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in Agent frameworks and SDKs
See all 153| Rank | Project | Score |
|---|---|---|
| 1 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 96 out of 100 |
| 2 | copilotkitReact and Angular SDK for adding agent chat and generative UI to apps | 92 out of 100 |
| 3 | openai-agents-pythonPython SDK for multi-agent workflows with handoffs, guardrails and tracing | 92 out of 100 |
| #4 | mastraTypeScript framework for building AI agents, workflows and MCP servers | 92 out of 100 |
| #5 | browser-usePython library that lets an LLM agent control a web browser | 90 out of 100 |