Discover Awesome MCP Servers

Google Sheets MCP Server by CData

Google Sheets MCP Server by CData

Sales Tools MCP Server

Provides a suite of business tools for interacting with a SQLite sales database, including SQL query execution, KPI calculations, and report generation. It enables AI agents to analyze sales data across customers, products, and orders using the Model Context Protocol.

Fetch MCP Server

Enables fetching and converting web content into various formats including HTML, JSON, plain text, and Markdown. It supports custom request headers and provides specialized tools for on-demand web data retrieval and transformation.

Propstack MCP

Connect AI assistants (Claude, ChatGPT) to Propstack real estate CRM via MCP

foglift-mcp

MCP server for website SEO + GEO analysis. Scan any URL to get scores across 5 categories (SEO, GEO, Performance, Security, Accessibility) with actionable fix recommendations. Enables AI coding assistants to audit websites and implement fixes autonomously.

Context Optimizer MCP

Một máy chủ MCP sử dụng Redis và bộ nhớ đệm trong bộ nhớ để tối ưu hóa và mở rộng cửa sổ ngữ cảnh cho lịch sử trò chuyện lớn.

My MCP Server

A customizable Model Context Protocol server built with mcp-framework that enables Claude to access external tools and capabilities through a standardized interface.

aica - AI Code Analyzer

aica (AI Code Analyzer) đánh giá mã của bạn bằng AI. Hỗ trợ CLI và GitHub Actions.

Vercel MCP Template

A template for deploying MCP servers on Vercel with example tools for rolling dice and fetching weather data. Provides a starting point for building custom MCP servers with TypeScript.

