Kubernetes MCP Server is an open-source Model Context Protocol server that lets AI assistants interact directly with Kubernetes and OpenShift clusters. Connect it to your favourite MCP client — Claude, ChatGPT, Cursor and others — and the assistant can inspect resources, check workloads and help operate your clusters through natural language. The project has 2,174 stars on GitHub.
Built by the containers community, this MCP server speaks to the Kubernetes API natively rather than shelling out to kubectl, which makes interactions faster and more reliable. Point it at a cluster and your AI assistant gains structured tools for exploring namespaces, pods and other resources, troubleshooting deployments, and performing day-2 operations — all gated by your existing kubeconfig credentials and RBAC permissions, so the assistant can only do what you are allowed to do.
Because it speaks the Model Context Protocol, the same server works across the growing ecosystem of MCP clients, and it can be run locally or deployed in-cluster depending on your setup. It is part of the broader containers community effort — the same ecosystem behind Podman and Buildah — which shows in its focus on straightforward, composable tooling rather than a monolithic platform.
Kubernetes MCP Server is completely free and open source under the Apache-2.0 license. You only need a cluster to point it at and an MCP-compatible AI client.
This tool is for DevOps engineers, SREs and developers who already use AI assistants and want them to help with cluster operations. It is especially handy for exploration and troubleshooting, where asking your assistant beats memorising long kubectl command chains.