Skip to content
An Agentic AI Foundation (AAIF) Project

Your agents
Your sandbox

Run AI agents on your own infrastructure. Give them isolated environments to execute code, use tools, and explore the web.

Start with Docker. Scale on Kubernetes.

Open source · Apache 2.0
An isolated sandbox connects your agent to code, browsers, and tools. sandbox / isolated
AI agent
Code · Browser · Tools
sandbox.pyExecution preview
from opensandbox import Sandboxsandbox = await Sandbox.create("alpine")try:    result = await sandbox.commands.run(        "echo 'Hello OpenSandbox!'"    )finally:    await sandbox.destroy()
Hello OpenSandbox!

A workspace for every workload

Let your agents get to work.

Explore examples
Fast Sandbox runtime

Less waiting.
More running.

Fast, high-density sandboxes on Kubernetes. Restore Firecracker microVMs from pre-warmed capacity, and pause with memory and disk state preserved.

Explore Fast Sandbox
Firecracker warm creation
64ms

P50 · Serial creation

125.4ms

P99 · 10 concurrent

Cached templateRestore microVMReady
Read the benchmark & conditions

Your infrastructure. Your boundaries.

Give agents access.
Keep control.

One sandbox API, from your laptop to your cluster. Define what agents can reach and how they use credentials.

From your first sandbox to your next idea

Make room
for your next agent.

Start the server, pick your SDK, and run your first command. Your infrastructure takes it from there.

Follow the quick start
1

Start a local server

Docker + Python 3.10+
# Initialize your Docker configuration
uvx opensandbox-server init-config \
  ~/.sandbox.toml --example docker

# Start the server
uvx opensandbox-server
2

Install your SDK

pip install opensandbox
Read the Python guide

Copyright © OpenSandbox a Series of LF Projects, LLC