Let's Explore a Secure Sandbox for AI Agents
This sandbox can be used for running and testing LLM-generated code, and you can install dependencies, create and restore state using snapshots, and manage the running processes.
1. Run your first sandbox
Watch Python code execute in a secure cloud environment.
2. Install packages and run something heavy
See how dependency installation and execution works.
3. Deploy a web server
You can run arbitrary code in sandboxes. Let's spin up a web server.
4. Create a snapshot
Save the memory and filesystem state of a sandbox.
5. Restore a snapshot
Now we'll restore a snapshot, which will create a sandbox which has the same environment variables and memory state as the sandbox we created in the previous step.
6. Process Management
The Sandbox provides process management through the process property. You can execute Python code, run shell commands, and manage long-running processes with real-time output streaming.
Ready to take sandboxes for a spin?
Create a free account and start building