> 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/why-pipekit.md).

# Why Pipekit

Pipekit is the control plane for [Argo Workflows](https://argoproj.github.io/argo-workflows/). Platform teams use it to operate Argo at scale and give their developers a self-serve surface; data and ML teams use it to find, debug, and re-run their pipelines without learning the cluster.

Pipekit sits on top of Argo, not in place of it. Your workflows still execute as Argo `Workflows` on your own Kubernetes clusters; Pipekit adds the dashboards, access control, multi-cluster management, log handling, and integrations that teams typically rebuild themselves.

Four reasons teams adopt Pipekit:

* [**Observability**](/why-pipekit/observability.md): unified UI across clusters, persisted Run history and logs, log-level detection and search, OpenTelemetry workflow metrics.
* [**Governance**](/why-pipekit/governance.md): workspaces, IdP-backed RBAC, per-environment secrets, default-deny access, audit-quality submission enforcement.
* [**Scale**](/why-pipekit/scale.md): multi-cluster routing, node-status offloading, Vector-based log collection, cross-cluster disaster recovery.
* [**Security**](/why-pipekit/security.md): bring-your-own-cluster architecture, SBOMs and signed containers, optional self-hosted control plane for air-gapped deployments.

## Architecture at a glance

![Pipekit architecture](/files/y1Cbl9j5aMLKWLt954ql)

Pipekit deploys a small [agent](/concepts/pipekit-agent.md) into each Kubernetes cluster alongside Argo Workflows. The agent brokers commands from Pipekit's control plane and reports status, logs, and metrics back. Your workflows, data, and compute stay in your cluster; the control plane is hosted by Pipekit, or you can [self-host](/self-hosting-pipekit.md) the whole stack.

For the two deployment models side by side, see [Pipekit Cloud vs Self-Hosted](/concepts/cloud-vs-self-hosted.md).

To try Pipekit, follow [Get Started > Evaluate Pipekit Cloud](/get-started/evaluate-cloud.md). It's a 5-minute path from sign-up to a running workflow.