If you manage Kubernetes clusters, you've made peace with a tradeoff.
Terminal clients like K9s are fast but sacrifice visibility. Visual dashboards like Lens are slow to launch, heavy on resources, and disconnected from everything else in your stack. And neither was built for teams.
Today we're open-sourcing KubeStudio, a Rust-native Kubernetes cluster management dashboard that eliminates that tradeoff and adds something neither category of tool has ever offered: a direct connection to Strike48's autonomous agent layer.
Every dev and ops team we talk to manages clusters with some combination of Lens, K9s, and raw kubectl. The tools don't talk to each other. They don't share state across a team. And they don't connect to anything downstream.
The result is a workflow held together by individual installs, manually shared kubeconfig files, and a lot of context-switching. It works until your team grows past a handful of people, or until you want to automate the routine work that eats hours every week.
KubeStudio is built on a single Rust/Dioxus codebase. It launches in under 500ms, supports 20+ resource types, and streams live updates through the Kubernetes watch API.
Everything practitioners expect from a cluster management tool works out of the box: pod log streaming, multi-container shell access, port forwarding, deployment scaling, and job triggering. The command palette, keyboard shortcuts, and resource aliasing keep you moving at terminal speed. The visual interface gives you the context a terminal can't.
KubeStudio also auto-discovers Custom Resource Definitions, so you're not limited to built-in resource types.
This is where KubeStudio breaks from everything else on the market.
Deploy a shared instance with configurable permission sets. Your entire dev org manages clusters from one place. Junior developers get view-only access to staging. DevOps engineers get deploy permissions across production and staging. Platform leads get admin access to everything.
No more passing around kubeconfig files. No more juggling individual Lens installs across the team. One deployment, team-wide access, proper controls.
KubeStudio's Matrix connector mode plugs directly into Strike48's Prospector Studio. This gives autonomous agents the ability to inspect cluster state, execute kubectl commands, and take action through tool-calling.
Your Kubernetes environment becomes another operational surface that Strike48's agents can see, reason about, and act on, alongside your logs, your network data, and everything else in the platform.
The practical implication: point an agent at a workflow you've been doing manually and it works. Restart a deployment when health checks fail. Scale pods based on queue depth. Monitor rollout progress and alert on failures. No custom scripting, no glue code.
Routine Kubernetes operations become autonomous workflows. Your team focuses on the work that actually requires human judgment.
The same Rust/Dioxus codebase produces three deployment targets:
A native desktop app for individual use with no Electron overhead and no browser lag. A standalone web server for shared team instances with permission controls. And a Matrix connector variant that bridges KubeStudio into AI-enhanced environments through Strike48.
Choose the deployment model that fits your infrastructure. Run all three if different teams have different needs.
KubeStudio is available today under an open-source license on GitHub. We built it because the Kubernetes management tools that exist today were designed for individual practitioners in a pre-agentic world. We're open-sourcing it because Kubernetes tooling should be fast, shareable, and extensible.
Inspect the codebase. Fork it. Extend it. Contribute back.
KubeStudio is available now on GitHub. Learn more here: https://www.strike48.com/platform/kubestudio
If you're running Lens today and want something faster, lighter, and team-ready, give it a try.