# SolKube

> SolKube is a native desktop Kubernetes client. Live dashboards, a command palette, YAML with dry runs, pod shells, logs and port forwards, at a fraction of the memory of web-based clients.

SolKube is a desktop Kubernetes client written in Rust on NSH, which draws with AppKit on macOS and GTK 4 on Linux. Builds are published for macOS on Apple silicon. It has no account and no telemetry; it talks to API servers directly with the user's kubeconfig and cloud CLIs.

## Features

- A dashboard that moves: Every cluster opens on live CPU, memory, pods and nodes. Charts glide to new values every few seconds and numbers roll like an odometer, so a change catches your eye without a redraw.
- One palette for every action: Open a cluster or a resource list, restart or scale a workload, or jump to a pod's logs or shell without leaving the keyboard. Anything that changes the cluster asks first.
- Tables built for triage: Resource lists follow the cluster through watch streams. Filter by status, pick rows for bulk actions, and read live CPU and memory from metrics-server right in the table.
- Details, events and YAML together: Open any object to see its fields, its latest events with warnings in amber, and its YAML. Shell, logs and port forwards are one click from a pod.
- Edit with a dry run first: Edit YAML in a tab, dry run it against the API server and read the diff before you save. Create from templates with server-side apply, custom resources included.
- kubeconfig, EKS, AKS and GKE discovery
- Live watch streams, polling fallback
- Helm releases: values, history, rollback
- Pod shell on alacritty_terminal
- Port forwarding with a live list
- Node cordon, uncordon and drain
- Every CRD, with its printer columns
- Health dots for every cluster

## Pages

- [Benchmarks](https://solkube.com/benchmarks): launch time, memory and CPU on real Macs, with the method. Raw data: https://solkube.com/benchmarks.json
- [Release notes](https://solkube.com/release-notes): every change, release by release
- [Demo](https://solkube.com/demo): a tour of the app
- [Blog](https://solkube.com/blog): engineering notes. Each post is also served as Markdown at /blog/<slug>.md

## Blog posts

- [Native views, half the memory: what moving SolKube to NSH changed](https://solkube.com/blog/native-views-half-the-memory.md): SolKube moved from a GPU-drawn UI to the platform's own views. On a MacBook Air it now rests at 25 MB instead of 50 MB and idles at a sixth of the CPU. Here are the numbers and why they moved.
- [How we benchmark a desktop app](https://solkube.com/blog/how-we-benchmark-a-desktop-app.md): Launch time, memory footprint and CPU, measured the same way on every build. The method behind SolKube's benchmark numbers, and the mistakes it is designed to avoid.

## FAQ

### What is SolKube?

SolKube is a desktop Kubernetes client, in the same family as Lens, Freelens and k9s. It is written in Rust and draws with each platform's own views (AppKit on macOS), so it opens in about a quarter of a second and uses around 25 MB of memory at rest.

### Does SolKube send my cluster data anywhere?

No. SolKube talks to your API servers straight from your Mac with your own kubeconfig and cloud CLIs. There is no SolKube account and no telemetry. The only other request it makes is the update check to downloads.solkube.com.

### Which clusters does it work with?

Any cluster your kubeconfig reaches, including exec-plugin auth. SolKube can also find your Amazon EKS, Azure AKS and Google GKE clusters through the aws, az and gcloud CLIs you already use.

### How much does SolKube cost?

Pricing isn't final yet. We'll publish it here before the first public release.

### Is there a Linux or Windows version?

SolKube builds and runs on Linux with GTK 4 today, and our end-to-end tests drive it there. Published builds start with macOS on Apple silicon.

### Can I try it in the browser?

Not yet. A browser build is planned, and when it lands the demo page will run the real app against a sample cluster.

Full text of every post and release note: https://solkube.com/llms-full.txt