MCP-server-typescript
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MCP-server-typescriptadd 10 and 20"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP Server - Model Context Protocol Tool Server
A production-ready Model Context Protocol (MCP) server built with Node.js, TypeScript, and Express. This server provides a tool-based execution system with input validation, error handling, and a scalable architecture.
📋 Table of Contents
Related MCP server: MCP Server Foundation Template
🤔 What is MCP?
Model Context Protocol (MCP) is a pattern for enabling AI assistants and other clients to execute backend tools/functions through a standardized API. Think of it as a bridge between AI models and your backend services.
How It Works:
Client (AI assistant, frontend app) sends a request to the MCP server
MCP Server validates the input using schemas
Tool executes the requested function
Response is returned in a standardized format
[AI Assistant] → POST /mcp → [MCP Server] → [Tool Execution] → [Response]✨ Features
✅ TypeScript - Full type safety
✅ Express - Fast and minimal web framework
✅ Zod Validation - Runtime input validation
✅ Error Handling - Comprehensive error management
✅ Logging - Winston logger with file and console output
✅ Security - Helmet for HTTP headers, CORS support
✅ Scalable Architecture - Easy to add new tools
✅ Production Ready - Proper error codes, health checks
📁 Folder Structure
mcp-server/
├── src/
│ ├── tools/ # Tool definitions
│ │ ├── sum.tool.ts # Sum tool implementation
│ │ ├── getUser.tool.ts # GetUser tool implementation
│ │ └── index.ts # Tool registry
│ ├── types/ # TypeScript type definitions
│ │ └── tool.types.ts # Core MCP types
│ ├── middleware/ # Express middleware
│ │ ├── errorHandler.ts # Global error handler
│ │ └── logger.ts # Winston logger configuration
│ ├── utils/ # Utility functions
│ │ └── validator.ts # Input validation helper
│ ├── server.ts # Express app configuration
│ └── index.ts # Server entry point
├── logs/ # Log files (auto-generated)
├── package.json
├── tsconfig.json
├── .env.example
└── README.md🚀 Installation
1. Install Dependencies
npm install2. Set Up Environment Variables
Create a .env file in the root directory:
PORT=3000
NODE_ENV=development
LOG_LEVEL=info3. Build the Project
npm run build4. Start the Server
Development mode (with auto-reload):
npm run devProduction mode:
npm startThe server will start at http://localhost:3000
🎯 Usage
Quick Start Example
Request:
curl -X POST http://localhost:3000/mcp \
-H "Content-Type: application/json" \
-d '{
"tool": "sum",
"input": {
"a": 10,
"b": 25
}
}'Response:
{
"success": true,
"data": {
"result": 35,
"operation": "10 + 25 = 35"
},
"timestamp": "2026-04-24T10:30:00.000Z"
}🔌 API Endpoints
1. Health Check
GET /healthResponse:
{
"status": "healthy",
"timestamp": "2026-04-24T10:30:00.000Z",
"uptime": 123.45
}2. List Available Tools
GET /toolsResponse:
{
"success": true,
"data": {
"count": 2,
"tools": [
{
"name": "sum",
"description": "Adds two numbers together and returns the result"
},
{
"name": "getUser",
"description": "Retrieves a user by their ID from the database"
}
]
},
"timestamp": "2026-04-24T10:30:00.000Z"
}3. Execute Tool (Main MCP Endpoint)
POST /mcp
Content-Type: application/jsonRequest Body:
{
"tool": "toolName",
"input": {
// Tool-specific input
}
}Success Response:
{
"success": true,
"data": {
// Tool-specific output
},
"timestamp": "2026-04-24T10:30:00.000Z"
}Error Response:
{
"success": false,
"error": {
"code": "VALIDATION_ERROR",
"message": "Input validation failed",
"details": {
"errors": [
{
"path": "a",
"message": "Expected number, received string",
"code": "invalid_type"
}
]
}
},
"timestamp": "2026-04-24T10:30:00.000Z"
}🛠️ Example Tools
1. Sum Tool
Adds two numbers together.
Request:
{
"tool": "sum",
"input": {
"a": 15,
"b": 30
}
}Response:
{
"success": true,
"data": {
"result": 45,
"operation": "15 + 30 = 45"
},
"timestamp": "2026-04-24T10:30:00.000Z"
}2. GetUser Tool
Retrieves user information by ID.
