All-in-One (AIO) Sandbox Example

This guide demonstrates how to create and access an All-in-One (AIO) Sandbox via Agent-Sandbox, featuring multiple pre-installed tools including VNC, VSCode, Jupyter, and Terminal in one unified environment.

This example demonstrates how to create and access an All-in-One (AIO) Sandbox via Agent-Sandbox.

Create an AIO Sandbox

Apply the sandbox manifest with AIO Sandbox runtime. You can find the latest version here.

kubectl apply -f aio-sandbox.yaml
# sandbox.agents.x-k8s.io/aio-sandbox-example created

They can then check the status of the applied resource. Verify sandbox and pod are running:

# wait until the sandbox is ready
kubectl wait --for=condition=Ready sandbox aio-sandbox-example

kubectl get sandbox
# NAME                  AGE
# aio-sandbox-example   41s
kubectl get pod aio-sandbox-example
# NAME                  READY   STATUS    RESTARTS   AGE
# aio-sandbox-example   1/1     Running   0          49s

The AIO Sandbox has multiple tools pre-installed, including VNC, VSCode, Jupyter and Terminal in one unified environment.

Accessing the AIO Sandbox Server

Port forward the aio-sandbox server port.

# set ingress host and port
export INGRESS_HOST="localhost"
export INGRESS_PORT="8080"

# port forward to the aio-sandbox pod
kubectl port-forward --address ${INGRESS_HOST} pod/aio-sandbox-example ${INGRESS_PORT}:8080

Setup GATEWAY_URL environment variable:

export GATEWAY_URL="http://${INGRESS_HOST}:${INGRESS_PORT}"

echo "$GATEWAY_URL"
# http://localhost:8080

Connect to the aio-sandbox on a browser via http://<INGRESS_HOST>:<INGRESS_PORT>

Access the AIO Sandbox via Python SDK

# set up a virtual environment if needed
python3 -m venv venv
source venv/bin/activate

# install the agent-sandbox package
pip install agent-sandbox==0.0.18

Note: The agent-sandbox Python SDK installed above is for controlling tools inside the AIO sandbox (e.g., browser, shell, file system). If you want to programmatically provision or manage sandbox infrastructure (Kubernetes resource lifecycle), use the complementary agentic-sandbox-client Python SDK. These two SDKs can be used together for a complete infrastructure + application workflow.

Run the basic python example:

python3 main.py

More examples

For more examples of using the AIO Sandbox, refer to agents-infra/sandbox examples.

Cleanup

To remove the sandbox and related resources when you are done, run:

kubectl delete -f aio-sandbox.yaml

# Cleanup sandbox_screenshot.png from main.py if needed
rm sandbox_screenshot.png

References


Last modified January 20, 2026: Initial push for website reorg. (ce9ee0d)