Rabobank MCP-Server Training

Welcome to the Rabobank MCP-Server Training repository. This project contains a fully functional demo Model Context Protocol (MCP) server built with TypeScript that you can use as a starting point during the training.

What is MCP?

The Model Context Protocol is an open standard that allows AI assistants (like GitHub Copilot and Claude) to call external tools and data sources in a structured, secure way. An MCP server exposes tools, resources, and prompts that any MCP-compatible client can discover and invoke.

┌──────────────┐   MCP (stdio/SSE)   ┌─────────────────────┐
│  AI Client   │ ◄──────────────────► │  Demo MCP Server    │
│ (Copilot /   │                      │  (this repository)  │
│  Claude …)   │                      └─────────────────────┘
└──────────────┘

Related MCP server: FastMCP Demo Server

Demo tools included

Prerequisites

• Node.js v18 or higher

• npm (comes with Node.js)

Quick start

# 1. Clone the repository
git clone https://github.com/RemseyMailjard/rabobank-mcp-server-training.git
cd rabobank-mcp-server-training

# 2. Install dependencies
npm install

# 3. Build
npm run build

# 4. Run (stdio)
node build/index.js

Use with VS Code (GitHub Copilot)

The .vscode/mcp.json file is already configured. Open the project in VS Code and Copilot will automatically discover the server.

Use with Claude for Desktop

Add the following entry to your claude_desktop_config.json:

{
  "mcpServers": {
    "rabobank-demo": {
      "command": "node",
      "args": ["C:\\path\\to\\rabobank-mcp-server-training\\build\\index.js"]
    }
  }
}

Project structure

rabobank-mcp-server-training/
├── src/
│   └── index.ts          # MCP server implementation
├── build/                # Compiled output (after npm run build)
├── .vscode/
│   └── mcp.json          # VS Code MCP configuration
├── .github/
│   └── copilot-instructions.md
├── package.json
├── tsconfig.json
└── README.md

Exercises

During the training you will:

1. Explore the existing tools in src/index.ts and understand how they are registered.

2. Add a new tool – e.g. get_interest_rate that returns savings account interest rates.

3. Add a resource – expose a static JSON file with product information.

4. Add a prompt – create a reusable prompt template for a customer service scenario.

5. Connect the server to GitHub Copilot Chat or Claude for Desktop and test the tools interactively.

Useful links

• MCP Documentation

• TypeScript SDK

• MCP Inspector – GUI for testing MCP servers

• Claude for Desktop