Discover Awesome MCP Servers

HowToCook-MCP Server

Enables AI assistants to recommend recipes, plan weekly meals, and solve the "what to eat today" problem by providing access to a comprehensive Chinese recipe database with smart filtering by categories, allergies, and dietary restrictions.

mcp-agent-opt

Provides code-aware context compression by stripping comments, docstrings, and whitespace while maintaining full logic fidelity for AI agents. It features tools for architectural mapping, symbol searching, and token-budgeted multi-file reading.

ascript-mcp

Enables AI programming tools to query AScript API documentation and control real Android/iOS devices for automation scripting, including device interaction, development deployment, and file management.

Viraly MCP Server

A thin MCP proxy that connects LLM clients like Claude and ChatGPT to Viraly's platform API, enabling AI agents to perform actions via natural language.

ORKL MCP Server

Máy chủ MCP cho Thư viện Thông tin Tình báo về Mối đe dọa ORKL

TypeScript MCP Server Boilerplate

A boilerplate project for quickly developing Model Context Protocol (MCP) servers using TypeScript SDK, with example implementations of calculator and greeting tools, plus resource management capabilities.

PDF Knowledgebase MCP Server

A Model Context Protocol server that enables intelligent document search and retrieval from PDF collections, providing semantic search capabilities powered by OpenAI embeddings and ChromaDB vector storage.

mcp-github

Gitbub mcp

raydium-launchlab-mcp

raydium-launchlab-mcp

zarq-risk-intelligence

Real-time crypto risk scoring for AI agents. Trust Score, crash probability, and distance-to-default for 205 tokens. Free, no API key needed.

sentinel-mcp

An MCP server for AI-powered API testing that enables automated positive, negative, and security testing directly from AI chat interfaces. It supports multiple AI providers and generates detailed security reports.

Symbol Blockchain MCP Server (REST API tools)

Máy chủ MCP Blockchain Symbol. (Công cụ REST API)

Physics MCP Server

Enables physicists to perform computer algebra calculations, create scientific plots, solve differential equations, work with tensor algebra and quantum mechanics, and parse natural language physics problems. Supports unit conversion, physical constants, and generates comprehensive reports with optional GPU acceleration.

SendGrid MCP Server

Enables comprehensive email marketing and transactional email operations through SendGrid's API v3. Supports contact management, campaign creation, email automation, list management, and email sending with built-in read-only safety mode.

CData Sync MCP Server

Enables AI assistants to manage CData Sync operations, including data synchronization jobs, connections, and ETL processes through stdio or HTTP transports. It provides tools for executing jobs, monitoring real-time progress via Server-Sent Events, and handling comprehensive workspace configurations.

Gemini Audio MCP

Gemini Audio MCP is a high-performance Model Context Protocol (MCP) server that leverages the power of the Gemini 2.0 Multimodal Live API to generate high-fidelity, environmental soundscapes on-demand.

Image Processor MCP Server

Enables optimization, conversion to WebP, and uploading of images to Vercel Blob storage, supporting both local files and external URLs.

Google Drive MCP Server

This server enables multi-agent conversations that interact with the Google Drive API, allowing agents to perform file operations, manage permissions, and handle Google Drive content through a standardized protocol.

