经 AI Skill Hub 精选评估,OpenSandbox Agent工作流 获评「强烈推荐」。在 GitHub 上收获超过 10.6k 颗 Star,这款Agent工作流在功能完整性、社区活跃度和易用性方面表现出色,AI 评分 8.2 分,适合有一定技术背景的用户使用。
OpenSandbox Agent工作流 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
OpenSandbox Agent工作流 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 方式一:pip 安装(推荐)
pip install opensandbox
# 方式二:虚拟环境安装(推荐生产环境)
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install opensandbox
# 方式三:从源码安装(获取最新功能)
git clone https://github.com/alibaba/OpenSandbox
cd OpenSandbox
pip install -e .
# 验证安装
python -c "import opensandbox; print('安装成功')"
# 命令行使用
opensandbox --help
# 基本用法
opensandbox input_file -o output_file
# Python 代码中调用
import opensandbox
# 示例
result = opensandbox.process("input")
print(result)
# opensandbox 配置文件示例(config.yml) app: name: "opensandbox" debug: false log_level: "INFO" # 运行时指定配置文件 opensandbox --config config.yml # 或通过环境变量配置 export OPENSANDBOX_API_KEY="your-key" export OPENSANDBOX_OUTPUT_DIR="./output"
<p align="center"> <a href="https://github.com/opensandbox-group/OpenSandbox"><img src="https://img.shields.io/github/stars/opensandbox-group/OpenSandbox?style=flat-square&logo=github&logoColor=white&label=Stars&color=181717" alt="Stars" /></a> <a href="https://www.bestpractices.dev/projects/12588"><img src="https://img.shields.io/badge/OpenSSF-Best-4C566A?style=flat-square" alt="OpenSSF Best Practices" /></a> <a href="https://landscape.cncf.io/?item=orchestration-management--scheduling-orchestration--opensandbox"><img src="https://img.shields.io/badge/CNCF-Landscape-0C66E4?style=flat-square" alt="CNCF Landscape" /></a> <a href="https://discord.gg/g7FuPs8YeD"><img src="https://img.shields.io/badge/Discord-Join-5865F2?style=flat-square&logo=discord&logoColor=white" alt="Discord" /></a> <a href="https://qr.dingtalk.com/action/joingroup?code=v1,k1,A4Bgl5q1I1eNU/r33D18YFNrMY108aFF38V+r19RJOM=&_dt_no_comment=1&origin=11"><img src="https://img.shields.io/badge/DingTalk-Join-0089FF?style=flat-square" alt="DingTalk" /></a> <a href="https://github.com/opensandbox-group/OpenSandbox/actions"><img src="https://img.shields.io/github/actions/workflow/status/opensandbox-group/OpenSandbox/real-e2e.yml?branch=main&label=TEST&style=flat-square&logo=github&logoColor=white" alt="E2E Status" /></a> <a href="https://github.com/opensandbox-group/OpenSandbox/actions"><img src="https://img.shields.io/github/actions/workflow/status/opensandbox-group/OpenSandbox/kubernetes-nightly-build.yml?branch=main&label=K8S&style=flat-square&logo=kubernetes&logoColor=white" alt="Kubernetes nightly build status" /></a> </p>
<p align="center"> <a href="https://trendshift.io/repositories/21828" target="_blank"><img src="https://trendshift.io/api/badge/repositories/21828" alt="opensandbox-group%2FOpenSandbox | Trendshift" style="width: 320px; height: 70px;" width="320" height="70" /></a> </p>
<hr /> </div>
Run AI agents in sandboxes on your own infrastructure.
OpenSandbox gives AI applications isolated environments to execute code, run commands, manage files, and operate browsers or desktops. Start locally with Docker and deploy on Kubernetes through a unified sandbox API.
