Discover Awesome MCP Servers

Extend your agent with 84,517 capabilities via MCP servers.

All84,517
Odoo MCP Gateway

Odoo MCP Gateway

Enables managing multiple Odoo instances from a single panel and exposes MCP tools for interacting with them via a unified service.

mcp-server-saxo

mcp-server-saxo

Enables interaction with Saxo Bank OpenAPI for trading, portfolio management, market data, and account operations through natural language.

Databricks MCP Server

Databricks MCP Server

Enables interaction with Databricks for running SQL queries, managing clusters, triggering jobs, running notebooks, exploring Unity Catalog metadata, managing secrets, and interacting with DBFS.

Git-Fabric Gateway

Git-Fabric Gateway

A central tool aggregator that routes and orchestrates multiple fabric apps through a single MCP surface. It enables users to access a combined suite of specialized tools from various registered apps via a unified server connection.

blender-mcp

blender-mcp

Connects Blender 3D modeling software with Claude AI, enabling natural language control of 3D modeling and scene manipulation tasks.

MCP Server

MCP Server

This MCP server provides access to resources and prompts from GitHub repositories or the local filesystem, enabling teams to share coding standards, documentation, and reusable prompts with AI tools like Claude.

Indonesia Financial Data MCP

Indonesia Financial Data MCP

Citation-first MCP server for official Indonesian financial data from IDX, BPS, and KSEI, providing tools to access company profiles, financial reports, statistical tables, and investor demographics.

402-mcp

402-mcp

L402 + x402 client MCP. AI agents discover, pay for, and consume any payment-gated API autonomously. Supports Lightning (NWC), Cashu ecash, stablecoins, and human-in-the-loop payments.

Knock MCP Server

Knock MCP Server

Enables AI assistants to manage notification infrastructure—workflows, channels, templates, users, and more—via Knock's API with OAuth authentication.

Kodama MCP

Kodama MCP

AST-aware code exploration MCP server for AI agents, optimized for token efficiency.

Dropbox MCP Server by CData

Dropbox MCP Server by CData

This read-only MCP Server allows you to connect to Dropbox data from Claude Desktop through CData JDBC Drivers. Free (beta) read/write servers available at https://www.cdata.com/solutions/mcp

chaoslimba-mcp-server

chaoslimba-mcp-server

MCP server for ChaosLimba — provides read-only access to the Romanian language learning platform's database via the Model Context Protocol.

