Introducing CodePier

Today we're opening the doors on CodePier. One window where you can hot-swap your laptop into a running pod, edit your manifests in a real editor, track what's drifted from your cluster, snapshot a release, and ship it. Your whole team, working out of the same place.

We've been building this for a while. Today it's yours.

The tool we always wanted to work with

Most people on a Kubernetes team already know where the real kubeconfig is. They know which dashboard to open, which terminal has the port-forward in it, and which directory has the manifests that actually match production. That knowledge lives in a few people's heads, and on the days they're out, everyone else is guessing.

CodePier is what that knowledge looks like when you put it in one window. Your manifests are grouped by service, the way you'd group them yourself. Overlays describe what changes per environment. Port Forwards and the Ingress Proxy bring the cluster onto your desk. A Manifest Editor with drift detection keeps everyone honest. And it's a desktop app, so the person next to you can open the same project on their laptop and see the same thing you do.

What ships today

There's a lot in the first release. Here's the shape of it.

The dev loop

• Hot-swap — codepier up takes your laptop and drops it into any deployment. Live file sync, ports forwarded, and the original deployment restored automatically when you exit.
• Port Forwards — pin any cluster Service to your machine and reach it at a friendly hostname like api.local. The proxy publishes over mDNS, so the same URL works from your laptop, your phone, or anything else on the network.
• Ingress Proxy — cluster ingresses come through an HTTPS reverse proxy with certificates managed by mkcert. No more self-signed certificate warnings.
• Pod Logs — tail any pod and any container, with multiple streams side by side and filters that actually work.

Your cluster, imported

• Cluster Import — point CodePier at a cluster you already run. It scans Deployments, StatefulSets, DaemonSets, Services, Ingresses, ConfigMaps, Secrets, and PVCs, and imports them grouped by service. Nothing restarts.
• Matched Builds — CodePier watches your CI registry and matches every new image back to the workload it belongs to. One click swaps a running deployment to a fresh tag.

Manifests, Overlays, Releases

• Manifest Editor with drift detection — a real editor for Deployments, Services, Ingresses, and the rest. Every resource carries a sync status, and the moment cluster state diverges from what you wrote, it shows up on the resource.
• Environment Overlays — write your manifests once. Each Overlay layers on its own image tags, replica counts, labels, annotations, and name prefix per environment.
• Versioned Releases — snapshot your manifests and per-environment overrides as a tagged Release. Pin a build per environment and promote it forward when you're ready.
• Git Sync — connect a GitHub or GitLab repo and CodePier keeps your manifests in sync as you edit them.

The rest of the toolbox

• Built-in K8s Assistant — a Claude-powered assistant that reads your manifests, your environments, and your live cluster state before it answers.
• Cost Visibility — OpenCost wired in. Drill from a cluster down to a single workload and see what it costs you per hour.
• Helm, without the CLI — browse charts, set values per environment, and install to any cluster from inside CodePier.
• Secrets via External Secrets Operator — encrypted secrets sit alongside your project and sync into the cluster through ESO. No second product to set up.
• Deploy Hooks — outbound webhooks on Ship It, service deploy, and image swap events.
• Volumes, per cluster — define a volume once and bind its storage per cluster, so the same project lands correctly on every cluster you deploy it to.
• Passkeys & 2FA — sign in with a passkey or add TOTP two-factor to your account, and invite teammates by email.

One window, one team

The thing we're most proud of isn't any single feature on that list. It's that they all live in the same app. You don't have to tab between a dashboard, a terminal, a YAML editor, and a cost tool. You don't have to Slack your teammate to ask which cluster they're pointed at. Projects, manifests, environments, and releases are shared. Everyone on your team is looking at the same thing.

How to get it

Download the desktop app from our download page, point it at a cluster you already run, and you'll be working against it in about a minute. The CLI is installed and managed by the app, so there's nothing to brew or curl by hand.

What's next

Preview environments — an isolated namespace per pull request with a shareable URL, torn down when the PR merges — is the next big thing we're shipping. A few other items on the roadmap we're excited about: custom build pipelines, CI/CD integration, and a plugin system for teams that want to extend CodePier with their own commands.

This is only the beginning

Everything above is what shipped on day one. There's still a long list of things we want to improve, a longer list of features we want to add, and an even longer list of rough edges we haven't found yet. That's where you come in.

Please give us feedback. File bugs, request features, tell us what's confusing, tell us what's missing, tell us what you wish worked differently. Everything goes in one place: github.com/codepier/codepier/issues. A human on our team reads every single one.

If you'd rather email, we're at [email protected]. If you'd like a guided walkthrough on your own cluster, book a demo from our contact page.

Thanks for being here on day one. Let's build this together.