GitOps made Git the single source of truth for Kubernetes delivery, and Argo CD does an excellent job of continuously reconciling cluster state against what's declared in a repo. What it doesn't solve on its own is release promotion — the structured, governed movement of an artifact from dev to staging to production with traceability and approval gates along the way. That's the gap this article walks through, using Kargo — Akuity's open-source continuous promotion platform — as the missing layer on top of Argo CD.
Why traditional GitOps promotion breaks down
In a typical Argo CD setup you end up with one Application resource per environment, each pointing at a different branch or values file. Promoting a build means editing the right file, committing, and waiting for reconciliation — usually stitched together with ad-hoc CI scripts. There's no single view of what version is running where, no built-in concept of a verification gate ("only promote after staging has been healthy for 24 hours"), and rollback is just a manual Git revert.
Where Kargo fits
Kargo doesn't replace Argo CD — it sits alongside it and owns the decision of what gets promoted, where, and when, then writes the corresponding change to Git so Argo CD can do the actual reconciliation. It's fully declarative (promotion pipelines are Kubernetes custom resources), tooling-agnostic (Helm, Kustomize, or raw manifests), and treats every promotion as an auditable, first-class operation rather than a side effect of a CI job.
The full piece on Medium goes further into the enterprise rollout details — governance policies, approval workflows, and the promotion pipeline configuration end to end.