Kubernetes MCP servers
MCP servers that connect an AI agent to Kubernetes: pods, deployments and cluster state via the Kubernetes API. We have classified 73 so far, listed below and ranked by their independent trust score.
73 catalogued · how we score
Kubernetes at a glance
- Highest score
- 96 io.github.dockndevai/mcp-kubernetes
- Median score
- 46
- Scored
- 73 of 73
- Rescanned in 7 days
- 72
npm 25 · oci 24 · pypi 15 · remote 15 · mcpb 6 · figures as of 2026-11-05 02:39 UTC · how we score
All 73 by trust score Trust score 0–100
- npm 96
io.github.dockndevai/mcp-kubernetes
Kubernetes monitoring & ops for AI agents — safe-by-default access modes and guards.
- oci · pypi 96
io.github.ocm-mcp-server/ocm-mcp-server
Guardrailed fleet ops for AI agents: multi-cluster Kubernetes via OCM with policy, approval, audit.
- npm 94
Cloud DevOps analysis with guarded Git, infrastructure, cloud and observability operations.
- npm 90
io.github.rafaelgaspar/longhorn-mcp
MCP server for the Longhorn distributed storage manager API, with a --read-only mode.
- remote 89
io.github.karandaid/karandeepsingh-blog
Search and read Karandeep Singh's DevOps and AWS blog: articles and cheatsheets.
- mcpb · npm 88
Architecture diagrams with official AWS, Azure, GCP and Kubernetes icons, to SVG, PNG and draw.io
- remote 84
Free browser-based calculators and analyzers for cloud cost, DevOps, security, and data governance.
- npm 84
26 deterministic, offline pre-action safety gates for AI agents, each with a signed verdict.
- npm · oci 81
io.github.apatilgtn/k8scortex-mcp
Kubernetes MCP server: 75 tools for GKE, AKS, EKS, RBAC, audit logging, GitOps, multi-cluster.
- npm 80
Generate and edit AWS, Azure, GCP and Kubernetes architecture diagrams as editable draw.io files.
- npm 79
MCP server and CLI that resolves config/env-var cascade across .env, docker-compose, and Kubernetes
- pypi 78
MCP server for natural-language Kubernetes operations across single or multi-cluster environments.
- npm 78
io.github.jmshinhwa/cron-schedule-lint
Names every schedule line in the file you have open that fires at the wrong hour, twice, or never at
- npm 78
io.github.jmshinhwa/k8s-removed-api-lint
15 checks for the apiVersions kubectl no longer serves, on the manifest open in your editor
- npm 77
io.github.jmshinhwa/cloud-cost-landmine-lint
Names every line in the file you have open that starts a recurring cloud charge, with the published
- npm 77
io.github.jmshinhwa/cloud-manifest-audit-kit
For platform engineers on AKS: 28 rules naming every removed apiVersion, literal credential and :lat
- npm 77
Read-only Model Context Protocol MCP server enabling code-driven AI analysis of Kubernetes clusters.
- npm 76
io.github.jmshinhwa/ingress-nginx-retirement-lint
Names the Gateway API field that replaces each retired ingress-nginx annotation — or tells you there
- oci · pypi 76
io.github.SCGIS-Wales/helm-mcp
The full Helm CLI surface as 46 MCP tools, backed by the Helm v3 and v4 Go SDKs.
- npm 75
io.github.NotHarshhaa/devops-mcp
Unified MCP server for Kubernetes, ArgoCD, Prometheus, PagerDuty, and Loki
- oci · pypi 75
io.github.sandeepbazar/ocm-mcp-server
Guardrailed fleet ops for AI agents: multi-cluster Kubernetes via OCM with policy, approval, audit.
- npm 74
Structured JSON wrappers for git, kubectl, terraform, helm, argocd, gh, glab and other CLIs.
- npm · oci 74
OpenShift/Kubernetes cluster management via the oc CLI for MCP clients
- remote 73
Provides read access to your GKE and Kubernetes resources.
- remote 65
Kubernetes kind token, manifest discarded
Kubernetes kind token, manifest discarded
- oci 60
AI-powered Kubernetes diagnostics, RBAC analysis, security checks, and app deployment via MCP
- mcpb 52
io.github.bryanjbelanger/talos-mcp-server
Talos Linux cluster operations as one thin MCP tool over the official machine API.
- pypi 50
Governed Kubernetes ops — 55 MCP tools with audit, budget, undo, risk-tier audit labels.
- remote 50
Causal reasoning for reliability: root cause, blast radius, and service health for agents.
- npm · oci · pypi 50
io.github.containers/kubernetes-mcp-server
A Model Context Protocol (MCP) server for Kubernetes and OpenShift
- oci · pypi 50
Secure Kubernetes MCP server: scoped, redacted diagnostics and human-approved writes
- npm 49
io.github.dockndevai/mcp-openshift
Safe-by-default MCP for OpenShift / Kubernetes: projects, pods, logs, deployments, routes.
- npm 49
Operate self-hosted OpenClaw and Hermes agent fleets on Docker or Kubernetes from MCP clients.
- npm 48
Dynamic pod spawner & proxy for ephemeral AI agent workspaces on Kubernetes without CRDs
- remote · oci 48
Manage Rackspace Spot Kubernetes Cloudspaces, node pools, and VMs from your AI assistant.
- oci · pypi 48
io.github.sophotechlabs/kube-lint-mcp
Validate Kubernetes manifests, Helm charts, Kustomize overlays, and FluxCD resources
- oci 46
Inspect Kairon VMs, network policy and VM-edge flows from an agent. Read-only by default.
