Control Kubernetes infrastructure through natural-language MCP workflows
Let MCP-compatible agents inspect, debug, deploy, audit, and manage Kubernetes clusters through a controlled kubectl-backed server.
npx skills add agentskillexchange/skills --skill control-kubernetes-infrastructure-through-natural-language-mcp-workflows
Use kubectl-mcp-server when an operator wants an agent to help with Kubernetes work from inside an MCP-compatible client instead of switching between chat, kubectl, Helm, dashboards, and runbooks. The agent can inspect pods and services, debug failed workloads, manage Helm-related operations, audit cluster state, and guide deployment or cost-optimization tasks through a Kubernetes-specific MCP surface. The boundary is live Kubernetes operations through this server, not a generic Kubernetes tutorial, DevOps platform, or broad infrastructure SDK.
What this skill actually does
Inputs and prerequisites: Node.js or Python, kubectl, Kubernetes cluster access, MCP-compatible client.
Setup notes: Run with npx -y kubectl-mcp-server, install globally with npm install -g kubectl-mcp-server, or install the Python package with pip install kubectl-mcp-server[ui], then register the server with an MCP-compatible client that has access to the intended kubeconfig.
Source and verification boundary: use https://github.com/rohitg00/kubectl-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.