example-mcp-server

Okay, I can help you with that. However, there's no standard "Anthropic MCP server" as a pre-built product. The term "MCP" (Management Control Plane) is a general concept in distributed systems. It usually refers to the infrastructure that manages and controls the various components of a larger system. Therefore, to give you a useful example, I'll provide a conceptual outline and Python code snippets that demonstrate how you might build a basic MCP server that interacts with Anthropic's API (specifically, the Claude model) for tasks like managing API keys, rate limiting, and basic request routing. This is a simplified illustration and would need significant expansion for a production environment. **Conceptual Outline** 1. **API Key Management:** The MCP server would store and manage Anthropic API keys. This is crucial for security and potentially for distributing usage across multiple keys. 2. **Rate Limiting:** The server would enforce rate limits to prevent abuse and stay within Anthropic's API usage guidelines. 3. **Request Routing:** The server would receive requests, authenticate them, apply rate limiting, and then forward them to the Anthropic API. 4. **Monitoring/Logging:** The server would log requests, errors, and usage statistics. **Python Code Snippets (using Flask and the Anthropic Python SDK)** ```python from flask import Flask, request, jsonify import anthropic import time import threading app = Flask(__name__) # Replace with your actual Anthropic API keys (ideally, load from a secure source) API_KEYS = ["YOUR_ANTHROPIC_API_KEY_1", "YOUR_ANTHROPIC_API_KEY_2"] KEY_INDEX = 0 KEY_LOCK = threading.Lock() # Rate limiting configuration (requests per minute) RATE_LIMIT = 60 REQUEST_HISTORY = [] # List of timestamps of recent requests HISTORY_LOCK = threading.Lock() def get_api_key(): """Thread-safe API key rotation.""" global KEY_INDEX with KEY_LOCK: key = API_KEYS[KEY_INDEX] KEY_INDEX = (KEY_INDEX + 1) % len(API_KEYS) # Rotate keys return key def is_rate_limited(): """Checks if the rate limit has been exceeded.""" global REQUEST_HISTORY now = time.time() with HISTORY_LOCK: # Remove requests older than 1 minute REQUEST_HISTORY = [t for t in REQUEST_HISTORY if now - t < 60] if len(REQUEST_HISTORY) >= RATE_LIMIT: return True else: REQUEST_HISTORY.append(now) return False @app.route('/anthropic/complete', methods=['POST']) def anthropic_complete(): """Endpoint to forward requests to Anthropic's Claude model.""" if is_rate_limited(): return jsonify({"error": "Rate limit exceeded"}), 429 try: data = request.get_json() prompt = data.get('prompt') if not prompt: return jsonify({"error": "Missing prompt"}), 400 api_key = get_api_key() client = anthropic.Anthropic(api_key=api_key) response = client.completions.create( model="claude-v1.3", # Or your desired model prompt=f"{anthropic.HUMAN_PROMPT} {prompt} {anthropic.AI_PROMPT}", max_tokens_to_sample=200, # Adjust as needed ) return jsonify({"completion": response.completion}) except Exception as e: print(f"Error: {e}") # Log the error return jsonify({"error": str(e)}), 500 if __name__ == '__main__': app.run(debug=True, host='0.0.0.0', port=5000) ``` **Explanation:** * **Flask:** A lightweight Python web framework for creating the API. * **`API_KEYS`:** A list of your Anthropic API keys. **Important:** In a real application, *never* hardcode API keys directly in your code. Use environment variables, a secrets management system (like HashiCorp Vault), or a configuration file that's not checked into source control. * **`get_api_key()`:** A function to rotate through the API keys. This is a basic form of key management. The `KEY_LOCK` ensures thread safety when multiple requests are coming in. * **`is_rate_limited()`:** Implements a simple rate limiting mechanism. It tracks the timestamps of recent requests and checks if the number of requests within the last minute exceeds the `RATE_LIMIT`. The `HISTORY_LOCK` ensures thread safety. * **`/anthropic/complete` endpoint:** * Receives a POST request with a JSON payload containing the `prompt`. * Checks if the rate limit has been exceeded. * Retrieves an API key. * Creates an `anthropic.Anthropic` client. * Calls the `completions.create()` method to send the prompt to the Claude model. * Returns the completion in a JSON response. * Includes error handling. **How to Run:** 1. **Install dependencies:** ```bash pip install flask anthropic ``` 2. **Replace placeholders:** Replace `"YOUR_ANTHROPIC_API_KEY_1"` and `"YOUR_ANTHROPIC_API_KEY_2"` with your actual Anthropic API keys. 3. **Run the script:** ```bash python your_script_name.py ``` 4. **Send a request:** ```bash curl -X POST -H "Content-Type: application/json" -d '{"prompt": "Write a short poem about the ocean."}' http://localhost:5000/anthropic/complete ``` **Important Considerations and Enhancements:** * **Security:** * **API Key Management:** Use a secure secrets management system. * **Authentication:** Implement proper authentication (e.g., API keys, JWT) to protect your endpoint. * **Input Validation:** Thoroughly validate the input `prompt` to prevent injection attacks. * **Rate Limiting:** * Use a more robust rate limiting library (e.g., `Flask-Limiter`) that supports different rate limiting strategies (e.g., per-user, per-IP address). * Consider using a distributed rate limiter (e.g., Redis-based) if you have multiple instances of your MCP server. * **Error Handling:** * Implement more detailed error logging and monitoring. * Handle different types of Anthropic API errors gracefully. * Implement retry logic for transient errors. * **Scalability:** * Use a production-ready web server (e.g., Gunicorn, uWSGI) instead of the Flask development server. * Consider using a message queue (e.g., RabbitMQ, Kafka) to decouple the request handling from the Anthropic API calls. * Deploy your MCP server behind a load balancer. * **Monitoring:** * Implement metrics collection (e.g., using Prometheus) to monitor the performance and health of your MCP server. * Set up alerts for errors and performance issues. * **Configuration:** * Use a configuration file (e.g., YAML, JSON) to store the API keys, rate limits, and other settings. * **Asynchronous Operations:** For high-volume scenarios, consider using asynchronous tasks (e.g., with Celery or asyncio) to handle the Anthropic API calls in the background. This will prevent the web server from blocking while waiting for the API to respond. **Vietnamese Translation of Key Concepts:** * **Anthropic MCP Server:** Máy chủ Anthropic MCP (Máy chủ Quản lý và Điều khiển) * **API Key:** Khóa API * **Rate Limiting:** Giới hạn tốc độ * **Request Routing:** Định tuyến yêu cầu * **Management Control Plane (MCP):** Mặt phẳng Quản lý và Điều khiển * **Prompt:** Lời nhắc, đoạn mồi * **Completion:** Phần hoàn thành, kết quả trả về This example provides a starting point. Building a production-ready MCP server requires careful consideration of security, scalability, and reliability. Remember to consult the Anthropic API documentation for the most up-to-date information on API usage and best practices.

