Discover Awesome MCP Servers

pocketbase-mcp-server MCP Server

Claude AI が自然言語で PocketBase データベースとやり取りできるようにする、コレクションのリスト表示やアクセスを可能にするモデルコンテキストプロトコルサーバー。

simple-mcp-server

シンプルな MCP サーバー

SQLite MCP Server

SQL操作（SELECT、INSERT、UPDATE、DELETE）とテーブル管理を、標準化されたインターフェースを通じてSQLiteデータベースで可能にする、モデルコンテキストプロトコルサーバー。

MCP Manager

MCPサーバーを管理するためのシンプルなGUI。MCPサーバーの簡単な切り替えが可能です。

ClickUp MCP Server

ClickUp MCPサーバーを使用すると、AIアシスタントがClickUpワークスペースとシームレスに連携できます。この強力な統合により、Claudeは自然な会話を通じて、タスクの作成と管理、ドキュメントへのアクセス、フォルダーとリストの整理、コメントの追加、チェックリストの処理などを行うことができます。

octomind mcp server for tools, resources and prompts

mcpサーバーはOctomindプラットフォームと連携します。Octomindは、E2Eウェブテストのための作成、実行、修正ソリューションを提供します。詳細については、https://octomind.dev をご覧ください。

Rust Docs MCP Server

鏡 (Kagami)

Goal Story MCP Server

AIを活用した目標管理システム。従来の目標追跡をストーリーテリングに変え、パーソナライズされた物語と洞察によって、ユーザーが一度に一つの目標に集中できるよう支援し、モチベーションと達成度を高めます。

WebDAV MCP Server

Claude Desktopやその他のMCPクライアントが、自然言語コマンドを通じてWebDAVファイルシステムとやり取りし、CRUD操作を可能にするModel Context Protocolサーバー。

Plugin de Estatísticas de Uso de IA

AI Code Assist の利用状況統計を収集するための MPC サーバー

MCP Custom Server

Model Context Protocol TypeScript SDK を使用したカスタム MCP サーバー実装

Simple Weather MCP Server example from Quickstart

鏡 (Kagami)

MCP Server for sensor device

Node.js アプリケーション。CO2 センサーデータとやり取りするための JSON-RPC インターフェースを提供し、シミュレーションモードと実際の Raspberry Pi Pico ハードウェア接続の両方で動作します。

PagerDuty MCP Server

PagerDuty APIの機能をLLMに公開するサーバー。構造化された入出力により、インシデント、サービス、チーム、ユーザーの管理を可能にする。

Anthropic MCP Code Analyzer

オープンソースプロジェクトを分析し、コードベースの統合を支援するMCPサーバー

MCP Figma to React Converter

Figmaのデザインを、Figmaファイルからコンポーネントを抽出し、すぐに使えるコードに変換することで、TypeScriptとTailwind CSSを用いたReactコンポーネントに変換します。

Ldoce MCP Server

Longman現代英英辞典のウェブサイトから構造化された辞書データを抽出し、Model Context Protocolを通じて提供することで、AIエージェントが詳細な単語の定義、例文、言語情報にアクセスできるようにします。

Medical_calculator_MCP

医療計算用MCPサーバー (Iryō keisan-yō MCP sābā)

Uniswap Trader MCP

複数のブロックチェーンにわたるUniswap DEXでのトークンスワップを自動化するためのAIエージェント用MCPサーバー。

PBIXRay MCP Server

AIクライアントが、PBIXRay Pythonパッケージを通じてメタデータをクエリすることで、PowerBIモデルとインタラクトできるようにするモデルコンテキストプロトコル。

