Discover Awesome MCP Servers
Extend your agent with 16,166 capabilities via MCP servers.
- All16,166
- Developer Tools3,867
- Search1,714
- Research & Data1,557
- AI Integration Systems229
- Cloud Platforms219
- Data & App Analysis181
- Database Interaction177
- Remote Shell Execution165
- Browser Automation147
- Databases145
- Communication137
- AI Content Generation127
- OS Automation120
- Programming Docs Access109
- Content Fetching108
- Note Taking97
- File Systems96
- Version Control93
- Finance91
- Knowledge & Memory90
- Monitoring79
- Security71
- Image & Video Processing69
- Digital Note Management66
- AI Memory Systems62
- Advanced AI Reasoning59
- Git Management Tools58
- Cloud Storage51
- Entertainment & Media43
- Virtualization42
- Location Services35
- Web Automation & Stealth32
- Media Content Processing32
- Calendar Management26
- Ecommerce & Retail18
- Speech Processing18
- Customer Data Platforms16
- Travel & Transportation14
- Education & Learning Tools13
- Home Automation & IoT13
- Web Search Integration12
- Health & Wellness10
- Customer Support10
- Marketing9
- Games & Gamification8
- Google Cloud Integrations7
- Art & Culture4
- Language Translation3
- Legal & Compliance2
MCP Git Commit Generator
Automatically generates conventional commit messages from staged git changes and checks repository status. Analyzes git diffs to create properly formatted commit messages following conventional commit standards.
LibSQL Memory
Server MCP berperforma tinggi yang memanfaatkan libSQL untuk memori persisten dan kemampuan pencarian vektor, memungkinkan pengelolaan entitas yang efisien dan penyimpanan pengetahuan semantik.
Graphiti MCP Server 🧠
Blackbaud FE NXT MCP Server by CData
This project builds a read-only MCP server. For full read, write, update, delete, and action capabilities and a simplified setup, check out our free CData MCP Server for Blackbaud FE NXT (beta): https://www.cdata.com/download/download.aspx?sku=OZZK-V&type=beta
Wikipedia Summarizer MCP Server
Sebuah server MCP (Model Context Protocol) yang mengambil dan meringkas artikel Wikipedia menggunakan Ollama LLM, dapat diakses melalui antarmuka baris perintah dan Streamlit. Sempurna untuk mengekstrak informasi penting dari Wikipedia dengan cepat tanpa membaca seluruh artikel.
MCP DeepSeek 演示项目
Tentu, berikut adalah contoh kasus penggunaan minimal DeepSeek yang terintegrasi dengan MCP (MicroConfig Protocol) dalam bahasa Inggris dan terjemahannya ke dalam bahasa Indonesia, termasuk kode klien dan server: **English:** **Scenario:** A simple key-value store where the client can set and get values from the server using DeepSeek for inference and MCP for configuration management. **Components:** * **Client:** Sends requests to the server to set or get a value. * **Server:** Stores the key-value pairs and uses DeepSeek to potentially enhance the value before storing or returning it. MCP is used to configure the server's behavior (e.g., DeepSeek model endpoint). **Minimal Code (Conceptual):** **Server (Python):** ```python # server.py import socket import json import deepseek_client # Hypothetical DeepSeek client library import mcp_client # Hypothetical MCP client library # MCP Configuration mcp = mcp_client.MCPClient(server_address="mcp_server:8080") config = mcp.get_config("key_value_store") deepseek_endpoint = config.get("deepseek_endpoint", "default_endpoint") # DeepSeek Client deepseek = deepseek_client.DeepSeekClient(endpoint=deepseek_endpoint) def handle_request(data): try: request = json.loads(data.decode()) action = request.get("action") if action == "set": key = request.get("key") value = request.get("value") # Potentially enhance the value using DeepSeek enhanced_value = deepseek.infer(f"Enhance this value: {value}") store[key] = enhanced_value return json.dumps({"status": "success", "message": f"Set {key} to {enhanced_value}"}) elif action == "get": key = request.get("key") value = store.get(key, None) return json.dumps({"status": "success", "value": value}) else: return json.dumps({"status": "error", "message": "Invalid action"}) except Exception as e: return json.dumps({"status": "error", "message": str(e)}) # Simple in-memory store store = {} # Basic Socket Server (Example - Replace with a proper framework) HOST = 'localhost' PORT = 12345 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) if not data: break response = handle_request(data) conn.sendall(response.encode()) ``` **Client (Python):** ```python # client.py import socket import json def send_request(action, key=None, value=None): request = {"action": action} if key: request["key"] = key if value: request["value"] = value message = json.dumps(request).encode() HOST = 'localhost' PORT = 12345 with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.connect((HOST, PORT)) s.sendall(message) data = s.recv(1024) return data.decode() # Example Usage response = send_request("set", key="my_key", value="initial value") print(f"Set Response: {response}") response = send_request("get", key="my_key") print(f"Get Response: {response}") ``` **Explanation:** 1. **MCP Integration:** The server uses a hypothetical `mcp_client` to fetch configuration data from an MCP server. This configuration includes the endpoint for the DeepSeek model. This allows you to change the DeepSeek model without redeploying the server. 