MCP.so
ログイン

EKS MCP Server

@mahmoudsalah84

基本情報

カテゴリ

その他

ライセンス

MIT

トランスポート

stdio

作者

mahmoudsalah84

EKS MCP Server について

概要はまだありません

EKS MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "eks-mcp-server-new": {
      "command": "python",
      "args": [
        "main.py"
      ]
    }
  }
}

EKS MCP Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is EKS Model Context Protocol (MCP) Server?

A lightweight server implementing the Model Context Protocol for EKS operations, providing a standardized interface for GenAI agents to interact with EKS clusters. It includes comprehensive EKS and Kubernetes resource management capabilities.

How to use EKS Model Context Protocol (MCP) Server?

Run locally with Python (pip install -r requirements.txt, then python main.py) or with Docker (docker build -t eks-mcp-server . && docker run -d -p 8000:8000 --name mcp-server eks-mcp-server). Docker Compose (docker-compose up -d) and ECS deployment scripts are also provided. Tests are available via test_*.py scripts.

Key features of EKS Model Context Protocol (MCP) Server

  • Fast response times with proper timeout handling
  • Comprehensive EKS and Kubernetes operations support
  • Multiple authentication methods for EKS clusters
  • Robust error handling and detailed logging
  • Docker containerization for easy deployment
  • ECS deployment support with client using Amazon Bedrock

Use cases of EKS Model Context Protocol (MCP) Server

  • GenAI agents listing EKS clusters and describing nodegroups
  • Retrieving pod logs, deployments, and services from a cluster
  • Managing Kubernetes resources across namespaces
  • Automating cluster inspection and troubleshooting via MCP

よくある質問

1

What operations does the server support?

It supports EKS operations (list/describe clusters, nodegroups) and Kubernetes operations (list pods, deployments, services, get pod logs, etc.) via POST to `/mcp/v1/query`.

2

How does authentication work?

The server supports AWS SDK authentication, kubectl with generated kubeconfig, and direct Kubernetes API calls. The current implementation uses the kubectl method.

3

What are the prerequisites?

Docker (for containerized deployment) or Python 3.8+ (for local development), AWS credentials for production use, and ECS task role added to the EKS cluster’s aws-auth ConfigMap when deployed on ECS.

4

How is the server deployed?

Deploy locally with Python/Docker, or to ECS using scripts in `ecs-deployment/`. The server is already deployed to ECS at `http://3.90.45.69:8000`.

5

Is there a client included?

Yes, a client implementation using Amazon Bedrock for conversational interaction is provided in the `client/` directory.

コメント

「その他」の他のコンテンツ