> For the complete documentation index, see [llms.txt](https://docs.pipekit.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.pipekit.io/get-started/operate.md).

# Operate Pipekit

A 10-minute path from "I have a Kubernetes cluster with Argo Workflows" to "Pipekit is managing it."

This quickstart assumes you're connecting your own cluster to **Pipekit Cloud**. For a fully self-hosted control plane (no Pipekit Cloud connection), see [Self-Hosting Pipekit > Evaluate Self-Hosted](/self-hosting-pipekit/evaluate.md) instead.

## Prerequisites

* A Kubernetes cluster with [Argo Workflows installed](https://argo-workflows.readthedocs.io/en/stable/installation/#production-installation).
* `kubectl` configured against that cluster.
* `helm` (v3) installed locally.
* A Pipekit Cloud account. If you don't have one, follow [Evaluate Pipekit Cloud](/get-started/evaluate-cloud.md) steps 1 to 2, then come back.

## 1. Create the Cluster in Pipekit

1. Log in to <https://pipekit.io>.
2. Decline the offer for a Pipekit-hosted free-trial cluster.
3. Click `Connect A Cluster`.
4. Select the [Organization](/concepts/organization.md) the cluster will belong to.
5. Give the [Cluster](/concepts/cluster.md) a name (lowercase, no whitespace) and a description.
6. Click `Submit`.

Pipekit generates the install command and Helm values for your cluster.

## 2. Install the Pipekit Agent

Pipekit's UI prints a Helm install command of the form:

```bash
helm upgrade pipekit-agent oci://registry.pipekit.io/charts/pipekit-agent \
  --install \
  --namespace=argo \
  --set secrets.pipekitSecretAccessKey="<provided>" \
  --set secrets.pipekitClusterId="<provided>"
```

Run that command against your cluster. For the full chart values and configuration options, see [Reference > Helm Chart Values > Agent](/reference/helm-values/agent.md).

## 3. Verify the agent is connected

```bash
kubectl get pods -n argo -l app=pipekit-agent
```

You should see a Pipekit Agent pod in `Running` state within a minute. The Pipekit UI's Cluster page also flips from `Connecting` to `Active`.

## 4. Submit a sample workflow

```bash
brew install pipekit/tap/cli   # or see other install methods in Reference > CLI
pipekit login
git clone https://github.com/pipekit/examples.git
cd examples
pipekit submit --cluster-name=<your-cluster-name> examples/dag-diamond/workflow.yaml
```

The CLI prints a link to the [Run](/concepts/run.md) in the Pipekit UI. Open it to see the [Run Graph](/using-pipekit/runs/run-graph.md), [Pod Logs](/using-pipekit/runs/pod-logs.md), and the submitted workflow YAML.

## What's next

* **Configure SSO and access control** before inviting teammates: [Access Control](/concepts/access-control.md), [SSO setup](/organizations-and-access/sso.md).
* **Connect a Git provider** so Workflows trigger on commits and pull requests: [Pipes](/using-pipekit/pipes.md).
* **Set up alerts** for workflow failures: [Alert Providers](/organizations-and-access/alert-providers.md).
* **Scale your cluster** beyond the defaults: [Operating at Scale](/using-pipekit/operating-at-scale.md).

## If something goes wrong

The Pipekit Agent logs are your first stop:

```bash
kubectl logs -n argo -l app=pipekit-agent --tail=200
```

If the agent reports an auth or registration error, double-check the secret values from step 2. If it's reaching Pipekit but the Cluster still shows `Connecting` in the UI, check that the `pipekitClusterId` matches the Cluster you created in step 1.