kubectl-ai

Why we included this project

kubectl-ai sits between you and your cluster: type a request in plain English, like "how's nginx doing in my cluster" or "scale this deployment," and it works out which kubectl commands get you there, showing you what it plans to run before it runs anything. That transparency is what makes it useful to experienced engineers who want to move faster during routine cluster work, and to people still learning their way around namespaces and resource state. You can install it as a kubectl plugin through krew or as a standalone binary, so it fits into the tooling your team already uses rather than asking you to adopt a separate platform. The MCP client and server modes extend that further, letting you plug the same assistant into other agent-based tooling.