Operate Kubernetes and OpenShift clusters through MCP
Expose Kubernetes and OpenShift cluster operations to MCP clients with native API-backed tools for resources, pods, events, Helm, Tekton, and diagnostics.
npx skills add agentskillexchange/skills --skill operate-kubernetes-and-openshift-clusters-through-mcp
Use Kubernetes MCP Server when an agent needs controlled access to inspect or operate a Kubernetes or OpenShift cluster from an MCP client. The operator workflow is to attach the server to a client, point it at an approved kubeconfig or in-cluster configuration, and ask for bounded cluster actions such as listing resources, reading pod logs, checking events, managing Helm releases, or starting Tekton runs. The boundary is cluster operations through the documented MCP tool surface; this should be configured with least-privilege credentials and is not a generic Kubernetes tutorial or a blanket approval for destructive cluster changes.
What this skill actually does
Inputs and prerequisites: Access to a Kubernetes or OpenShift cluster, kubeconfig or in-cluster configuration, MCP-compatible client, optional npm, Python, Docker, or native binary install path.
Setup notes: For npm-based clients, add an MCP server command such as npx -y kubernetes-mcp-server@latest. For other environments, use the upstream native binary, Python package, container image, or VS Code/Cursor install links, then configure the server with an approved kubeconfig or in-cluster identity.
Source and verification boundary: use https://github.com/containers/kubernetes-mcp-server as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a MCP workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.