My motivation was to give PR/MR reviewers a very low-friction way to see a Helm chart change running.
The workflow is intentionally simple: install a GitHub App (or call a REST API in other workflows), open a PR/MR, and you get a live preview. That’s it.
There’s no ArgoCD setup, no Helmfile, no cluster provisioning, no DNS wiring to build or maintain. The goal was to make it trivial for reviewers to see “this PR running” — especially for public Helm charts where contributors and reviewers can’t realistically be expected to set up infrastructure just to demo a change.
If you already run ephemeral previews via ArgoCD or Helmfile, this probably isn’t adding much value. Those approaches work well once they’re in place. Chart Preview is aimed at the cases where teams want PR previews without having to design, build, and maintain that machinery themselves.
That makes sense — thanks for clarifying. Framing it as “zero infra ownership, just a reviewer convenience” really helps explain where this fits compared to ArgoCD-style previews.
Thanks for the feedback. I agree your setup sounds solid.
What you're describing - different values, Argo apps per branch, pipelines that create and clean up - absolutely works. In my experience though, getting to that point involves a lot of work: Argo setup, application generation per branch, DNS and ingress wiring, and ongoing maintenance. It's reliable once built, but non-trivial to design and keep running.
The scope of what I'm trying to solve is narrower. In my case, Helm chart changes (often from external contributors) were hard to review because validating them meant manually standing up environments or spinning up short-lived clusters. Many reviewers couldn't realistically run the pipeline themselves, so reviews often stalled. That was the core motivation.
This isn't aimed at teams that already have a mature GitOps setup like yours — I agree Chart Preview probably won't add much value there. It's for teams that don't have that infrastructure yet, or don't want to invest in building and maintaining it, and want to get to the same outcome with less effort.
I appreciate you taking the time to describe your approach — it's a good reference point for where many teams eventually end up.
Thanks for the feedback — you’re spot on about the setup this is trying to speed up. The namespace-per-branch approach works well (and that’s what this does), but the setup around ingress, DNS, secrets, and cleanup tends to be the real time sink. Glad it resonates.
Thanks for the thoughtful feedback — these are all fair concerns.
The app doesn’t access your production clusters. Previews run in managed, isolated clusters, and each preview gets its own namespace with deny-all NetworkPolicies, quotas, and automatic teardown. That said, if the concern is about installing charts into any external K8 cluster at all, then I agree this won’t be a fit — and that’s a reasonable constraint.
It’s GitHub-first today simply because that’s where I personally hit the problem. GitLab is supported via the REST API using a personal access token that you can scope as tightly as you want, so you can trigger previews from GitLab CI today.
Native GitLab App integration (auto-triggering on MRs, status updates, etc.) is something I’ve thought about, but I wanted to validate the core workflow first.
It is intentionally Helm-only for now. The specific pain I was trying to solve was reviewing Helm changes — values layering, dependencies, and template changes — by seeing them running in a real environment, rather than trying to generalise across all deployment models.
I’m not trying to replace or compete with Flux or Argo CD. The idea is to validate Helm changes before they land in a GitOps repo or get promoted through environments — essentially answering the question of “does this look OK and actually work when deployed, so should be safe to merge?”
It doesn’t expose rendered manifest diffs today, but I agree that would be valuable — especially a readable “what changed after Helm rendering” view tied back to the PR. I’m still thinking through the cleanest way to do that without adding a lot of complexity to the workflow.
Appreciate you taking the time give your feedback. Thanks.
My motivation was to give PR/MR reviewers a very low-friction way to see a Helm chart change running.
The workflow is intentionally simple: install a GitHub App (or call a REST API in other workflows), open a PR/MR, and you get a live preview. That’s it.
There’s no ArgoCD setup, no Helmfile, no cluster provisioning, no DNS wiring to build or maintain. The goal was to make it trivial for reviewers to see “this PR running” — especially for public Helm charts where contributors and reviewers can’t realistically be expected to set up infrastructure just to demo a change.
If you already run ephemeral previews via ArgoCD or Helmfile, this probably isn’t adding much value. Those approaches work well once they’re in place. Chart Preview is aimed at the cases where teams want PR previews without having to design, build, and maintain that machinery themselves.