MCP 만들면서 원리 파헤쳐보기

Okay, here's a breakdown of a server and client implementation for a system conceptually similar to the MCP (Master Control Program), along with considerations for a Vietnamese audience. I'll focus on the core concepts and provide examples using Python, a widely accessible language. Remember that a *true* MCP is a complex operating system kernel, so this is a simplified, illustrative example. **Conceptual Overview** The MCP, in its essence, is a central control point. In a modern context, we can think of it as a central server managing resources, tasks, and communication between multiple clients. * **Server (MCP):** * Listens for client connections. * Authenticates clients (optional, but important for security). * Manages a queue of tasks or requests. * Allocates resources (e.g., processing time, memory, data access). * Monitors system health and performance. * Provides a central logging and auditing facility. * Can enforce security policies. * **Client:** * Connects to the server. * Authenticates itself (if required). * Submits tasks or requests to the server. * Receives results or status updates from the server. * Handles errors and retries. **Python Implementation (Simplified)** This example uses Python's `socket` library for basic network communication and `threading` for handling multiple clients concurrently. **1. Server (mcp_server.py):** ```python import socket import threading import time HOST = '127.0.0.1' # Standard loopback interface address (localhost) PORT = 65432 # Port to listen on (non-privileged ports are > 1023) def handle_client(conn, addr): """Handles communication with a single client.""" print(f"Connected by {addr}") try: while True: data = conn.recv(1024) # Receive data from the client if not data: break # Client disconnected message = data.decode('utf-8') print(f"Received from {addr}: {message}") # **Simulated Task Processing** time.sleep(2) # Simulate some work being done response = f"MCP: Task '{message}' processed successfully." conn.sendall(response.encode('utf-8')) except ConnectionResetError: print(f"Client {addr} disconnected abruptly.") finally: conn.close() print(f"Connection with {addr} closed.") def main(): with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.bind((HOST, PORT)) s.listen() print(f"MCP Server listening on {HOST}:{PORT}") while True: conn, addr = s.accept() # Accept incoming connections thread = threading.Thread(target=handle_client, args=(conn, addr)) thread.start() print(f"Active Connections: {threading.active_count() - 1}") #Subtract main thread if __name__ == "__main__": main() ``` **2. Client (mcp_client.py):** ```python import socket HOST = '127.0.0.1' # The server's hostname or IP address PORT = 65432 # The port used by the server def main(): with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: try: s.connect((HOST, PORT)) message = input("Enter a task for the MCP: ") s.sendall(message.encode('utf-8')) data = s.recv(1024) print(f"Received: {data.decode('utf-8')}") except ConnectionRefusedError: print("Connection to the MCP server refused. Is the server running?") except Exception as e: print(f"An error occurred: {e}") if __name__ == "__main__": main() ``` **How to Run:** 1. Save the code as `mcp_server.py` and `mcp_client.py`. 2. Open a terminal or command prompt. 3. Run the server: `python mcp_server.py` 4. Open another terminal or command prompt. 5. Run the client: `python mcp_client.py` 6. The client will prompt you to enter a task. Type something and press Enter. 7. Observe the output on both the client and server consoles. **Explanation:** * **Sockets:** The `socket` library provides the foundation for network communication. We create sockets for both the server and the client. * **Server:** * `socket.socket(socket.AF_INET, socket.SOCK_STREAM)`: Creates a TCP socket (reliable, connection-oriented). * `s.bind((HOST, PORT))`: Binds the socket to a specific IP address and port. * `s.listen()`: Starts listening for incoming connections. * `s.accept()`: Accepts a connection from a client, creating a new socket (`conn`) for communication with that client. * `threading.Thread`: Creates a new thread to handle each client concurrently. This allows the server to handle multiple clients simultaneously. * `conn.recv(1024)`: Receives data from the client (up to 1024 bytes at a time). * `conn.sendall(response.encode('utf-8'))`: Sends data back to the client. * `conn.close()`: Closes the connection with the client. * **Client:** * `s.connect((HOST, PORT))`: Connects to the server at the specified IP address and port. * `s.sendall(message.encode('utf-8'))`: Sends data to the server. * `s.recv(1024)`: Receives data from the server. * **Encoding:** `encode('utf-8')` and `decode('utf-8')` are used to convert strings to bytes (for sending over the network) and bytes back to strings. UTF-8 is a common character encoding that supports a wide range of characters, including Vietnamese. * **Error Handling:** The `try...except` blocks handle potential errors like connection refused or abrupt disconnections. **Improvements and Considerations (Quan Trọng):** * **Authentication:** Implement a proper authentication mechanism (e.g., username/password, API keys, certificates) to prevent unauthorized access to the MCP. This is *crucial* for security. * **Task Queue:** Use a proper task queue (e.g., `queue.Queue` in Python, or a more robust message queue like RabbitMQ or Redis) to manage tasks efficiently. This allows the server to handle tasks asynchronously and prevent blocking. * **Resource Management:** Implement resource allocation and monitoring. The MCP should track resource usage (CPU, memory, disk I/O) and prevent clients from consuming excessive resources. * **Error Handling:** Implement more robust error handling and logging. The MCP should log all errors and exceptions to a file or database for debugging and auditing. * **Security:** Address security vulnerabilities such as injection attacks, denial-of-service attacks, and data breaches. Use secure coding practices and regularly update your software. * **Scalability:** Consider scalability requirements. If you need to handle a large number of clients, you may need to use a more scalable architecture, such as a distributed system with multiple MCP servers. * **Data Serialization:** Instead of sending plain text, use a data serialization format like JSON or Protocol Buffers to send structured data between the client and server. This makes it easier to handle complex data structures. * **Vietnamese Language Support:** * Ensure that your code uses UTF-8 encoding throughout to properly handle Vietnamese characters. * Consider translating any user-facing messages (e.g., error messages, prompts) into Vietnamese for a better user experience. * If you are storing data in a database, make sure that the database is configured to use UTF-8 encoding. **Vietnamese Translation of Key Concepts:** * **Server:** Máy chủ * **Client:** Máy khách * **Master Control Program (MCP):** Chương trình Điều khiển Chính * **Task:** Nhiệm vụ * **Resource:** Tài nguyên * **Connection:** Kết nối * **Authentication:** Xác thực * **Error:** Lỗi * **Message:** Tin nhắn * **Queue:** Hàng đợi **Example of Vietnamese Error Message:** ```python print("Lỗi: Không thể kết nối đến máy chủ MCP. Máy chủ có đang chạy không?") ``` **Important Considerations for Vietnamese Developers:** * **Character Encoding:** Always use UTF-8 encoding for all text files and data. * **Localization:** Consider localizing your application for Vietnamese users, including translating the user interface and providing support for Vietnamese date and time formats. * **Network Infrastructure:** Be aware of the network infrastructure in Vietnam, which may have different characteristics than in other countries (e.g., slower internet speeds, higher latency). This provides a basic framework. Building a real-world MCP-like system requires significantly more effort and expertise. Remember to prioritize security, scalability, and maintainability. Good luck!

