opensandbox-group/opensandbox is a general-purpose sandbox platform for AI applications that provides unified APIs, lifecycle management, and Docker/Kubernetes runtimes. It addresses the need for secure, isolated, and scalable execution environments where AI agents can run code and commands.
Project overview
It combines sandbox lifecycle management with support for both Docker and Kubernetes runtimes, enabling local runs and large-scale distributed scheduling for AI agent workloads within a single platform.
Project type
MCP · AI Agent · Infrastructure
Use cases
Automation
Deployment
Refer to project documentation
License
Apache-2.0
Best for
Developers building AI agent applications that require secure, isolated execution environments for code and commands.
AI engineers and operations teams who need sandbox lifecycle management across Docker and Kubernetes runtimes.
Key capabilities
Built-in lifecycle management supporting Docker and Kubernetes runtime, enabling both local runs and large-scale distributed scheduling.
Provides multi-language SDKs, the osb CLI, and MCP server integration for sandbox creation, command execution, and file operations.
Built-in Command, Filesystem, and Code Interpreter implementations.
Unified ingress gateway with multiple routing strategies plus per-sandbox egress controls.
Secure credential injection for sandbox outbound requests without exposing real secrets to workloads.
Supports secure container runtimes like gVisor, Kata Containers, and Firecracker microVM for enhanced isolation.
Limitations and risks
Minimum hardware requirements for deployment are not documented.
Getting started
Install Docker and Python 3.10+ as prerequisites.
Run uvx opensandbox-server init-config ~/.sandbox.toml --example docker to generate the configuration.
Run uvx opensandbox-server to launch the platform. Using the SDKs requires code execution.
Evidence and sources
README: OpenSandbox is a **general-purpose sandbox platform** for AI applications, offering multi-language SDKs, unified sandbox APIs, and Docker/Kubernetes runtimes for scenarios like Co…
README: - 🧩 SDKs, CLI, and MCP: Provides multi-language SDKs, the osb CLI, and MCP server integration for sandbox creation, command execution, and file operations. See SDKs, CLI, and MCP.…
README: OpenSandbox also provides `osb`, a terminal CLI for the common sandbox workflow: create sandboxes, run commands, move files, inspect diagnostics, and manage runtime egress policy.