- oci 45
io.github.basphere/ratatosk-mcp
CNCF release intelligence: typed changes from release notes; check_stack compares versions locally.
- oci 45
io.github.garlicKim21/ratatosk-mcp
CNCF release intelligence: typed changes from release notes; check_stack compares versions locally.
- oci 44
eBPF network observability and safe emergency control for Linux and Kubernetes.
- pypi 40
Capability-aware MCP server for Rancher-managed Kubernetes: discovery, resources, workflows.
- npm 39
Troubleshoot Kubernetes agentically: natural-language cluster diagnosis via ask_kubently
- pypi 39
VMware AVI (NSX ALB) load balancer plus AKO Kubernetes ops — 28 MCP tools.
- remote 36
Read-only server insight: why a Linux server or K8s cluster is slow, failing or changed.
- npm 36
AI-powered development platform for Kubernetes deployments and intelligent automation
- pypi 34
io.github.Michael-WhiteCapData/WhiteCapData-Dev
Operate a k3s / Kubernetes cluster from your AI agent — safe-by-default MCP server.
- remote 34
kube-linter audit for Kubernetes manifests — 63 checks: security, availability, RBAC, network.
- pypi 34
VMware AVI (NSX ALB) load balancer plus AKO Kubernetes ops — 29 MCP tools.
- remote 33
Deploy and operate Kubernetes: clusters, stacks, apps, logs, metrics and security, with OAuth.
- pypi 32
vSphere with Tanzu (VKS): Namespace and TanzuKubernetesCluster lifecycle. Requires vSphere 8.x+.
- npm 31
TypeScript MCP server for AI-powered containerization workflows with Docker and Kubernetes support
- remote 31
Manage Kubernetes clusters, deployments, databases, secrets and observability on Mengi Cloud.
- pypi 24
AI-powered SRE observability for Kubernetes/OpenShift with 40+ Tekton debugging tools
- remote 23
Give your AI assistant access to real Helm chart data. No more hallucinated values.yaml files.
- pypi 21
MCP server for Rancher API integration
- npm 21
io.github.oijusti/k8s-port-forward-mcp
Kubernetes Port Forward — MCP Server: port-forward Kubernetes services with terminal integration
- mcpb 20
Kubernetes port forwarding for local development with automatic /etc/hosts entries.
- oci 10
Kubernetes MCP server with RBAC-style, context-scoped guardrails for AI agents.
- oci 10
SSH & Kubernetes access broker for AI agents; the model never touches a credential.
- oci 10
Read-only Kubernetes diagnostics: pods, logs, events, workloads, services, PVCs and nodes.
- oci 10
Read-only access to a Crossplane control plane: resources, claims, packages, compositions.
- mcpb 9
Safe Kubernetes access for AI agents via MCP. Read-only by default, with explicit permission modes.
- oci 9
MimiOps mcp server for Kubernetes
MimiOps is an opinionated MCP server for Kubernetes
- oci 7
SDK for creating, managing, and operating Kubernetes clusters and workloads with ease.
- oci 7
Kubernetes MCP server for diagnosis, resource management, GitOps, and RBAC-enforced operations.
- oci 6
Runs a command in a throwaway Kubernetes pod and returns exit code, output and artifacts.
- oci 6
K8gentS — Kubernetes RCA MCP Server
Kubernetes RCA agent that sandboxes its own AI with OPA Gatekeeper. MCP-ready for any client.
- remote · oci 6
io.github.kubeshop/testkube-mcp
MCP server for Testkube - Manage test workflows, executions, and artifacts via AI assistants
- mcpb 5
Real-time Kubernetes network traffic visibility and API analysis for HTTP, gRPC, Redis, Kafka, DNS.
- mcpb 5
Connect to Signadot to manage ephemeral environments and route traffic to local services.
- remote 0
Willform Agent dead
Deploy containers on Kubernetes with x402 billing. 9 workload types and source builds.
- remote 0
Ramen: self-hosted MCP for teams
Multi-zone MCP server for GKE/EKS/AKS: Python tools from git, NeMo guardrails, SSO, autoscaling.
- remote 0
Deploy apps and manage Kubernetes on AWS, GCP, Azure, and On-Premise with natural language.
Every server here has been scored. A server is listed here because its description and tool definitions say it is built for Kubernetes. That is what the server declares about itself, not an integration we have tested, so check the score breakdown before you trust one. This classification currently runs only over servers that publish a hosted endpoint, so a Kubernetes server distributed solely as an installable package will not appear here yet, however clearly it belongs. Treat the ranking as a partial field rather than the whole one.
Common questions
What is a Kubernetes MCP server?
A Kubernetes MCP server exposes pods, deployments and cluster state via the Kubernetes API to an AI assistant over the Model Context Protocol, so the assistant can work with Kubernetes directly instead of you moving data in and out by hand. These come from the public MCP registry, and being listed there is not a review of the code, so the publisher is worth checking as closely as the feature list.
How do I get started with a Kubernetes MCP server?
Start from the ranked list on this page, then open a server's own page and read its score breakdown before you install anything. Every server listed here is a hosted endpoint, so that breakdown covers Endpoint Security, Transport & Reachability, Schema Quality & AI Usability, Stability & Change Management, Tool Coverage, Tool Safety and Capabilities. A Kubernetes server acts on your Kubernetes data, so check what access it asks for before you connect it. The same page carries install snippets for the common MCP clients.