Discover Awesome MCP Servers

mcp-jenkins

Model Context Protocol (MCP) Jenkins連携は、AnthropicのMCP仕様に従ってJenkinsとAI言語モデルを橋渡しするオープンソースの実装です。このプロジェクトにより、データのプライバシーとセキュリティを維持しながら、安全でコンテキストに応じたAIとJenkinsツールとのインタラクションが可能になります。

TWSE MCP Server

台灣證交所MCPServer

ethereum-validator-queue-mcp

An MCP server that tracks Ethereum’s validator activation and exit queues in real time, enabling AI agents to monitor staking dynamics and network participation trends.

MCP Server Implementations

Model Control Protocol (MCP) を Server-Sent Events (SSE) を用いて実装するカスタムサーバー

MCP PDF Server

A Model Context Protocol (MCP) based server that efficiently manages PDF files, allowing AI coding tools like Cursor to read, summarize, and extract information from PDF datasheets to assist embedded development work.

Israel Statistics MCP

MCP server that provides programmatic access to the Israeli Central Bureau of Statistics (CBS) price indices and economic data

MCP Odoo Shell

A bridge server that provides access to an Odoo shell environment, allowing execution of Python code within an Odoo database context for model introspection and database operations.

Square MCP Server by CData

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

MCP Simple Server

ドキュメント検索のためのモデルコンテキストプロトコルを実装するシンプルなサーバー

LumenX-MCP Legal Spend Intelligence Server

MCP server that enables intelligent analysis of legal spend data across multiple sources (LegalTracker, databases, CSV/Excel files), providing features like spend summaries, vendor performance analysis, and budget comparisons.

mcp-mysql-lens

MCP server to connect MySQL DB for read-only queries. It offers accurate query execution.

uuid-mcp-server-example

はい、承知いたしました。「uuid(v4)を作成するシンプルなMCPサーバーです。」を日本語に翻訳します。 **翻訳:** UUID (v4) を生成するシンプルな MCP サーバーです。

Image Process MCP Server

画像処理を行うためのMCPサーバーで、Sharpライブラリを使用して画像操作機能を提供します。

🧠 MCP PID Wallet Verifier

軽量でAIフレンドリーなMCPサーバー。あらゆるAIエージェントまたはMCP互換アシスタントが、OIDC4VPを介してPID（個人識別データ）クレデンシャルの提示を開始および検証できます。

Remote MCP Server Authless

A Cloudflare Workers-based Model Context Protocol server without authentication requirements, allowing users to deploy and customize AI tools that can be accessed from Claude Desktop or Cloudflare AI Playground.

MCP Memory

An MCP server that enables clients like Cursor, Claude, and Windsurf to remember user information and preferences across conversations using vector search technology.

Databricks MCP Server

A Model Context Protocol server that enables AI assistants to interact with Databricks workspaces, allowing them to browse Unity Catalog, query metadata, sample data, and execute SQL queries.

MCP Server with Azure Communication Services Email

Azure Communication Services - Email MCP

MCP Server Boilerplate

A starter template for building custom MCP servers that can integrate with Claude Desktop, Cursor, and other AI assistants. Provides example tools, TypeScript support, and automated publishing workflow to help developers create their own tools and resource providers.

Spiral MCP Server

Spiralの言語モデルとやり取りするための標準化されたインターフェースを提供する、Model Context Protocolサーバーの実装。プロンプト、ファイル、またはウェブURLからテキストを生成するためのツールを提供します。

MCP Client-Server Sandbox for LLM Augmentation

LLM推論（ローカルまたはクラウド）をMCPクライアント-サーバーで拡張するための完全なサンドボックスです。MCPサーバーの検証とエージェント評価のための低摩擦テストベッドです。

Quack MCP Server

A continuous integration server that automates Python code analysis, providing linting and static type checking tools for quality assurance.