MCP Middleware Adapter for Express Servers

MCP Middleware Adapter for Express Servers

Okay, here's a breakdown of how you can run multiple Minecraft Protocol (MCP) clients on a NodeJS Express server, acting as an adapter or middleware. This is a more complex setup, so I'll break it down into key concepts and provide code snippets to illustrate the ideas. **Conceptual Overview** 1. **Minecraft Protocol (MCP) Clients:** You'll need a library that handles the Minecraft protocol. Popular choices include: * `minecraft-protocol`: A low-level, flexible library. Requires more manual handling of packets. * `mineflayer`: A higher-level library built on top of `minecraft-protocol`. Provides a bot-like API for interacting with the server. 2. **NodeJS Express Server:** This will be your central hub. It will: * Receive requests from your front-end or other clients. * Manage the lifecycle of multiple MCP client instances. * Route commands/data to the appropriate MCP client. * Receive data from the MCP clients (e.g., chat messages, player positions) and relay it back to the requesting client. 3. **Client Management:** You'll need a way to keep track of your MCP clients. A simple object or array can work, but consider using a more robust data structure if you need advanced features like searching or filtering. 4. **Routing:** You'll need a mechanism to determine which MCP client a request should be sent to. This could be based on: * A unique ID assigned to each client. * The Minecraft server address the client is connected to. * Some other application-specific logic. 5. **Asynchronous Operations:** MCP communication is inherently asynchronous. You'll need to use `async/await` or Promises to handle the asynchronous nature of the protocol. **Code Example (Illustrative)** This example uses `mineflayer` for simplicity, but the core concepts apply to `minecraft-protocol` as well. ```javascript const express = require('express'); const mineflayer = require('mineflayer'); const app = express(); const port = 3000; app.use(express.json()); // For parsing JSON request bodies // Store MCP clients in an object, keyed by a unique ID const minecraftClients = {}; let nextClientId = 1; // Simple ID generator // Function to create a new Minecraft client async function createMinecraftClient(host, port, username, password) { try { const bot = mineflayer.createBot({ host: host, port: port, username: username, password: password, // Optional version: "1.20.1" // Replace with the correct version }); bot.on('login', () => { console.log(`Bot ${bot.username} logged in to ${host}:${port}`); }); bot.on('kicked', (reason, loggedIn) => { console.log(`Bot ${bot.username} was kicked: ${reason}, loggedIn: ${loggedIn}`); // Handle reconnection or cleanup here }); bot.on('error', (err) => { console.error(`Bot ${bot.username} error: ${err}`); // Handle errors appropriately }); bot.on('chat', (username, message) => { if (username === bot.username) return; // Ignore self-messages console.log(`<${username}> ${message}`); // Broadcast the message to connected clients (e.g., via WebSocket) }); return bot; // Return the bot instance } catch (error) { console.error("Error creating bot:", error); throw error; // Re-throw to be caught by the route handler } } // Route to create a new Minecraft client app.post('/clients', async (req, res) => { const { host, port, username, password } = req.body; try { const bot = await createMinecraftClient(host, port, username, password); const clientId = nextClientId++; minecraftClients[clientId] = bot; res.status(201).json({ message: 'Client created', clientId: clientId }); } catch (error) { console.error("Error in /clients route:", error); res.status(500).json({ error: 'Failed to create client' }); } }); // Route to send a command to a specific Minecraft client app.post('/clients/:clientId/command', (req, res) => { const clientId = parseInt(req.params.clientId); const command = req.body.command; const bot = minecraftClients[clientId]; if (!bot) { return res.status(404).json({ error: 'Client not found' }); } try { bot.chat(command); res.json({ message: 'Command sent' }); } catch (error) { console.error("Error sending command:", error); res.status(500).json({ error: 'Failed to send command' }); } }); // Route to get information about a specific client (example: player list) app.get('/clients/:clientId/players', (req, res) => { const clientId = parseInt(req.params.clientId); const bot = minecraftClients[clientId]; if (!bot) { return res.status(404).json({ error: 'Client not found' }); } const playerList = Object.keys(bot.players).map(username => ({ username: username, // Add other player properties as needed (e.g., position) })); res.json(playerList); }); // Start the server app.listen(port, () => { console.log(`Server listening at http://localhost:${port}`); }); ``` **Explanation:** * **Dependencies:** Install the necessary packages: `npm install express mineflayer` * **`minecraftClients`:** This object stores the `mineflayer` bot instances, keyed by a unique ID. * **`createMinecraftClient`:** This function creates a new `mineflayer` bot, sets up event listeners (login, kicked, error, chat), and returns the bot instance. Crucially, it's `async` because `mineflayer.createBot` can take some time. Error handling is important here. * **`/clients` (POST):** This route handles the creation of new Minecraft clients. It receives the server details (host, port, username, password) from the request body, calls `createMinecraftClient`, stores the bot in the `minecraftClients` object, and returns a success response with the new client's ID. * **`/clients/:clientId/command` (POST):** This route sends a command to a specific Minecraft client. It retrieves the client ID from the URL parameters, finds the corresponding bot in the `minecraftClients` object, and uses `bot.chat()` to send the command. * **`/clients/:clientId/players` (GET):** This route retrieves a list of players on the server the bot is connected to. * **Error Handling:** The code includes basic error handling. You should expand this to handle different types of errors and implement appropriate logging and recovery mechanisms. * **Asynchronous Operations:** The `createMinecraftClient` function and the route handlers use `async/await` to handle the asynchronous nature of the Minecraft protocol. **Key Considerations and Enhancements:** * **Authentication/Authorization:** Implement proper authentication and authorization to protect your API endpoints. You don't want just anyone creating and controlling Minecraft bots. Consider using JWTs or other authentication mechanisms. * **WebSocket Integration:** For real-time updates (e.g., chat messages, player positions), use WebSockets. The Express server can act as a WebSocket server (using libraries like `ws` or `socket.io`). The `chat` event handler in the `createMinecraftClient` function would then broadcast the message to connected WebSocket clients. * **Rate Limiting:** Implement rate limiting to prevent abuse of your API. * **Error Handling and Logging:** Implement robust error handling and logging. Use a logging library like `winston` or `pino`. * **Client Lifecycle Management:** Implement mechanisms to handle client disconnections, reconnections, and cleanup. The `kicked` and `error` events are important here. Consider using a more sophisticated client management system (e.g., a database) if you need to persist client state. * **Configuration:** Externalize configuration settings (e.g., server addresses, usernames, passwords) into environment variables or a configuration file. * **Scalability:** For high-scale deployments, consider using a message queue (e.g., RabbitMQ, Kafka) to decouple the Express server from the MCP clients. The Express server would publish commands to the message queue, and the MCP clients would subscribe to the queue and process the commands. * **Security:** Be extremely careful about storing Minecraft account credentials. Use environment variables and avoid hardcoding them in your code. Consider using a secure vault to store sensitive information. * **Minecraft Version Compatibility:** Ensure that your `mineflayer` or `minecraft-protocol` version is compatible with the Minecraft server version you are connecting to. The example code includes a `version` option in `mineflayer.createBot`. * **Resource Management:** Be mindful of the resources consumed by each Minecraft client. Each client will consume CPU, memory, and network bandwidth. Monitor your server's resources and adjust the number of clients accordingly. * **Command Handling:** Instead of just sending raw chat commands, consider creating a more structured command API. This will make it easier to manage and control the bots. **Example Usage (from a client):** 1. **Create a new client:** ```bash curl -X POST -H "Content-Type: application/json" -d '{"host": "your_server_ip", "port": 25565, "username": "your_bot_username", "password": "your_bot_password"}' http://localhost:3000/clients ``` This will return a JSON response with the `clientId`. 2. **Send a command:** ```bash curl -X POST -H "Content-Type: application/json" -d '{"command": "/say Hello from the bot!"}' http://localhost:3000/clients/1/command # Replace 1 with the actual clientId ``` 3. **Get the player list:** ```bash curl http://localhost:3000/clients/1/players # Replace 1 with the actual clientId ``` **Spanish Translation of Key Concepts:** * **Minecraft Protocol (MCP) Clients:** Clientes del Protocolo de Minecraft (MCP) * **NodeJS Express Server:** Servidor NodeJS Express * **Adapter/Middleware:** Adaptador/Middleware * **Client Management:** Gestión de Clientes * **Routing:** Enrutamiento * **Asynchronous Operations:** Operaciones Asíncronas * **Unique ID:** ID Único * **Server Address:** Dirección del Servidor * **Authentication/Authorization:** Autenticación/Autorización * **WebSocket Integration:** Integración de WebSocket * **Rate Limiting:** Limitación de Tasa * **Error Handling and Logging:** Manejo de Errores y Registro * **Client Lifecycle Management:** Gestión del Ciclo de Vida del Cliente * **Configuration:** Configuración * **Scalability:** Escalabilidad * **Security:** Seguridad * **Resource Management:** Gestión de Recursos * **Command Handling:** Manejo de Comandos This comprehensive guide should give you a solid foundation for building your multi-MCP client adapter. Remember to adapt the code and concepts to your specific needs and requirements. Good luck!