Quick Start · Examples · Documentation · Fast Sandbox
| Feature | What it enables | Learn more |
|---|---|---|
| **Fast Sandbox runtime** | Fast, high-density sandboxes on Kubernetes. Firecracker creation: **64 ms P50 (serial)** / **125.4 ms P99 (10 concurrent)**. Firecracker sandboxes support pause/resume with memory and disk state preserved. | [Integration](docs/architecture/fast-sandbox/index.md) · [Performance](docs/architecture/fast-sandbox/performance.md) · [Pause/resume](docs/architecture/fast-sandbox/checkpoints.md) |
| **Agent working environments** | Execute commands, manage files, and run code with built-in APIs. Integration examples show how to run coding agents, browsers, and desktops inside sandboxes. | [Examples](docs/examples/index.md) |
| **Network access control** | Route inbound traffic through a unified ingress gateway and control outbound access with per-sandbox egress policies. | [Ingress](docs/architecture/network/ingress.md) · [Egress](docs/architecture/network/egress.md) |
| **Credential Vault** | Let agents call external services without exposing real credentials to sandbox workloads. | [Credential Vault](docs/guides/credential-vault.md) |
| **Local to cluster** | Start with Docker and deploy on Kubernetes through a unified lifecycle API. Resource pools and batch creation support agent evaluation and RL training workloads. | [Kubernetes runtime](docs/architecture/control-plane/operator.md) |
| **SDKs, CLI, and MCP** | Integrate with Python, Java/Kotlin, TypeScript/JavaScript, C#/.NET, or Go SDKs. Use osb from the terminal or connect agents through MCP. | [SDKs](#sdks) · [CLI](#cli) · [MCP](#mcp) |
| **Extensible sandbox protocol** | Build custom runtime integrations against defined sandbox lifecycle and execution APIs. | [API specs](specs/README.md) |
Requirements:
```bash uvx opensandbox-server init-config ~/.sandbox.toml --example docker
uvx opensandbox-server
Explore examples by what you want your agent to do. Runnable source code lives in examples/.
| Use case | What you can build | Examples |
|---|---|---|
| **Coding agents** | Run coding agents in isolated environments to edit files, execute commands, and complete development tasks. | [Claude Code](docs/examples/claude-code.md) · [Codex CLI](docs/examples/codex-cli.md) · [DeerFlow](docs/examples/deer-flow.md) |
| **Code execution and data analysis** | Execute model-generated code and work with results through the Code Interpreter SDK. | [Code Interpreter](docs/examples/code-interpreter.md) |
| **Browser and desktop automation** | Automate web interactions and testing, or give agents access to a desktop environment. | [Playwright](docs/examples/playwright.md) · [Chrome](docs/examples/chrome.md) · [Desktop](docs/examples/desktop.md) |
| **Agent evaluation** | Run evaluations with a separate sandbox for each trial. | [Harbor Evaluation](docs/examples/harbor-evaluation.md) |
See the full example catalog for more coding agents, framework integrations, remote development environments, Kubernetes deployment, and storage patterns.
Pick your language:
<details> <summary><b>Python</b></summary>
pip install opensandbox
</details>
<details> <summary><b>Java/Kotlin (Gradle Kotlin DSL)</b></summary>
dependencies {
implementation("com.alibaba.opensandbox:sandbox:{latest_version}")
}
</details>
<details> <summary><b>Java/Kotlin (Maven)</b></summary>
<dependency>
<groupId>com.alibaba.opensandbox</groupId>
<artifactId>sandbox</artifactId>
<version>{latest_version}</version>
</dependency>
</details>
<details> <summary><b>JavaScript/TypeScript</b></summary>
npm install @alibaba-group/opensandbox
</details>
<details> <summary><b>C#/.NET</b></summary>
dotnet add package Alibaba.OpenSandbox
</details>
<details> <summary><b>Go</b></summary>
go get github.com/alibaba/OpenSandbox/sdks/sandbox/go
</details>
OpenSandbox是企业级AI基础设施的优秀选择,结合安全隔离、高性能和Kubernetes支持,解决agent执行的关键痛点,社区活跃度高。
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ Apache 2.0 — 宽松开源协议,可商用,需保留版权声明和 NOTICE 文件,含专利授权条款。
AI Skill Hub 点评:OpenSandbox Agent工作流 的核心功能完整,质量优秀。对于自动化工程师和运维人员来说,这是一个值得纳入个人工具库的选择。建议先在非生产环境试用,再逐步推广。
| 原始名称 | OpenSandbox |
| 原始描述 | 开源AI工作流:Secure, Fast, and Extensible Sandbox runtime for AI agents.。⭐10.6k · Python |
| Topics | AI代理沙箱运行时工作流编排Kubernetes安全隔离 |
| GitHub | https://github.com/alibaba/OpenSandbox |
| License | Apache-2.0 |
| 语言 | Python |
收录时间:2026-05-13 · 更新时间:2026-05-16 · License:Apache-2.0 · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端