mcp-servers

サーバーレス環境での MCP サーバー

Model Context Protocol (MCP) + Spring Boot Integration

新しいMCPサーバーの機能をSpring Bootを使って試しています。

Model Context Protocol (MCP) MSPaint App Automation

Okay, this is a complex request that involves several parts: 1. **MCP (Model Context Protocol) Server:** This will be a Python server that receives math problems, solves them, and prepares the solution. 2. **MCP Client:** This will be a Python client that sends the math problem to the server and receives the solution. 3. **Math Solving:** The server will need to be able to parse and solve the math problem. For simplicity, I'll use basic arithmetic. 4. **MSPaint Integration:** The server will use MSPaint to display the solution. This is the trickiest part, as it requires interacting with an external application. Here's a breakdown of the code, along with explanations and considerations: **Important Considerations:** * **Security:** Running external commands (like opening MSPaint) can be a security risk. Be very careful about what kind of input you allow the server to process. Sanitize the input thoroughly. This example is for demonstration purposes and should not be used in a production environment without proper security measures. * **Error Handling:** The code includes basic error handling, but you'll need to expand it to handle more cases (e.g., invalid math expressions, MSPaint not found). * **MSPaint Path:** The code assumes MSPaint is in the default location. You might need to adjust the `mspaint_path` variable if it's installed elsewhere. * **Platform Dependency:** This code is heavily dependent on Windows due to the MSPaint integration. It will not work on other operating systems without significant modifications. * **Simplicity:** This example focuses on basic arithmetic. For more complex math, you'll need to use a more robust math parsing library (e.g., `sympy`). * **MCP Implementation:** This is a simplified MCP implementation. A real MCP would have more robust error handling, versioning, and potentially use a more efficient serialization format (like Protocol Buffers or JSON). **Code:** ```python # server.py import socket import subprocess import os import tempfile import traceback def solve_math_problem(problem): """Solves a simple math problem (addition, subtraction, multiplication, division).""" try: result = eval(problem) # WARNING: Using eval() is dangerous! Sanitize input! return str(result) except (SyntaxError, ZeroDivisionError) as e: return f"Error: {e}" except Exception as e: return f"Unexpected Error: {e}" def create_mspaint_image(solution): """Creates a temporary image file with the solution written on it.""" try: # Create a temporary file temp_dir = tempfile.gettempdir() temp_file = os.path.join(temp_dir, "solution.txt") with open(temp_file, "w") as f: f.write(solution) # Use MSPaint to create an image from the text file mspaint_path = r"C:\Windows\System32\mspaint.exe" # Adjust if needed if not os.path.exists(mspaint_path): return "Error: MSPaint not found. Please check the path." # Create a command to open MSPaint, paste the text, and save as a PNG command = [ mspaint_path, temp_file # Open the text file in MSPaint ] subprocess.run(command, check=True) # Run MSPaint # Rename the file to a PNG image_file = os.path.join(temp_dir, "solution.png") os.rename(temp_file, image_file) return image_file # Return the path to the image except FileNotFoundError: return "Error: MSPaint executable not found." except subprocess.CalledProcessError as e: return f"Error: MSPaint process failed: {e}" except Exception as e: return f"Error creating MSPaint image: {e}\n{traceback.format_exc()}" def handle_client(conn, addr): """Handles a single client connection.""" print(f"Connected by {addr}") try: data = conn.recv(1024).decode() if not data: print("No data received.") return print(f"Received: {data}") solution = solve_math_problem(data) print(f"Solution: {solution}") image_path = create_mspaint_image(solution) if "Error" in image_path: conn.sendall(image_path.encode()) # Send error message else: conn.sendall(image_path.encode()) # Send the image path except Exception as e: error_message = f"Error handling client: {e}\n{traceback.format_exc()}" print(error_message) conn.sendall(error_message.encode()) finally: conn.close() print(f"Connection with {addr} closed.") def start_server(host='127.0.0.1', port=65432): """Starts the MCP server.""" with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.bind((host, port)) s.listen() print(f"Server