Discover Awesome MCP Servers

Minimal MCP Server for AutoCAD DWG

A FastAPI-based MCP server scaffold designed to extract material tables from AutoCAD DWG files. It enables users to automate CAD data processing by configuring drawing folder paths and layer-specific extraction parameters.

HubSpot MCP Server

Enables AI clients to seamlessly take HubSpot actions and interact with HubSpot data, allowing users to create/update CRM records, manage associations, and gain insights through natural language.

Rec-MCP

A Model Context Protocol server that enables searching for camping facilities and recreational areas using Recreation.gov's API and Google Maps Geocoding.

Pipedrive MCP Server

Enables LLMs like Claude to interact with Pipedrive CRM data through a standardized interface, providing 22 tools for managing deals, persons, organizations, activities, and more.

mcp-reunion

MCP server providing 96 tools across 21 modules for La Réunion (French overseas region) open data: economy, demographics, geography, transport, health, education, elections, tourism, housing, and environment. Powered by data.regionreunion.com and data.gouv.fr.

Metabase AI Assistant

Enables AI-powered interaction with Metabase instances and PostgreSQL databases through natural language. Creates models, SQL queries, metrics, and dashboards using both Metabase API and direct database connections.

TickTick MCP

Enables AI assistants to manage TickTick tasks through natural language, including creating, updating, completing, and deleting tasks, as well as managing projects.

SD Elements MCP Server

A Model Context Protocol server that provides SD Elements API integration, enabling LLMs to interact with SD Elements security development lifecycle platform.

postgres-mcp

Enables Claude to interact with a local PostgreSQL database through SQL queries, schema inspection, and CRUD operations.

CVE Checker for Node Modules

Enables checking npm packages for known security vulnerabilities using the OSV API before installation. Supports both single package checks and bulk vulnerability scanning for multiple packages at once.

Meshy MCP Server

Enables AI assistants to generate 3D models, textures, and images via the Meshy API using natural language, supporting text-to-3D, image-to-3D, remeshing, retexturing, rigging, and animation workflows.

MCP Registry Server

Enables searching and retrieving detailed information about MCP servers from the official MCP registry. Provides tools to list servers with filtering options and get comprehensive details about specific servers.

Simple MCP Server

Claro, aquí tienes un ejemplo minimalista de cómo construir un servidor MCP (Message Center Protocol) en Python, usando la biblioteca `asyncio`: ```python import asyncio async def handle_client(reader, writer): """Maneja la conexión de un cliente.""" addr = writer.get_extra_info('peername') print(f"Conexión de {addr}") try: while True: data = await reader.readline() # Lee una línea del cliente if not data: break # Cliente se desconectó message = data.decode().strip() # Decodifica y elimina espacios print(f"Recibido de {addr}: {message}") response = f"Servidor: Recibido '{message}'\n".encode() # Crea una respuesta writer.write(response) # Envía la respuesta al cliente await writer.drain() # Asegura que los datos se envíen except Exception as e: print(f"Error con {addr}: {e}") finally: print(f"Cerrando conexión con {addr}") writer.close() await writer.wait_closed() async def main(): """Función principal para iniciar el servidor.""" server = await asyncio.start_server( handle_client, '127.0.0.1', 8888 # Escucha en localhost:8888 ) addr = server.sockets[0].getsockname() print(f"Servidor escuchando en {addr}") async with server: await server.serve_forever() # Mantiene el servidor en ejecución if __name__ == "__main__": asyncio.run(main()) ``` **Explicación:** 1. **`handle_client(reader, writer)`:** - Esta función maneja la conexión individual de cada cliente. - `reader` y `writer` son objetos `StreamReader` y `StreamWriter` de `asyncio`, que permiten leer y escribir datos asíncronamente. - Lee líneas del cliente usando `reader.readline()`. - Decodifica el mensaje recibido (`data.decode().strip()`). - Imprime el mensaje recibido en la consola del servidor. - Crea una respuesta simple y la envía de vuelta al cliente usando `writer.write()`. - `writer.drain()` asegura que los datos se envíen al cliente. - Maneja excepciones y cierra la conexión cuando el cliente se desconecta o hay un error. 2. **`main()`:** - Esta función es la principal para iniciar el servidor. - `asyncio.start_server()` crea el servidor y lo asocia con la función `handle_client` para manejar las conexiones. - El servidor escucha en la dirección IP `127.0.0.1` (localhost) y el puerto `8888`. - `server.serve_forever()` mantiene el servidor en ejecución indefinidamente, esperando nuevas conexiones. 3. **`if __name__ == "__main__":`:** - Asegura que la función `main()` se ejecute solo cuando el script se ejecuta directamente (no cuando se importa como un módulo). - `asyncio.run(main())` ejecuta la función `main()` dentro de un bucle de eventos `asyncio`. **Cómo ejecutarlo:** 1. Guarda el código como un archivo Python (por ejemplo, `mcp_server.py`). 2. Abre una terminal y ejecuta el script: `python mcp_server.py` **Cómo probarlo:** Puedes usar `telnet` o `netcat` para conectarte al servidor: ```bash telnet 127.0.0.1 8888 ``` O: ```bash nc 127.0.0.1 8888 ``` Luego, escribe un mensaje y presiona Enter. Verás la respuesta del servidor. **Puntos clave:** * **Asíncrono:** Usa `asyncio` para manejar múltiples conexiones concurrentemente sin bloquear el hilo principal. * **Minimalista:** El código es lo más simple posible para demostrar la estructura básica de un servidor MCP. * **Líneas:** Este ejemplo asume que los mensajes se envían como líneas (terminadas con un salto de línea). Este es un punto de partida. Para un servidor MCP más robusto, necesitarías: * **Definir un protocolo MCP específico:** Especificar el formato de los mensajes, los tipos de datos, etc. * **Manejo de errores más robusto:** Manejar diferentes tipos de errores y desconexiones de manera más elegante. * **Autenticación y autorización:** Implementar mecanismos para verificar la identidad de los clientes y controlar su acceso a los recursos. * **Escalabilidad:** Considerar cómo escalar el servidor para manejar un gran número de conexiones. * **Logging:** Implementar un sistema de registro para rastrear la actividad del servidor y diagnosticar problemas. Espero que esto te sea útil. Si tienes alguna pregunta, no dudes en preguntar.