markapatent-mcp

markapatent-mcp

Enables searching Turkish patent, trademark, and industrial design data via the TÜRKPATENT database using the MCP protocol.

MCPedia

MCPedia

MCP server for a content-first knowledge base, enabling AI agents to search (full-text, semantic, hybrid) and retrieve Markdown documents, list content, and find related docs.

racksmith

racksmith

MCP server for bare-metal lab operations, enabling inventory, commissioning, deployment, fabric audit, and job scheduling via Hermes agent skills and a tool server.

Team Sync MCP Server

Team Sync MCP Server

A self-hosted synchronization hub for frontend and backend teams using Cursor, enabling publishing and reading of shared project state with a local React dashboard.

Nature Vision MCP Server

Nature Vision MCP Server

Enables LLMs to identify biological species from images using the Nature Vision API. It returns Latin scientific names and confidence scores for various categories including plants, animals, and fungi.

Oneiros MCP Server

Oneiros MCP Server

Exposes a JEPA latent world model for planning in a 2D point-mass environment via MCP tools like encode_observation, predict_rollout, and plan_to_goal.

clickup-mcp-full

clickup-mcp-full

Enables reading entire nested ClickUp subtask trees in a single call, plus task, list, folder, comment, tag, and time-tracking operations through natural language.

GenFilesMCP

GenFilesMCP

Generates PowerPoint, Excel, Word, and Markdown files from natural language requests and reviews Word documents with AI comments.

ClickUp MCP Server

ClickUp MCP Server

An MCP server that enables AI assistants to manage ClickUp workspaces, tasks, comments, and time entries through natural language, deployable on Google Cloud Run.

Prolog MCP Server

Prolog MCP Server

Enables LLMs to define Prolog facts and rules and execute queries via SWI-Prolog for logical reasoning.

Procuradoria Geral do Estado MT: Dívida Ativa

Procuradoria Geral do Estado MT: Dívida Ativa

Enables querying of active debt information from the official source of the Attorney General of the State of Mato Grosso. Provides a read-only MCP tool to consult debt status via a hosted HTTP API with prepaid credits.

Wealth Management MCP

Wealth Management MCP

Provides specialized tools for portfolio health analysis, rebalance simulations, and trade execution via a single interface powered by MongoDB. It enables AI agents to manage investment portfolios while adhering to organizational governing rules, clearance guards, and rate limits.

Metabase MCP Server

Metabase MCP Server

Connects Metabase business intelligence platform to AI assistants, enabling users to query data, create dashboards and charts, manage databases, and interact with BI assets using natural language through the Model Context Protocol.

AI Process Manager

AI Process Manager

Enables AI agents to query Windows process, window, and console information via structured JSON instead of screenshots, reducing token usage by 94-98%.

iTechSmart MCP Server

iTechSmart MCP Server

Enables AI agents to verify cryptographic ProofLink receipts, query UAIO platform status, and simulate infrastructure attacks, providing autonomous IT operations with audit trails.

@cyanheads/openchargemap-mcp-server

@cyanheads/openchargemap-mcp-server

Find EV charging stations by location and connector, get full station detail, resolve reference IDs, and read community reliability check-ins via MCP.

SmartCut

SmartCut

SmartCut is a hosted cutting-optimisation API. It turns a list of parts and available stock into machine-ready cutting patterns for sheet materials (plywood, MDF, glass, plastic, sheet metal), linear stock (timber, bar, pipe, extrusion) and roll goods. Guillotine and true-shape nesting modes, with grain direction, per-part orientation locks, edge banding, blade kerf and stock trim.