MCP server proxy

Okay, I understand. You want to extend the functionality of an MCP (Minecraft Protocol) server to act as a worker in a distributed system. This means you want the MCP server to be able to receive tasks, process them (likely related to Minecraft), and return results. Here's a breakdown of the concepts, potential approaches, and considerations for achieving this: **1. Understanding the Core Concepts** * **MCP Server:** This is your existing Minecraft server, likely built using a library like `minecraft-server` or a similar framework that handles the Minecraft protocol. It currently manages players, world state, and game logic. * **Worker:** In this context, a worker is a process or service that receives tasks from a central system (a task queue, a scheduler, etc.), performs those tasks, and reports the results. * **Task Queue/Scheduler:** This is the central component that distributes tasks to available workers. Examples include: * **RabbitMQ:** A robust message broker. * **Celery (Python):** A distributed task queue. * **Redis:** Can be used as a simple task queue. * **Custom Solution:** You could build your own task queue using a database or other mechanism. * **Task Definition:** A task is a unit of work. For example: * "Generate a specific chunk of the world." * "Find the nearest diamond ore to coordinates X, Y, Z." * "Simulate the growth of a crop over a period of time." * "Execute a specific command on the server." * **Serialization/Deserialization:** You'll need a way to convert tasks and results into a format that can be sent over the network (e.g., JSON, Protocol Buffers). **2. Potential Approaches** Here are a few ways to integrate the MCP server as a worker, ranked from simpler to more complex: * **A. Command-Based Integration (Simplest):** * **Concept:** The task queue sends commands to the MCP server via the Minecraft protocol itself (e.g., using the `/execute` command or a custom plugin). * **Implementation:** 1. **Plugin/Mod:** Create a plugin or mod for your MCP server that listens for specific commands. 2. **Task Queue:** The task queue sends these commands to a designated player (or a "bot" account) on the server. 3. **Command Processing:** The plugin executes the command and captures the output. 4. **Result Reporting:** The plugin sends the result back to the task queue (e.g., via a separate HTTP endpoint or by sending another command). * **Pros:** Relatively easy to implement, leverages existing Minecraft protocol. * **Cons:** Limited by the capabilities of Minecraft commands, potential security risks if not carefully implemented, can be slow due to protocol overhead. * **B. Direct API Integration (More Flexible):** * **Concept:** Expose an API (e.g., a REST API) on the MCP server that allows external systems to trigger specific actions. * **Implementation:** 1. **API Server:** Embed a web server (e.g., using Flask, FastAPI, or Node.js) within your MCP server process. 2. **API Endpoints:** Create API endpoints that correspond to the tasks you want to perform (e.g., `/generate_chunk`, `/find_diamonds`). 3. **Task Queue:** The task queue sends HTTP requests to these endpoints with the task parameters. 4. **Task Processing:** The API endpoint executes the task within the MCP server's environment. 5. **Result Reporting:** The API endpoint returns the result as a JSON response. * **Pros:** More flexible than command-based integration, allows for more complex tasks. * **Cons:** Requires more development effort, needs careful security considerations. * **C. Dedicated Worker Process (Most Robust):** * **Concept:** Create a separate worker process that interacts with the MCP server's internal data structures directly (if possible) or through a well-defined API. * **Implementation:** 1. **Worker Process:** Develop a separate application (e.g., in Python, Java, or Go) that acts as the worker. 2. **Task Queue:** The task queue sends tasks to the worker process. 3. **MCP Server Interaction:** The worker process connects to the MCP server (either directly to its internal data structures, if accessible, or through a dedicated API). 4. **Task Processing:** The worker process performs the task using the MCP server's resources. 