# kubectl-mcp-server

An MCP server that lets an AI assistant manage a Kubernetes cluster in natural language, covering pod debugging, cost optimization, network diagnostics, security auditing and Helm chart management.

kubectl-mcp-server is a Model Context Protocol server that exposes Kubernetes cluster operations to an AI assistant so a user can debug crashed pods, deploy applications, audit RBAC permissions and secrets, manage Helm charts, and view dashboards through natural-language requests instead of memorized kubectl commands. It is listed in the CNCF Landscape.

It ships 253 tools, 8 workflow prompts and 8 data resources covering crash diagnosis with logs and events, deployment with best-practice defaults, AI-assisted cost optimization recommendations, network connectivity tracing from ingress to pods, and interactive HTML dashboards with live metrics and event timelines.

It installs with npx for zero-install use, with npm install -g, or with pip, including an optional [ui] extra for interactive dashboards, and supports in-cluster deployment and multi-cluster configurations.

Website: https://github.com/rohitg00/kubectl-mcp-server