Discover Awesome MCP Servers

Aki

Aki is a local stdio MCP server that gives AI coding assistants durable project memory, allowing them to remember decisions, search prior notes, and save context across sessions.

Custom Search MCP Server

An MCP (Multi-Agent Conversation Protocol) Server that enables interaction with Google's Custom Search API, allowing agents to perform customized web searches through natural language requests.

nowsecure-mcp-server

Enables interaction with NowSecure Platform for listing applications, retrieving remediation findings, and generating remediation PDFs via REST and GraphQL APIs.

ArXiv MCP Server

Enables AI assistants to search arXiv's research repository, download papers, and access their content programmatically. Includes specialized prompts for comprehensive academic paper analysis covering methodology, results, and implications.

mcp-server-ovh

A Model Context Protocol (MCP) server that provides standardized access to OVH API services, enabling AI assistants to manage OVH cloud infrastructure.

ketcher-mcp-server

MCP server for Ketcher chemical structure editor integration, enabling SMILES/MOL/InChI conversion, image generation, molecular property calculation, and validation.

TradingWizard No-Chase MCP

Enforces pre-trade planning discipline by checking entry, stop, target, and risk/reward via a gate tool, and provides proof receipts, terminal links, and risk prompts for AI assistants.

Claude-to-Gemini MCP Server

Enables Claude to use Google Gemini as a secondary AI through MCP for large-scale codebase analysis and complex reasoning tasks. Supports both Gemini Flash and Pro models with specialized functions for general queries and comprehensive code analysis.

mercari-jp-mcp

MCP server to search Mercari Japan listings with query, price range, and exclude keywords.

A Simple MCP Server and Client

Okay, here's a simple example of an MCP (Minecraft Communications Protocol) client and server in Python. This is a very basic illustration and doesn't cover all the complexities of a real MCP implementation. It focuses on establishing a connection, sending a simple message, and receiving a response. **Important Considerations:** * **Security:** This example is *not* secure. It transmits data in plain text. For any real-world application, you'd need to implement proper encryption (e.g., TLS/SSL). * **Error Handling:** The error handling is minimal. A robust implementation would need more comprehensive error checking and recovery. * **MCP Complexity:** Real MCP involves much more complex data structures, authentication, and message types. This is a simplified demonstration. * **Dependencies:** This example uses the standard `socket` library, which is built into Python. **Server (server.py):** ```python import socket HOST = '127.0.0.1' # Standard loopback interface address (localhost) PORT = 25565 # Port to listen on (non-privileged ports are > 1023) with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.bind((HOST, PORT)) s.listen() print(f"Server listening on {HOST}:{PORT}") conn, addr = s.accept() with conn: print(f"Connected by {addr}") while True: data = conn.recv(1024) # Receive up to 1024 bytes if not data: break decoded_data = data.decode('utf-8') print(f"Received: {decoded_data}") # Simple response response = f"Server received: {decoded_data}".encode('utf-8') conn.sendall(response) ``` **Client (client.py):** ```python import socket HOST = '127.0.0.1' # The server's hostname or IP address PORT = 25565 # The port used by the server with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.connect((HOST, PORT)) message = "Hello, MCP Server!".encode('utf-8') s.sendall(message) data = s.recv(1024) print(f"Received: {data.decode('utf-8')}") ``` **How to Run:** 1. **Save:** Save the code as `server.py` and `client.py`. 2. **Run the Server:** Open a terminal or command prompt and run `python server.py`. The server will start listening for connections. 3. **Run the Client:** Open *another* terminal or command prompt and run `python client.py`. The client will connect to the server, send a message, and receive a response. **Explanation:** * **`socket.socket(socket.AF_INET, socket.SOCK_STREAM)`:** Creates a socket object. * `AF_INET`: Specifies the IPv4 address family. * `SOCK_STREAM`: Specifies a TCP socket (reliable, connection-oriented). * **`s.bind((HOST, PORT))` (Server):** Binds the socket to a specific address and port. * **`s.listen()` (Server):** Starts listening for incoming connections. * **`s.accept()` (Server):** Accepts a connection. This blocks until a client connects. It returns a new socket object (`conn`) representing the connection and the client's address (`addr`). * **`s.connect((HOST, PORT))` (Client):** Connects to the server at the specified address and port. * **`conn.recv(1024)` (Server & Client):** Receives data from the socket. The `1024` specifies the maximum number of bytes to receive at once. * **`conn.sendall(message)` (Server & Client):** Sends data to the socket. `sendall` ensures that all data is sent. * **`data.decode('utf-8')`:** Decodes the received bytes into a string using UTF-8 encoding. * **`message.encode('utf-8')`:** Encodes the string into bytes using UTF-8 encoding before sending. **Spanish Translation of Explanation:** * **`socket.socket(socket.AF_INET, socket.SOCK_STREAM)`:** Crea un objeto socket. * `AF_INET`: Especifica la familia de direcciones IPv4. * `SOCK_STREAM`: Especifica un socket TCP (confiable, orientado a la conexión). * **`s.bind((HOST, PORT))` (Servidor):** Vincula el socket a una dirección y puerto específicos. * **`s.listen()` (Servidor):** Comienza a escuchar las conexiones entrantes. * **`s.accept()` (Servidor):** Acepta una conexión. Esto se bloquea hasta que un cliente se conecta. Devuelve un nuevo objeto socket (`conn`) que representa la conexión y la dirección del cliente (`addr`). * **`s.connect((HOST, PORT))` (Cliente):** Se conecta al servidor en la dirección y el puerto especificados. * **`conn.recv(1024)` (Servidor y Cliente):** Recibe datos del socket. El `1024` especifica el número máximo de bytes para recibir a la vez. * **`conn.sendall(message)` (Servidor y Cliente):** Envía datos al socket. `sendall` asegura que todos los datos se envíen. * **`data.decode('utf-8')`:** Decodifica los bytes recibidos en una cadena utilizando la codificación UTF-8. * **`message.encode('utf-8')`:** Codifica la cadena en bytes utilizando la codificación UTF-8 antes de enviar. **Spanish Translation of Important Considerations:** * **Seguridad:** Este ejemplo *no* es seguro. Transmite datos en texto plano. Para cualquier aplicación del mundo real, necesitaría implementar un cifrado adecuado (por ejemplo, TLS/SSL). * **Manejo de errores:** El manejo de errores es mínimo. Una implementación robusta necesitaría una verificación y recuperación de errores más completa. * **Complejidad de MCP:** El MCP real implica estructuras de datos, autenticación y tipos de mensajes mucho más complejos. Esta es una demostración simplificada. * **Dependencias:** Este ejemplo utiliza la biblioteca estándar `socket`, que está integrada en Python. This provides a basic foundation. To build a more realistic MCP implementation, you would need to define specific message formats, handle different message types, implement authentication, and add robust error handling. Remember to prioritize security in any real-world application.

