badenson/eks-mcp-ai-devops

EKS MCP Server AI-Driven DevOps Project

This repository automates the provisioning of an Amazon EKS cluster, deploys the AWS EKS MCP Server, and demonstrates AI-assisted development and troubleshooting using Cline in VS Code with AWS Bedrock.

Special Thanks

References & Credits Original Demo Video: Watch on YouTube - https://www.youtube.com/watch?v=bIJGVn2SElk

Special thanks to Cloud With Raj for the inspiration and detailed walkthrough in the original video.

Features

• Infrastructure as Code: Uses Terraform to provision AWS EKS, VPC, and supporting resources.
• Kubernetes Manifests: Sample YAML for deploying the MCP server and a demo app.
• AI Integration: VS Code + Cline extension configured for AWS Bedrock LLMs.
• Python Examples: Scripts for troubleshooting and interacting with the MCP server.
• Local Development: Optimized for VS Code workflows.

---

Prerequisites

• AWS CLI configured with credentials and region ([see guide][6])
• Docker
• Terraform >= 1.0
• kubectl
• Python 3.8+
• VS Code with Cline extension
• AWS Bedrock access (for LLM integration)

---

Repository Structure

• README.md
• terraform/
  • main.tf
  • variables.tf
  • outputs.tf
• k8s/
  • deployment.yaml
  • service.yaml
  • namespace.yaml
  • configmap.yaml
• scripts/
  • deploy.sh
  • test-deployment.sh
• mcp-server/
  • server.py
  • Dockerfile
  • requirements.txt
• python-examples/
  • troubleshooting_example.py
• .vscode/
  • settings.json
• cline-config/
  • cline-settings.json

Quick Start

1. Clone the Repository

git clone cd

---

2. Set Environment Variables

export AWS_REGION=us-east-1 export AWS_ACCOUNT_ID= export CLUSTER_NAME=mcp-eks-cluster

3. Provision EKS Cluster with Terraform

cd terraform terraform init terraform apply

This will create the VPC, EKS cluster, node groups, and supporting IAM roles[2][9].

4. Build and Deploy MCP Server

cd ../mcp-server docker build -t mcp-server:latest .

Push to ECR if needed, or use locally for testing cd ../scripts ./deploy.sh all # Or phase by phase: infrastructure, app, etc.

5. Deploy Kubernetes Manifests

kubectl apply -f ../k8s/namespace.yaml kubectl apply -f ../k8s/deployment.yaml kubectl apply -f ../k8s/service.yaml

6. Configure Cline in VS Code

• Install the Cline extension from the VS Code Marketplace[4][6].
• In Cline settings, select AWS Bedrock as your API provider.
• Enter your AWS credentials and preferred region.
• Select your desired Bedrock model (e.g., Claude 3 Sonnet)[6].
• Test with a simple prompt to confirm integration.

7. AI-Assisted Development & Troubleshooting

• Use Cline to interact with your codebase, generate manifests, and troubleshoot deployments.
• Example: Ask Cline to check pod status, generate YAML, or debug deployment issues using natural language[1][7][9].

---

Example Python Troubleshooting Script

python-examples/troubleshooting_example.py import boto3

def list_foundation_models(): bedrock_client = boto3.client(service_name="bedrock") response = bedrock_client.list_foundation_models() for model in response["modelSummaries"]: print(f"Model: {model['modelName']}")

if name == "main": list_foundation_models()

See [AWS Bedrock Python examples][5][8] for more.

---

Project Structure

---

Troubleshooting Scenarios

• Use Cline to list pods, check deployment status, and fetch logs via the MCP server.
• Example prompt:

  "Cline, show me all pods in the mcp-server namespace and their status."

• If a deployment fails, ask Cline for recommended fixes or to regenerate YAML[1][7][9].

---

References

• [AWS EKS MCP Server documentation][7]
• [Terraform EKS Example][2]
• [Cline + AWS Bedrock Integration Guide][6]
• [AWS Bedrock Python Examples][5][8]

---

License

MIT