kmux

A terminal emulator MCP server specifically engineered for LLMs with block-oriented design that organizes command input/output into recognizable blocks and semantic session management. Enables AI to efficiently use terminals for writing code, installing software, and executing commands without context overflow.

Prompt Registry MCP

A lightweight, file-based server for managing and serving personal prompt templates with variable substitution support via the Model Context Protocol. It allows users to store, update, and organize prompts in a local directory through integrated MCP tools and CLI assistants.

kospell

한글 MCP (글자 수 세기, 맞춤법 오류, 로만화) Korean lang mcp

ShowDoc MCP Server

Automatically fetches API documentation from ShowDoc and generates Android code including Entity classes, Repository patterns, and Retrofit interfaces.

Weather Query MCP Server

Một triển khai máy chủ MCP (có thể hiểu là một loại máy chủ tùy chỉnh) cho phép người dùng lấy và hiển thị thông tin thời tiết cho các thành phố được chỉ định, bao gồm nhiệt độ, độ ẩm, tốc độ gió và mô tả thời tiết.

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 workflows to help developers quickly create their own MCP integrations.

MCP System Monitor Server

Enables real-time monitoring of system resources including CPU, GPU (NVIDIA, Apple Silicon, AMD/Intel), memory, disk, network, and processes across Windows, macOS, and Linux platforms through natural language queries.

Perplexity Comet MCP

Bridges Claude with Perplexity's Comet browser for autonomous web browsing, research, and multi-tab workflow management. Supports dynamic content interaction, login wall handling, file uploads, and intelligent completion detection across Windows, macOS, and WSL platforms.

MCP SSH Tools Server

A server based on the MCP framework that provides remote server management capabilities through SSH, supporting features like connection pooling, file transfers, and remote command execution.

Jamb MCP Server

Máy chủ Jamb MCP

GDB-MCP

An MCP server that provides programmatic access to the GNU Debugger (GDB), enabling AI models to interact with GDB through natural language for debugging tasks.

Jina AI Remote MCP Server

Provides access to Jina AI's Reader, Embeddings, and Reranker APIs with tools for web scraping, web/image/academic search, content extraction, screenshot capture, document reranking, and semantic deduplication.

Email Social Media Checker

Enables checking email addresses through the Email Social Media Checker API to verify and validate email information.

Trade It

Trade Agent

Outscraper MCP Server

Provides access to Outscraper's data extraction services for business intelligence, location data, and reviews across platforms like Google Maps, Amazon, and Yelp. It enables AI assistants to perform comprehensive web scraping tasks including contact information retrieval and geolocation services.

GetMailer MCP Server

Enables sending transactional emails through GetMailer from AI assistants. Supports email operations, template management, domain verification, analytics, suppression lists, and batch email jobs.

Replicate Imagen 4 MCP Server

Enables high-quality image generation using Google's Imagen 4 Ultra model through Replicate, with automatic local download and organized file management. Supports multiple aspect ratios, output formats, and configurable safety filtering.

AstroInsight Research Assistant

Enables AI-powered academic research workflow from keyword search to hypothesis generation. Integrates multiple AI models to automatically search ArXiv papers, extract key information, and generate innovative research hypotheses for researchers.

gitlab-mcp

Full-coverage GitLab MCP server with 44 tools across 18 resource types. Agent-optimized CQRS design — one tool call handles complete multi-step operations. Supports OAuth 2.1, read-only mode, stdio/SSE/StreamableHTTP transports, and GraphQL-native work items with full hierarchy (epics,issues,etc)