K8scortex Mcp
Operate Kubernetes on GKE, AKS, or EKS from your agent with a large toolkit for RBAC, GitOps, audits, and multi-cluster checks.
Overview
k8scortex-mcp is a MCP server for the Operate phase that exposes 75 Kubernetes tools for GKE, AKS, EKS, RBAC, audit logging, GitOps, and multi-cluster management to AI assistants.
What is this MCP server?
- 75 MCP tools for Kubernetes across GKE, AKS, and EKS
- RBAC, audit logging, GitOps, and multi-cluster scenarios called out explicitly
- npm package k8scortex-mcp v1.0.2 with stdio transport
- Optional KUBECONFIG path; DISABLE_AUTH=true recommended for local trusted use
- 75 Kubernetes MCP tools per server description
- Package version 1.0.2 (npm identifier k8scortex-mcp)
- Documented cloud targets: GKE, AKS, EKS
What problem does it solve?
You context-switch between kubectl, cloud consoles, and chat when an agent could help—but only if it has safe, structured cluster tools.
Who is it for?
Solo devs or tiny teams already on EKS/GKE/AKS who want agent-assisted RBAC reviews, audits, and multi-cluster visibility with local kubeconfig.
Skip if: Builders with no Kubernetes footprint, serverless-only stacks, or production environments where unauthenticated MCP access is unacceptable.
What do I get? / Deliverables
Your agent can run documented Kubernetes MCP operations against your kubeconfig so infra triage and GitOps checks stay in one session.
- Agent-invokable Kubernetes operations across 75 documented tool surfaces
- In-chat RBAC, audit, GitOps, and multi-cluster workflows tied to your real clusters
Recommended MCP Servers
Journey fit
Once you ship containerized workloads, day-two operations—clusters, RBAC, and GitOps—live in the operate phase rather than initial feature coding. Infra subphase covers cluster management, cloud control planes, and platform glue exactly matching this server's 75-tool Kubernetes surface.
How it compares
Broad 75-tool K8s MCP control plane, not a single-purpose helm skill or static infra diagram generator.
Common Questions / FAQ
Who is k8scortex-mcp for?
Developers operating real Kubernetes clusters on major clouds who use MCP agents for day-two infra work and want many cluster actions in one server.
When should I use k8scortex-mcp?
Use it in operate when debugging deployments, reviewing RBAC, following GitOps drift, or comparing resources across clusters without leaving your agent IDE.
How do I add k8scortex-mcp to my agent?
Add the npm stdio server k8scortex-mcp to your MCP config, point KUBECONFIG at your cluster credentials, and set DISABLE_AUTH=true for recommended local use per server metadata.