Nextcloud MCP Server

Enables LLMs to interact with Nextcloud instances through 30 tools across Notes, Calendar, Contacts, Tables, and WebDAV file operations, featuring a powerful unified search system for finding files without exact paths.

stata-mcp

Controls the Stata GUI through Windows Stata Automation COM, enabling do-file management, command execution, and data analysis within natural language workflows.

Security Scanner MCP Server

Enables comprehensive security scanning of code repositories to detect secrets, vulnerabilities, dependency issues, and configuration problems. Provides real-time security checks and best practice recommendations to help developers identify and prevent security issues.

elimu-mcp

MCP server for Kenya education — school registry, KCSE/KCPE results, HELB student loans, TVET programs, literacy resources.

Quick Data for Windows MCP

A Windows-optimized server providing universal data analytics for JSON and CSV files through over 32 tools including schema discovery and interactive visualizations. It is specifically designed for seamless integration with Claude Desktop on Windows.

ibapi-mcp-server

This project creates a middleware server that connects to Interactive Brokers Gateway and exposes its functionality through the FastMCP framework. This allows LLMs like Claude to interact with your IB account, retrieve portfolio information, and potentially execute trades.

Markdown2PDF MCP Server

Converts Markdown documents to PDF files with support for syntax highlighting, custom styling, Mermaid diagrams, optional page numbers, and configurable watermarks.

MDict MCP Server

A server that provides access to MDict dictionaries through the Model Context Protocol, enabling word lookups, searches, and dictionary management.

MCP-Server-Filesystem

An enhanced Model Context Protocol filesystem server that fixes common limitations and adds features like glob pattern search, head/tail reading, intelligent file editing with syntax validation, and new tools for delete and command execution.

MCP Hub

MCP Hub is a self-hosted AI operations platform that provides a unified MCP gateway with semantic tool routing, persistent vector memory, automation, and multi-agent flows. It enables connecting any MCP client to 130+ tools across 12 integrations through just 3 hub endpoints.

Social Media Analytics

Social media analytics toolkit that analyzes profiles, scores engagement, detects trending topics, researches hashtags, generates content calendars, and benchmarks against competitors. 6 tools across all major platforms.

V.I.S.O.R.

Smart MCP server that reads your code, understands file connections, and provides precise context to AI coding assistants, reducing token usage and preventing errors from outdated or irrelevant files.

Simple Python MCP Server

A simple MCP server implementation in Python using FastMCP and FastAPI, providing streamable HTTP transport with optional OAuth authorization for tools like echo and math.

testrail-mcp

An MCP server for TestRail that prepares everything needed to generate test cases from Jira tickets, Confluence pages, or free-form specs and pushes them back to TestRail.

readdown-mcp

Converts web pages and HTML strings into clean, LLM-optimized Markdown with metadata extraction and token estimation. It uses a lightweight, browserless approach to provide token-efficient output for more effective LLM processing.

shipstatic

MCP server for Shipstatic — deploy and manage static sites from AI agents. Works with Claude Code, Cursor, VS Code Copilot, and any MCP-compatible client.