MetaTrader 5 MCP Server

Enables comprehensive access to the MetaTrader 5 trading platform for retrieving market data, managing accounts, and executing trading operations. It provides 32 specialized tools for interacting with MT5 functionalities, including historical OHLCV data access, real-time position monitoring, and automated order management.

codeix

Fast semantic code search for AI agents — find symbols, references, and callers across any codebase.

PlainGov-MCP

Retrieves and explains government program information from official Canadian sources using a strict retrieval-first approach, with deterministic eligibility checks and full source attribution.

Z3 Theorem Prover with Functional Programming

Un servidor MCP para el demostrador de teoremas z3.

brainlayer

Local-first persistent memory layer for AI agents. Provides hybrid search (FTS5 keyword + vector embeddings) over 223K+ knowledge chunks via MCP. Tools: brain_search, brain_store, brain_entity, brain_subscribe. Features pub/sub with stable agent identity, delivery tracking, and Claude --channels integration. SQLite + BrainBar Swift daemon on Unix socket.

yt

An MCP server that provides YouTube data access without API keys or quotas. It enables agents to search videos, retrieve transcripts and metadata, and perform full-text search across cached content for AI context retrieval.

MCP Server Template

A production-ready TypeScript template for building MCP servers with dual transport support (stdio/HTTP), OAuth 2.1 foundations, SQLite caching, observability, and security features including PII sanitization and rate limiting.

ai-ssh-mcp

Enables natural language SSH server management via Claude Code, allowing users to read logs, check services, run commands, and transfer files across multiple servers.

fronius-mcp

Enables real-time solar data from Fronius inverters via Claude, allowing natural language queries about solar production, battery, and grid exchange.

VeoMCP

Google Veo AI video generation with text-to-video, image-to-video, multi-image fusion, 1080p upscaling, and multiple quality/speed models.

Agent-VC MCP Server

Provides AI agents with persistent state, version control, and task management capabilities powered by Fossil SCM. It enables agents to manage files in a sandboxed environment with full commit history and built-in ticket tracking.

mcp-obsidian-vault

Provides AI agents with direct filesystem access to an Obsidian vault for note management, task orchestration, context persistence, and git synchronization.

Ronen Tanchum MCP Server

An MCP server that acts as an artist alter-ego, enabling AI agents to browse Ronen Tanchum's portfolio, ask him questions in his authentic voice, and explore his artworks and philosophy.

Leave MCP Server

Enables LLM clients to handle leave applications by providing tools for initialization, organization selection, leave day calculation, attachment checks, uploads, and submission, with built-in business validation and environment switching.

GSC SEO MCP

Professional Google Search Console MCP server providing 40+ SEO tools for performance analysis, content decay, CTR opportunities, and more, enabling real search data in clients like Cursor and Claude.

local-web-search-service

Self-hosted web search and page fetch service using SearXNG metasearch and Scrapling browser rendering, exposing MCP tools for direct agent use.

mcp-fetch

A minimal MCP server that enables HTTP requests with any method, headers, and body types, supporting large responses through chunked transfers with disk-backed caching.

MCP Notes Connector

Enables integration with Evernote API to access and manage Evernote resources including notes, notebooks, and tags through the Model Context Protocol.

Azure AI Foundry MCP Server

Enables interaction with Azure AI Foundry services for model exploration, deployment, and performance evaluation. It provides tools for managing knowledge bases via AI Search Service, executing fine-tuning jobs, and orchestrating AI agents through natural language.