2. **DeepSeek Integration:** The server uses a hypothetical `deepseek_client` to interact with a DeepSeek model. In the `set` action, it sends the value to DeepSeek for potential enhancement. 3. **Client-Server Communication:** The client sends JSON-formatted requests to the server using sockets. The server processes the requests and sends back JSON-formatted responses. 4. **Minimal Example:** This is a simplified example. A real-world application would likely use a more robust framework (e.g., Flask, FastAPI, gRPC) for the server and handle errors more gracefully. **Indonesian Translation:** **Skenario:** Sebuah penyimpanan nilai-kunci sederhana di mana klien dapat mengatur dan mendapatkan nilai dari server menggunakan DeepSeek untuk inferensi dan MCP untuk manajemen konfigurasi. **Komponen:** * **Klien:** Mengirim permintaan ke server untuk mengatur atau mendapatkan nilai. * **Server:** Menyimpan pasangan nilai-kunci dan menggunakan DeepSeek untuk berpotensi meningkatkan nilai sebelum menyimpannya atau mengembalikannya. MCP digunakan untuk mengkonfigurasi perilaku server (misalnya, endpoint model DeepSeek). **Kode Minimal (Konseptual):** **Server (Python):** ```python # server.py import socket import json import deepseek_client # Pustaka klien DeepSeek hipotetis import mcp_client # Pustaka klien MCP hipotetis # Konfigurasi MCP mcp = mcp_client.MCPClient(server_address="mcp_server:8080") config = mcp.get_config("key_value_store") deepseek_endpoint = config.get("deepseek_endpoint", "default_endpoint") # Klien DeepSeek deepseek = deepseek_client.DeepSeekClient(endpoint=deepseek_endpoint) def handle_request(data): try: request = json.loads(data.decode()) action = request.get("action") if action == "set": key = request.get("key") value = request.get("value") # Berpotensi meningkatkan nilai menggunakan DeepSeek enhanced_value = deepseek.infer(f"Tingkatkan nilai ini: {value}") store[key] = enhanced_value return json.dumps({"status": "success", "message": f"Mengatur {key} ke {enhanced_value}"}) elif action == "get": key = request.get("key") value = store.get(key, None) return json.dumps({"status": "success", "value": value}) else: return json.dumps({"status": "error", "message": "Aksi tidak valid"}) except Exception as e: return json.dumps({"status": "error", "message": str(e)}) # Penyimpanan dalam memori sederhana store = {} # Server Socket Dasar (Contoh - Ganti dengan kerangka kerja yang tepat) HOST = 'localhost' PORT = 12345 with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.bind((HOST, PORT)) s.listen() print(f"Server mendengarkan di {HOST}:{PORT}") conn, addr = s.accept() with conn: print(f"Terhubung oleh {addr}") while True: data = conn.recv(1024) if not data: break response = handle_request(data) conn.sendall(response.encode()) ``` **Klien (Python):** ```python # client.py import socket import json def send_request(action, key=None, value=None): request = {"action": action} if key: request["key"] = key if value: request["value"] = value message = json.dumps(request).encode() HOST = 'localhost' PORT = 12345 with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.connect((HOST, PORT)) s.sendall(message) data = s.recv(1024) return data.decode() # Contoh Penggunaan response = send_request("set", key="my_key", value="nilai awal") print(f"Respons Set: {response}") response = send_request("get", key="my_key") print(f"Respons Get: {response}") ``` **Penjelasan:** 1. **Integrasi MCP:** Server menggunakan `mcp_client` hipotetis untuk mengambil data konfigurasi dari server MCP. Konfigurasi ini mencakup endpoint untuk model DeepSeek. Ini memungkinkan Anda mengubah model DeepSeek tanpa menyebarkan ulang server. 2. **Integrasi DeepSeek:** Server menggunakan `deepseek_client` hipotetis untuk berinteraksi dengan model DeepSeek. Dalam aksi `set`, ia mengirimkan nilai ke DeepSeek untuk potensi peningkatan. 3. **Komunikasi Klien-Server:** Klien mengirim permintaan berformat JSON ke server menggunakan soket. Server memproses permintaan dan mengirim kembali respons berformat JSON. 4. **Contoh Minimal:** Ini adalah contoh yang disederhanakan. Aplikasi dunia nyata kemungkinan akan menggunakan kerangka kerja yang lebih kuat (misalnya, Flask, FastAPI, gRPC) untuk server dan menangani kesalahan dengan lebih baik. **Important Notes:** * **Hypothetical Libraries:** `deepseek_client` and `mcp_client` are placeholders. You'll need to replace them with actual libraries or implement your own. For DeepSeek, you'd likely use their API. For MCP, you'd need a client library that interacts with your MCP server. * **Error Handling:** The error handling is very basic. In a real application, you'd want to handle exceptions more gracefully and provide more informative error messages. * **Security:** This example does not include any security measures. In a production environment, you'd need to implement authentication, authorization, and data encryption. * **Socket Server:** The socket server is a very basic example. Consider using a more robust framework like Flask, FastAPI, or gRPC for a production application. * **MCP Server:** You'll need a running MCP server to provide the configuration data. The example assumes it's running at `mcp_server:8080`. This provides a minimal, conceptual example. You'll need to adapt it to your specific needs and environment. Remember to install the necessary libraries (or create your own) and configure the MCP server.