5. **Result Reporting:** The worker process sends the result back to the task queue. * **Pros:** Most robust and scalable, allows for the most complex tasks, can be optimized for performance. * **Cons:** Most complex to implement, requires a deep understanding of the MCP server's internals. **3. Key Considerations** * **Concurrency:** Minecraft servers are often single-threaded or have limited multi-threading capabilities. You'll need to carefully manage concurrency to avoid performance bottlenecks or crashes. Consider using asynchronous programming techniques (e.g., `asyncio` in Python) or offloading tasks to separate threads or processes. * **Resource Management:** Tasks can consume significant resources (CPU, memory, disk I/O). Implement resource limits and monitoring to prevent tasks from overloading the server. * **Security:** If you're exposing an API, ensure that it's properly secured to prevent unauthorized access. Use authentication, authorization, and input validation. * **Error Handling:** Implement robust error handling to gracefully handle task failures. Retry failed tasks, log errors, and provide informative error messages. * **Task Prioritization:** If you have different types of tasks, consider implementing task prioritization to ensure that important tasks are processed first. * **Data Consistency:** If tasks modify the world state, ensure that the changes are properly synchronized to avoid data inconsistencies. * **MCP Server Architecture:** The specific architecture of your MCP server implementation will heavily influence the best approach. If it's based on a well-defined API or plugin system, integration will be easier. If it's a more monolithic codebase, you may need to modify the core server code. * **Minecraft Protocol Limitations:** Be aware of the limitations of the Minecraft protocol. Some tasks may not be possible to perform directly through the protocol. **4. Example Scenario (API Integration - Approach B)** Let's say you want to implement a task to generate a specific chunk of the world. 1. **MCP Server (with API):** ```python from flask import Flask, request, jsonify import minecraft_server # Your MCP server library app = Flask(__name__) mcp_server = minecraft_server.MinecraftServer() # Initialize your server @app.route('/generate_chunk', methods=['POST']) def generate_chunk(): data = request.get_json() x = data.get('x') z = data.get('z') if x is None or z is None: return jsonify({'error': 'Missing x or z coordinate'}), 400 try: chunk_data = mcp_server.generate_chunk(x, z) # Call your chunk generation function return jsonify({'chunk_data': chunk_data}) except Exception as e: return jsonify({'error': str(e)}), 500 if __name__ == '__main__': mcp_server.start() # Start the minecraft server app.run(debug=True, port=5000) # Start the API server ``` 2. **Task Queue (using Celery):** ```python from celery import Celery import requests celery_app = Celery('tasks', broker='redis://localhost:6379/0') # Configure Celery @celery_app.task def generate_chunk_task(x, z): url = 'http://your_mcp_server_ip:5000/generate_chunk' data = {'x': x, 'z': z} response = requests.post(url, json=data) if response.status_code == 200: return response.json() else: raise Exception(f"Chunk generation failed: {response.text}") # Example usage: result = generate_chunk_task.delay(10, 20) # Enqueue the task print(f"Task ID: {result.id}") # Later, retrieve the result: # chunk_data = result.get() ``` **5. Steps to Implement** 1. **Choose an Approach:** Select the approach that best suits your needs and the architecture of your MCP server. 2. **Set up a Task Queue:** Choose a task queue system (RabbitMQ, Celery, Redis, etc.) and configure it. 3. **Implement the Worker Logic:** Write the code that will run on the MCP server to process tasks. This will involve creating API endpoints, plugins, or modifying the core server code. 4. **Implement the Task Queue Integration:** Write the code that will enqueue tasks and retrieve results from the task queue. 5. **Test Thoroughly:** Test your integration thoroughly to ensure that it's working correctly and that it's handling errors gracefully. 6. **Monitor Performance:** Monitor the performance of your system to identify bottlenecks and optimize performance. **Important Notes:** * Replace placeholders like `minecraft_server`, `your_mcp_server_ip`, and `redis://localhost:6379/0` with your actual values. * This is a high-level overview. The specific implementation details will depend on your MCP server and the task queue system you choose. * Security is paramount. Always validate input and protect your API endpoints. By carefully considering these factors and following a structured approach, you can successfully extend your MCP server to function as a worker in a distributed system. Good luck!

MCP Analytics Middleware

MCP SDKサーバー向けの軽量なTypeScriptミドルウェアで、分析機能を提供します。リクエストメトリクス、パフォーマンスデータ、および利用パターンを最小限のオーバーヘッドでキャプチャします。リアルタイム監視、設定可能なデータ収集、詳細なレポート機能を備えており、すべて完全な型安全性で実現されています。

mcp-dice: A MCP Server for Rolling Dicemcp-dice: A MCP Server for Rolling Dice

LLM (大規模言語モデル) がサイコロを振れるようにする MCP サーバー

state-server MCP Server

新しいリポジトリ (Atarashii ripojitori)

iFlytek Workflow MCP Server

MCP (Model Context Protocol) を通じて iFlytek のワークフローを呼び出すことを可能にする MCP サーバー実装。シーケンシャル、パラレル、ループ、ネストされた実行モードによるインテリジェントなワークフローのスケジューリングを可能にします。

Thingsboard MCP Server

Thingsboard MCP Server を LLM ツールで Thingsboard データをコンテキストとして使用するためのサーバー