🤖 MCP Server

This repository contains the Model Context Protocol (MCP) Server used by AI Agents. It is built using the @modelcontextprotocol/sdk, which allows integration with various AI models. The project uses Node.js and TypeScript to provide a robust implementation of the MCP protocol.

The MCP (Model Context Protocol) is a protocol that enables standardized communication between applications and different AI models, making integration easier and more consistent.

🚀 Getting started

To start using the MCP Server, follow these steps:

Prerequisites

- Node.js version 18 or higher
- npm

Installation and build

Install dependencies and build the application:

npm install && npm run build

⚙️ MCP Server Configuration

To configure the MCP Server, you'll need to set up some environment variables. Use the .env.example file provided in the project as a reference. Copy it to a new .env file in the project root and adjust the settings as needed:

cp .env.example .env

Make sure to update the values in your .env file to match your specific environment and requirements.

🔌 Integrating with GitHub Copilot

Configuration in VS Code

The project already includes a pre-configured file for GitHub Copilot integration located in the .vscode/mcp.json folder. This is the recommended per-project configuration.

For external projects that want to use this MCP server, copy the following file to the .vscode folder of your project:

"mcp-server": {
  "type": "stdio",
  "command": "node",
  "args": ["${workspaceFolder}/build/index.js"],
  "envFile": "${workspaceFolder}/.env"
}

Alternatively, you can configure Copilot globally in VS Code settings:

1. Open VS Code Settings (Ctrl+,)
2. Search for "Copilot: Plugin"
3. Click "Edit in settings.json"
4. Add the following configuration:

"github.copilot.plugins": [
  {
    "name": "MCP",
    "command": "npm run start",
    "cwd": "/path/to/your/project/mcp"
  }
]

Replace /path/to/your/project/mcp with the absolute path to your MCP project.

Troubleshooting Common Issues

- If Copilot does not recognize your plugin, try restarting VS Code.
- Ensure the MCP SDK version is up to date.
- Make sure the server is running before attempting to use the tools in Copilot.

🧰 Available Tools

This project comes with example tools:

- get-alerts: Fetches weather alerts for a U.S. state.
- get-forecast: Retrieves weather forecasts for specific coordinates.

📝 Testing with GitHub Copilot

After starting the server with npm run start and configuring VS Code, you can test your tools by asking Copilot to use them:

- "Use the get-forecast tool to get the weather forecast for Detroit."
- "Fetch weather alerts for California using the get-alerts tool."

🛠️ Creating New Tools

To create a new tool:

1. Create a new file in the folder src/tools/[tool-name]/[tool-name].tool.ts
2. Use the following structure as a template:

import { z } from "zod";
import server from "../../utils/server.config.js";
server.addTool(

  "tool-name",

  "Tool description",

  {

    parameter1: z.string().describe("Description of parameter 1"),

    parameter2: z.number().describe("Description of parameter 2"),

  },

  async ({ parameter1, parameter2 }) => {

    // Tool logic here
return {

      content: [

        {

          type: "text",

          text: "Tool result",

        },

      ],

    };

  }

);

3. Test the tool by calling it through the MCP Server API or using the provided tools.

📦 Creating New Resources

To create a new resource:

1. Create a new file in the folder src/resources/[resource-name]/[resource-name].resource.ts
2. Use the following structure as a template:

import { z } from "zod";
import server from "../../utils/server.config.js";
server.addResource(

  "resource-name",

  "Resource description",

  {

    parameter1: z.string().describe("Description of parameter 1"),

    parameter2: z.number().describe("Description of parameter 2"),

  },

  async ({ parameter1, parameter2 }) => {

    // Resource logic here
return {

      data: {

        key: "value", // Replace with actual resource data

      },

    };

  }

);

3. Test the resource by calling it through the MCP Server API or using the provided tools.

📓 Our conventions

Commit types / Branch prefixes

Here is the list of commit types and branch prefixes you can choose from :

| Commit Type | Title | Description |
| ----------- | --------------------------- | :----------------------------------------------------------------------------------------------------- |
| feat | ✨ Features | New feature |
| fix | 🐛 Bug Fixes | Bug fix |
| docs | 📚 Documentation | Documentation only changes |
| style | 💎 Styles | Changes that do not affect the meaning of the code (white-space, formatting, missing semi-colons, etc) |
| refactor | 📦 Code Refactoring | A code change that neither fixes a bug nor adds a feature |
| perf | 🚀 Performance Improvements | A code change that improves performance |
| test | 🚨 Tests | Adding missing tests or correcting existing tests |
| build | 🛠 Builds | Changes that affect the build system or external dependencies (example scopes: next, webpack, pnpm) |
| ci | ⚙️ Continuous Integrations | Changes to our CI configuration files and scripts |
| chore | ♻️ Chores | Other changes that don't modify src or test files |
| revert | 🗑 Reverts | Reverts a previous commit |

Branch names

See the regex validating the branch naming on husky branch-naming script.

Structure :

<type>/<short-description>

Examples :

feat/signup-page
ci/github-actions-setup
fix/calendar-input-focus-not-working

Commit messages

Structure :

<type>(optional scope): <description>

Examples :

ci(front): setup storybook tests
fix: send cors headers
feat(groups): add comment section