protein-bars-mcp-server
If you are the rightful owner of protein-bars-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
This project implements a serverless Model Context Protocol (MCP) server on AWS Lambda for interacting with a protein bar ordering system.
Serverless MCP Server: Protein Bar Ordering System
This project implements a serverless Model Context Protocol (MCP) server on AWS Lambda that enables AI assistants to interact with a protein bar ordering system.
Features
- MCP Server with tools for:
- Listing available protein bars
- Creating new orders
- Admin functionality for managing orders
- Serverless Architecture using:
- AWS Lambda with Express and Lambda Web Adapter
- API Gateway for HTTP endpoint
- DynamoDB for data storage
- Stateless Design that scales efficiently
Prerequisites
- Node.js 22 or higher
- AWS CLI configured with appropriate permissions
- AWS CDK installed (
npm install -g aws-cdk
)
Setup and Deployment
1. Install Dependencies
npm install
2. Build the Project
npm run build
3. Deploy to AWS
npm run cdk bootstrap # Only needed first time
npm run deploy
The CDK deployment will output:
- The API Gateway URL
- The MCP Server URL (used for client configuration)
4. Seed Initial Data
After deployment, seed the DynamoDB table with initial protein bar data:
# Set your AWS_PROFILE if needed
export PRODUCTS_TABLE=protein_products # Should match the table name in CDK stack
npm run seed-data
Testing Locally
To run the MCP server locally for testing:
npm run dev
This will start the server on port 3000, and you can send MCP requests to http://localhost:3000/mcp
.
Connecting Clients
VS Code (Copilot Agent Mode)
- Enable GitHub Copilot Chat and Agent Mode in VS Code
- Create a
.vscode/mcp.json
file with:
{
"servers": {
"ProteinBarMCP": {
"type": "http",
"url": "https://your-api-id.execute-api.your-region.amazonaws.com/prod/mcp"
}
}
}
- Reload VS Code and start a conversation with GitHub Copilot
- You can now use the protein bar tools in your AI interactions
Claude Desktop
As of 2025, Claude Desktop has limited support for remote MCP servers, but you can:
- Check for the latest Claude updates that might support direct remote connections
- Alternatively, use a local proxy that forwards requests to your AWS MCP server
Security Considerations
For a production deployment, you should add:
- API Key authentication for both public and admin endpoints
- IAM roles with least privilege for the Lambda function
- VPC configuration if needed
- Proper error handling and logging
License
MIT