Request:
{
"tool": "getUser",
"input": {
"id": 1
}
}Response:
{
"success": true,
"data": {
"id": 1,
"name": "Alice Johnson",
"email": "alice@example.com",
"role": "admin",
"createdAt": "2024-01-15T10:30:00Z"
},
"timestamp": "2026-04-24T10:30:00.000Z"
}User Not Found:
{
"success": true,
"data": null,
"timestamp": "2026-04-24T10:30:00.000Z"
}➕ Adding New Tools
Creating a new tool is simple! Follow these steps:
Step 1: Create Tool File
Create src/tools/myTool.tool.ts:
import { z } from "zod";
import { MCPTool } from "../types/tool.types";
// Define input schema
const myToolInputSchema = z.object({
name: z.string().min(1),
age: z.number().positive(),
});
type MyToolInput = z.infer<typeof myToolInputSchema>;
interface MyToolOutput {
message: string;
}
// Implement the tool
export const myTool: MCPTool<MyToolInput, MyToolOutput> = {
name: "myTool",
description: "Description of what my tool does",
inputSchema: myToolInputSchema,
execute: async (input: MyToolInput): Promise<MyToolOutput> => {
// Your tool logic here
return {
message: `Hello ${input.name}, you are ${input.age} years old!`,
};
},
};Step 2: Register the Tool
Add to src/tools/index.ts:
import { myTool } from './myTool.tool';
constructor() {
this.registerTool(sumTool);
this.registerTool(getUserTool);
this.registerTool(myTool); // ← Add your tool here
}Step 3: Test Your Tool
curl -X POST http://localhost:3000/mcp \
-H "Content-Type: application/json" \
-d '{
"tool": "myTool",
"input": {
"name": "John",
"age": 25
}
}'That's it! Your tool is now available in the MCP server.
🌍 Real-World Use Case
AI Assistant with Backend Integration
Scenario: You're building an AI customer support assistant that needs to access real backend systems.
User: "What's the status of order #12345?"
↓
AI Assistant: [Calls MCP server with "getOrderStatus" tool]
↓
MCP Server: [Validates input, queries database]
↓
AI Assistant: [Receives order data]
↓
Response: "Your order #12345 is currently being shipped and will arrive tomorrow."Benefits:
Separation of Concerns - AI logic separate from business logic
Security - Validate and sanitize all AI requests
Consistency - Standardized API for all AI interactions
Auditability - Log all AI actions and tool executions
Flexibility - Add new capabilities without modifying AI model
Example Tools for Production:
getOrderStatus- Check order informationsearchProducts- Find products in inventorycreateTicket- Create customer support ticketssendEmail- Send automated emailscheckAvailability- Check resource availabilityprocessRefund- Handle refund requests
⚠️ Error Handling
The MCP server uses standardized error codes:
Error Code | HTTP Status | Description |
| 404 | Requested tool doesn't exist |
| 400 | Input validation failed |
| 500 | Tool execution failed |
| 500 | Unexpected server error |
Example Error Response:
{
"success": false,
"error": {
"code": "VALIDATION_ERROR",
"message": "Input validation failed",
"details": {
"errors": [
{
"path": "id",
"message": "Number must be greater than 0",
"code": "too_small"
}
]
}
},
"timestamp": "2026-04-24T10:30:00.000Z"
}📊 Logging
Logs are stored in the logs/ directory:
combined.log- All logserror.log- Error logs only
Console output is colorized for better readability during development.
🔐 Security Features
Helmet - Secures HTTP headers
CORS - Configurable cross-origin requests
Input Validation - Zod schema validation for all inputs
Error Sanitization - Prevents sensitive data leakage
📝 License
MIT
🤝 Contributing
Create a new tool following the patterns in
src/tools/Add comprehensive input validation
Include tests for your tool
Update this README with examples
📞 Support
For issues or questions, please open an issue on the repository.
Happy coding! 🚀
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- Alicense-qualityDmaintenanceA TypeScript implementation of a Model Context Protocol server that provides a frictionless framework for developers to build and deploy AI tools and prompts, focusing on developer experience with zero boilerplate and automatic tool registration.Last updated1,86114MIT
- Flicense-qualityDmaintenanceA customizable, production-ready template for building Model Context Protocol servers with dual transport support (stdio and HTTP), TypeScript, Docker support, and extensible architecture for tools, resources, and prompts.Last updated
- Flicense-qualityDmaintenanceA Model Context Protocol (MCP) server template designed for building structured tools, prompts, and resources with built-in support for HTTP and STDIO transports. It provides a standardized framework for developers to create and deploy AI-driven services using TypeScript and Zod schema validation.Last updated15
- Flicense-qualityBmaintenanceA Model Context Protocol server implementation using Node.js, TypeScript, and Bun, designed for serverless and edge deployments on Vercel and Cloudflare.Last updated1
Related MCP Connectors
A Model Context Protocol server for Wix AI tools
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
MCP (Model Context Protocol) server for Appwrite
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Nayem707/MCP-server-typescript'
If you have feedback or need assistance with the MCP directory API, please join our Discord server