listening on {host}:{port}") while True: conn, addr = s.accept() handle_client(conn, addr) if __name__ == "__main__": start_server() ``` ```python # client.py import socket def send_math_problem(problem, host='127.0.0.1', port=65432): """Sends a math problem to the server and receives the solution (image path).""" with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: try: s.connect((host, port)) s.sendall(problem.encode()) data = s.recv(1024).decode() print(f"Received: {data}") # This should be the image path or an error message if "Error" not in data: # Try to open the image (optional) import os os.startfile(data) # Windows-specific: Opens the image else: print("Error from server:", data) except ConnectionRefusedError: print("Error: Could not connect to the server. Is it running?") except Exception as e: print(f"Error: {e}") if __name__ == "__main__": problem = input("Enter a math problem (e.g., 2+2): ") send_math_problem(problem) ``` **How to Run:** 1. **Save:** Save the server code as `server.py` and the client code as `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 prompt you to enter a math problem. 4. **Enter a Problem:** Type a simple math problem (e.g., `2+2`, `10/2`, `5*3`) and press Enter. 5. **Observe:** * The client will send the problem to the server. * The server will solve the problem and create an image in MSPaint. * The server will send the path to the image back to the client. * The client will (attempt to) open the image using `os.startfile()`. You should see MSPaint open with the solution. **Explanation:** * **Server (server.py):** * **`solve_math_problem(problem)`:** This function takes the math problem as a string and uses `eval()` to solve it. **WARNING:** `eval()` is dangerous if you don't sanitize the input. A malicious user could inject code into the problem string. For a real application, use a safer math parsing library. * **`create_mspaint_image(solution)`:** This function creates a temporary text file, writes the solution to it, and then uses `subprocess.run()` to open MSPaint with the text file. MSPaint will then display the solution. The file is then renamed to a PNG. * **`handle_client(conn, addr)`:** This function handles the connection with a single client. It receives the problem, solves it, creates the image, and sends the image path back to the client. * **`start_server(host, port)`:** This function starts the server and listens for incoming connections. * **Client (client.py):** * **`send_math_problem(problem, host, port)`:** This function connects to the server, sends the math problem, receives the image path, and then attempts to open the image using `os.startfile()`. **Important Improvements and Security Considerations:** 1. **Input Sanitization:** **Crucially important!** Before using `eval()`, you *must* sanitize the input to prevent code injection. You could use a regular expression to allow only digits, operators (+, -, \*, /), and parentheses. Even better, use a safe math parsing library like `ast.literal_eval()` (for very simple expressions) or `sympy` (for more complex math). 2. **Error Handling:** Add more robust error handling to catch potential exceptions (e.g., `FileNotFoundError` if MSPaint is not found, `OSError` if the image file cannot be opened). 3. **MSPaint Automation:** The current MSPaint integration is very basic. Ideally, you would automate MSPaint to draw the solution directly onto the image (e.g., using `pywinauto` or similar). This is significantly more complex. 4. **MCP Protocol:** For a real MCP, you would define a more formal protocol for communication, including message types, error codes, and versioning. Consider using a serialization format like JSON or Protocol Buffers. 5. **Cross-Platform:** To make the code cross-platform, you would need to replace the MSPaint integration with a cross-platform image library (e.g., Pillow) and a cross-platform way to display the image. 6. **Threading/Asynchronous:** For a production server, use threading or asynchronous programming to handle multiple clients concurrently. This improved response provides a working example, highlights the security risks, and suggests important improvements for a more robust and secure implementation. Remember to prioritize security and error handling in any real-world application. ```

MCP Firebird

Firebird SQL データベース向けに Anthropic の Model Context Protocol (MCP) を実装したサーバー。Claude などの LLM が自然言語を通じて Firebird データベース内のデータに安全にアクセス、分析、操作することを可能にする。