openaiDeveloper tools
Openai Agents Python
openai-agents-python
A Python framework by OpenAI for building multi-agent workflows, supporting text, voice, and sandbox-based agent applications.
- Backend
- Full stack
- Automation
- Collaboration
- Library/framework
- macOS
- Linux
- Windows
- Self-hostable
- Runs locally
- Docker supported
- Needs API key
- Popularity
- 29.0k Stars
- GitHub stars
- Recent activity
- 8/25/2026
- Updated in the last 30 days
- License
- MIT
- Permissive
Why it matters
We look beyond stars: what problem it solves, whether it creates real utility, and what makes its approach worth noticing.
Problem
Addresses the lack of a standardized framework for developers to manage complex multi-agent workflows, including session state, tool execution, and sandbox environments.
Practical value
Provides comprehensive abstractions ranging from basic text interactions to real-time voice streaming and sandbox code execution, lowering the engineering barrier for integrating OpenAI agent capabilities.
Innovation / differentiation
Deeply integrates sandbox isolation, real-time voice processing, and multi-agent orchestration while supporting the MCP protocol, offering a highly practical approach to production-grade agent development.
Leverage potential
As an official OpenAI SDK, it leverages underlying API updates for performance gains and maintains high extensibility for various LLM backends.
Why now
Released as agent applications transition from experimentation to production, aligning with the current developer demand for stable and observable agent frameworks.
Community activity
In the last 90 days there were 158 new issues and 965 pull requests; the bounded issue/PR samples include 53 issue authors and 37 PR contributors, with at least 17 releases.
Maintainer responsiveness
The 99-issue window sample had a 100% close rate, and the 98-pull-request sample had a 20% merge rate. Maintainer-response observations covered 49% of that issue sample, with a 14% response rate and median first response of 3.9 hours.
Key highlights
- Agent orchestration supporting multiple LLM providers
- Built-in sandbox for file operations and code execution
- Real-time voice streaming and multimodal interaction
Quick start
How it is installed, how hard it is, and where to start.
Where it runs
Runs locally
Difficulty
Medium — some setup needed
- 01Create and activate a Python virtual environment.
- 02Install the core SDK package using pip.
- 03Set the OPENAIAPIKEY environment variable.
- 04Write code to instantiate an Agent and execute tasks via the Runner.
Best for
- Teams that want data on their own servers
- Developers who want to try it on their machine
- People who prefer Docker deploys
More about it
OpenAI Agents SDK is a Python framework for building multi-agent systems. It allows developers to define agents with specific instructions, tool-calling capabilities, and guardrails, while utilizing sandbox environments for long-running tasks or real-time voice pipelines.
The framework supports delegating tasks between agents, and includes built-in session management, tracing, and human-in-the-loop mechanisms to help developers implement automated workflows in complex scenarios.
Sources
Each field shows its status and source — expand to review.
15 · Expand
Sources
Each field shows its status and source — expand to review.
capability tags
Verifiedautomation, collaboration
Source: admin_cms · cms editor · 8/24/2026
editor note
Verified{"en":"This SDK provides structured agent orchestration, ideal for developers handling complex tasks, file operations, or real-time voice interactions. Note the requirement for Python 3.10+.","zh":"该 SDK 提供了结构化的智能体编排能力，特别适合需要处理复杂任务、文件操作或实时语音交互的开发者。使用前需注意其对 Python 3.10+ 的版本要求。"}
Source: admin_cms · cms editor · 8/24/2026
how to use
Verified{"steps":[{"en":"Create and activate a Python virtual environment.","zh":"创建并激活 Python 虚拟环境。"},{"en":"Install the core SDK package using pip.","zh":"使用 pip 安装核心 SDK 包。"},{"en":"Set the OPENAI_API_KEY environment variable.","zh":"设置 OPENAI_API_KEY 环境变量。"},{"en":"Write code to instantiate an Agent and execute tasks via the Runner.","zh":"编写代码实例化 Agent 并通过 Runner 执行任务。"}],"installAt":"local","difficulty":"medium"}
Source: admin_cms · cms editor · 8/24/2026
intro
Verified{"en":"OpenAI Agents SDK is a Python framework for building multi-agent systems. It allows developers to define agents with specific instructions, tool-calling capabilities, and guardrails, while utilizing sandbox environments for long-running tasks or real-time voice pipelines.\n\nThe framework supports delegating tasks between agents, and includes built-in session management, tracing, and human-in-the-loop mechanisms to help developers implement automated workflows in complex scenarios.","zh":"OpenAI Agents SDK 是一个用于构建多智能体系统的 Python 框架。它允许开发者定义具备特定指令、工具调用能力和安全防护的智能体，并通过沙箱环境执行长期任务或构建实时语音管道。\n\n该框架支持将智能体作为工具进行嵌套调用，并内置了会话管理、追踪调试及人机协作功能，帮助开发者在复杂的业务场景中实现自动化工作流。"}
Source: admin_cms · cms editor · 8/24/2026
Latest release
Verifiedv0.22.0
Source: GitHub API · latest_release=v0.22.0 · 8/25/2026
License
VerifiedMIT
Source: GitHub API · license.spdx_id=MIT · 8/25/2026
needs api key
VerifiedYes
Source: admin_cms · cms editor · 8/24/2026
One-liner
Verified{"en":"A Python framework by OpenAI for building multi-agent workflows, supporting text, voice, and sandbox-based agent applications.","zh":"OpenAI 推出的 Python 多智能体工作流开发框架，支持构建文本、语音及沙箱环境下的智能体应用。"}
Source: admin_cms · cms editor · 8/24/2026
Platforms
Verifiedmacos, linux, windows
Source: admin_cms · cms editor · 8/24/2026
Category hint
Inferred from materialsai-apps
Source: Project README · hint=ai-apps · 8/17/2026
product forms
Verifiedlibrary_framework
Source: admin_cms · cms editor · 8/24/2026
role tags
Verifiedbackend, fullstack
Source: admin_cms · cms editor · 8/24/2026
supports docker
VerifiedYes
Source: Repository file · dockerfile=true; compose=false · 8/25/2026
supports local
VerifiedYes
Source: admin_cms · cms editor · 8/24/2026
supports self host
VerifiedYes
Source: admin_cms · cms editor · 8/24/2026
Related projects
Other verified projects matched by category, capabilities, and intended roles.
flipt
A self-hosted feature management platform that stores feature flags in Git and manages multi-environment delivery through a web UI, APIs, and real-time updates.
massing
Open-source, self-hosted, IFC-native AEC platform featuring in-browser BIM modeling, a general contractor portal, and development proforma tools.
puppeteer
A JavaScript library that provides a high-level API to control Chrome or Firefox over DevTools Protocol or WebDriver BiDi.
compose
Docker’s official CLI tool for defining and running multi-container applications from a Compose file.
WSL
Windows Subsystem for Linux allows you to run Linux command-line tools and applications directly on Windows without the overhead of a traditional virtual machine.
opencodex
A universal local proxy that enables the use of any LLM, such as Claude, Gemini, or DeepSeek, within Codex and Claude Code, while managing ChatGPT account pools.