Docker Sandboxes

Docker Sandboxes run AI coding agents in isolated microVM sandboxes. Each sandbox gets its own Docker daemon, filesystem, and network — the agent can build containers, install packages, and modify files without accessing host resources beyond those you share.

[!NOTE] The sbx CLI is free to use, including for commercial work. Only organization governance requires a separate paid subscription.

Organization admins can centrally manage sandbox network, filesystem, and MCP policies, so the same controls apply uniformly across every developer's machine. Available on a separate paid subscription.

Get started#

Follow the installation guide to check the system requirements, install the sbx CLI, and sign in.

Then launch an agent in a sandbox:

$ cd ~/my-project
$ sbx run claude

See the get started guide for a first-session walkthrough, or jump to the usage guide for basic commands.

Learn more#

• Agents — supported agents and per-agent configuration
• Workflows — patterns for Git, local development, authentication, agent skills, and automation
• Configuration — manage credentials, declare project environments, turn on GPU passthrough, and configure an upstream proxy
• Integrations — connect editors and apps like VS Code and Cursor to a sandbox over SSH
• MCP gateway — register MCP servers and connect them to sandboxed agents
• Customize — reusable templates and declarative kits for extending or tailoring sandboxes
• Architecture — microVM isolation, workspace mounting, networking
• Security — isolation model, credential handling, and network policies
• CLI reference — full list of sbx commands and options
• Troubleshooting — common issues and fixes
• FAQ — login requirements, telemetry, etc

Feedback#

Your feedback shapes what gets built next. If you run into a bug, hit a missing feature, or have a suggestion, open an issue at github.com/docker/sbx-releases/issues.