Multi-Cluster MCP server

Multi-Cluster MCP server is an MCP server that provides an interface for Generative AI systems to interact with multiple Kubernetes clusters using Open Cluster Management. The server connects directly to a designated hub cluster via standard Kubernetes configurations and facilitates resource inspection across connected managed clusters. Platform engineers, DevOps teams, and site reliability specialists use this server to inspect live infrastructure, troubleshoot workload errors across fleets, and retrieve cluster health telemetry without manually switching context profiles. By exposing Kubernetes tools via the Model Context Protocol, the server allows AI assistants to retrieve resources from the hub cluster, inspect workloads on managed clusters, connect using specific ClusterRole permissions, and examine metrics, logs, and alerts across multiple environments. The tool treats the active cluster configuration as the primary control plane, allowing AI clients to navigate fleet-wide operational state consistently.

Category: Cloud & Infrastructure

Tags: cluster-management, infrastructure, k8s, kubernetes

Visit Multi-Cluster MCP server

How to install and configure Multi-Cluster MCP server

1. Ensure you have kubectl installed on your machine and set up with access to your primary hub cluster. 2. Confirm your KUBECONFIG environment variable points to your valid cluster configuration file. 3. Open your MCP client configuration file, such as claude_desktop_config.json. 4. Add the following entry to the mcpServers object: json { "mcpServers": { "multicluster-mcp-server": { "command": "npx", "args": [ "-y", "multicluster-mcp-server@latest" ] } } } 5. Save the configuration and restart your MCP client to enable multi-cluster operations.

What you can do with Multi-Cluster MCP server

• Querying Kubernetes workloads, deployments, and pods running across managed clusters directly from an AI prompt - Fetching resources and status configurations from the primary hub cluster using current context settings - Connecting to a target managed cluster with specific ClusterRole configurations for secure inspection - Retrieving runtime metrics, operational logs, and active alerts across multiple integrated Kubernetes clusters - Auditing fleet-wide configuration drift and workload states across Open Cluster Management environments using natural language

Key facts

• https://github.com/yanmxa/multicluster-mcp-server
• Cloud & Infrastructure, DevOps, CI/CD & Version Control
• cluster-management, infrastructure, k8s, kubernetes

Part of MCP Servers

How do I install Multi-Cluster MCP server?

You can install and run Multi-Cluster MCP server using npx. Add multicluster-mcp-server@latest under your client mcpServers configuration with the npx command, ensure kubectl is installed, and verify your KUBECONFIG environment variable is set.

What can Multi-Cluster MCP server do?

Multi-Cluster MCP server allows AI agents to interact with Kubernetes clusters via Open Cluster Management. It can fetch resources from the hub cluster and managed clusters, assume specific ClusterRoles, and retrieve cluster metrics, logs, and alerts.

How does Multi-Cluster MCP server authenticate with Kubernetes?

The server automatically relies on your local kubectl installation and your KUBECONFIG environment variable. In multi-cluster architectures, it connects to your configured cluster as the hub cluster and communicates with managed clusters through it.

Is Multi-Cluster MCP server open source?

Yes, Multi-Cluster MCP server is open source software published under the MIT License on GitHub.