Filmladder MCP Server

Provides movie listings, showtimes, and personalized recommendations for Amsterdam cinemas by scraping filmladder.nl, with support for filtering by date, cinema, rating, and preferred showtimes.

BlindPay MCP Server

Enables AI assistants to interact with BlindPay's stablecoin payment infrastructure, allowing users to create receivers, process payouts and payins across multiple blockchains, manage virtual accounts and wallets, and configure payment operations through natural language.

M/M/1 and M/M/c Queue Simulation Server

Enables simulation and analysis of M/M/1 and M/M/c queuing systems using SimPy, with tools for parameter validation, theoretical metric calculation, simulation execution, and comparison of separate vs pooled queue strategies.

OpenZeppelin Contracts MCP Server

A Model Context Protocol (MCP) server that allows AI agents to generate smart contracts using OpenZeppelin Contracts libraries.

TypeScript MCP Server Boilerplate

A boilerplate project for quickly developing Model Context Protocol (MCP) servers using TypeScript SDK, with example implementations of tools (calculator, greeting) and resources (server info).

Cloudflare Remote MCP Server Template

Enables the deployment of remote Model Context Protocol servers on Cloudflare Workers without authentication. It allows users to host custom tools and connect them to AI clients using Server-Sent Events (SSE).

Commerce-MCP

Enables natural-language control of e-commerce operations including product management, order processing, inventory tracking, customer service, content generation, and advertising analytics through 15 integrated MCP tools. Provides a local-first commerce automation solution with SQLite storage and extensible channel adapters for end-to-end online store workflows.

Composer Trade MCP

Enables AI assistants to create, backtest, and manage automated investment strategies (symphonies) on Composer, including searching through 1000+ existing strategies, monitoring portfolio performance, and executing trades with live market data.