multicluster-mcp-server

Multicluster MCP Server

Open Cluster Management connects GenAI clients with hub and spoke Kubernetes clusters through MCP. For Kubernetes operators, its tools cover cluster discovery, managed-cluster connections, and running kubectl commands or applying YAML.

Server facts

multicluster-mcp-server

Project

Author
yanmxa
License
Not specified

Tech

Runtime / language
python
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
4
Last commit
Data as of

Overview

Kubernetes work across a hub and its managed clusters is organized around three tools in multicluster-mcp-server. The clusters tool lists managed clusters, also called spoke clusters. After choosing a target, connect_cluster generates its KUBECONFIG and associates the connection with a chosen ClusterRole; cluster-admin is the default role. The kubectl tool accepts either a command or YAML to apply, providing a path from discovering clusters to working with their resources. This supports GenAI clients handling Kubernetes operations through Open Cluster Management rather than treating every cluster as an unrelated target.

The server communicates over stdio and requires kubectl to be installed. It uses the KUBECONFIG environment variable by default, with the configured cluster serving as the hub for access to other clusters. The documented resource retrieval covers both that hub and managed clusters. Reusable OCM prompts and documentation-backed MCP resources remain planned features, not current functionality. Interaction with broader multi-cluster APIs, such as policies and add-ons, is also marked as unavailable.

multicluster-mcp-server setup notes

  • multicluster-mcp-server records stdio only; confirm a network transport upstream before planning remote access.
  • multicluster-mcp-server has no recorded license; confirm reuse terms in its repository.

Related MCP servers