Model Context Protocol (MCP) Server

by MCP-Mirror

288 downloads
Not rated
GitHub

About

This repository contains the Kubernetes deployment configuration for the MCP server on Azure Kubernetes Service (AKS). It provides a containerized implementation of the Model Context Protocol, deployable using Docker and Azure Container Registry.

Details

Author
MCP-Mirror
Downloads
288
Categories
Other

- Exposes a /health endpoint for Kubernetes health checks.
- Supports scaling with 3 default replicas.
- Uses HTTPS, CORS, and Helmet.js for security.
- Rate limiting is implemented to prevent abuse.
- Monitored via Azure Monitor and kubectl logs.
- Runs on Node.js 18+ with TypeScript source code.

Setting up with Highlight

This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:

  1. Download and install Highlight from highlightai.com/download
  2. Navigate to the plugins tab and select "Add Custom Plugin"
  3. Configure the plugin with the settings below
    Plugin Name Model Context Protocol (MCP) Server
    Command (node, npx, python, etc.)

    Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.

  4. Enable "Start Automatically" if you want the plugin to start when Highlight launches

From the repository

Build the Docker image, push it to your Azure Container Registry, then apply the Kubernetes manifests (k8s/) to deploy. Configure the server through environment variables defined in the ConfigMap. Verify with kubectl get pods and kubectl get services.

Claude Desktop / Cursor

Paste into your MCP client config file to install this server.

{
    "mcpServers": {
        "model context protocol (mcp) server": {
            "adrianozagallo-home_mcp-server": {
                "command": "docker",
                "args": [
                    "build",
                    "-t",
                    "mcp-server:latest",
                    "."
                ]
            }
        }
    }
}

McpServers

{
    "adrianozagallo-home_mcp-server": {
        "command": "docker",
        "args": [
            "build",
            "-t",
            "mcp-server:latest",
            "."
        ]
    }
}

Model Context Protocol (MCP) Server

This repository contains the Kubernetes deployment configuration for the MCP server on Azure Kubernetes Service (AKS).

Prerequisites

- Azure CLI
- kubectl
- Docker
- Node.js 18+

Project Structure

.
├── k8s/
│   ├── deployment.yaml    # Kubernetes deployment configuration
│   ├── service.yaml      # Kubernetes service configuration
│   └── configmap.yaml    # Kubernetes configmap for environment variables
├── src/                  # Source code directory
├── Dockerfile           # Container build configuration
├── package.json         # Node.js dependencies
└── tsconfig.json        # TypeScript configuration

Deployment Steps

1. Build the Docker image:

docker build -t mcp-server:latest .

2. Push the image to Azure Container Registry (ACR):

az acr login --name <your-acr-name>
docker tag mcp-server:latest <your-acr-name>.azurecr.io/mcp-server:latest
docker push <your-acr-name>.azurecr.io/mcp-server:latest

3. Apply Kubernetes manifests:

kubectl apply -f k8s/

4. Verify deployment:

kubectl get pods
kubectl get services

Configuration

The application can be configured through environment variables defined in the ConfigMap (k8s/configmap.yaml).

Health Checks

The application exposes a /health endpoint for Kubernetes health checks.

Monitoring

- Resource usage can be monitored through Azure Monitor
- Application logs are available through kubectl logs

Security

- The application uses HTTPS
- CORS is configured for secure cross-origin requests
- Rate limiting is implemented to prevent abuse
- Helmet.js is used for security headers

Scaling

The deployment is configured with 3 replicas by default. You can scale up or down using:

kubectl scale deployment mcp-server --replicas=<number>
No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.