MCP Poisoning Attack - PoC
Repositori ini mendemonstrasikan berbagai **Serangan Peracunan MCP** yang memengaruhi alur kerja agen AI dunia nyata.
Marketfiyati MCP Server
Enables searching and comparing product prices across Turkish supermarket chains (BIM, A101, Migros, SOK, etc.) with location-based filtering. Provides access to Turkish market data from marketfiyati.org.tr for price tracking and comparison.
Disk Usage MCP Server (Demo)
Withings MCP Client
Enables retrieval of health data from Withings smart scales including weight measurements and comprehensive body composition metrics like fat mass, muscle mass, and hydration levels. Supports multiple users, unit preferences, and OAuth authentication for secure access to personal health data.
YDB MCP
Model Context Protocol server for YDB databases that enables AI-powered database operations and natural language interactions with YDB instances from any LLM that supports MCP.
Public Certificate Authority API MCP Server
An MCP Server that enables interaction with Google's Public Certificate Authority API for managing publicly trusted TLS certificates via natural language.
Qdrant MCP Server
Enables semantic code search across codebases using Qdrant vector database and OpenAI embeddings, allowing users to find code by meaning rather than just keywords through natural language queries.
Azure DevOps MCP Server with PAT Authentication
Enables interaction with Azure DevOps services through Personal Access Token authentication. Provides 75 tools for managing work items, builds, repositories, pull requests, and other DevOps operations via both CLI and HTTP API interfaces.
Redis MCP
Enables AI assistants to perform comprehensive Redis database operations including managing strings, hashes, lists, sets, sorted sets, TTL management, and data backup/restore. Supports secure connections and provides batch operations for efficient Redis interaction through natural language.
AWS Security MCP
A Model Context Protocol server that connects AI assistants like Claude to AWS security services, allowing them to autonomously query, inspect, and analyze AWS infrastructure for security issues and misconfigurations.
MCP Dynamic Tools
A dynamic MCP server that automatically discovers Python files in a directory and exposes them as tools to any MCP-compatible AI client, allowing users to easily create and deploy custom AI tools.
WeChat MCP Server
Enables AI assistants to send WeChat messages through automation, supporting both immediate and scheduled message delivery to contacts and groups using the Model Context Protocol standard.
EmailAssistant
An MCP-compatible server that integrates with Gmail API to fetch and summarize emails based on custom queries or recent activity.
CosmWasm MCP Server
Sebuah server Model Context Protocol (MCP) yang menyediakan alat untuk berinteraksi dengan kontrak pintar CosmWasm.
Total PC Control
Server MCP Kontrol PC Total - v2 dengan perbaikan dan kompresi
MCP Server - VSCode Tutorial
Baik, berikut adalah tutorial untuk membangun server MCP (Minecraft Coder Pack) dan menggunakan VS Code sebagai klien MCP: **Bagian 1: Mempersiapkan Lingkungan** 1. **Instal Java Development Kit (JDK):** * Unduh dan instal JDK versi 8 atau 17 dari Oracle atau OpenJDK. Pastikan Anda menginstal versi yang sesuai dengan sistem operasi Anda. * Atur variabel lingkungan `JAVA_HOME` ke direktori instalasi JDK Anda. * Tambahkan direktori `bin` JDK ke variabel lingkungan `PATH`. 2. **Instal Python:** * Unduh dan instal Python 3 dari situs web resmi Python. * Pastikan opsi "Add Python to PATH" dicentang selama instalasi. 3. **Instal Git:** * Unduh dan instal Git dari situs web resmi Git. 4. **Instal Visual Studio Code (VS Code):** * Unduh dan instal VS Code dari situs web resmi VS Code. * Instal ekstensi berikut di VS Code: * Java Extension Pack (oleh Microsoft) * Python (oleh Microsoft) **Bagian 2: Mengunduh dan Mengatur MCP** 1. **Unduh MCP:** * Kunjungi situs web MCP (misalnya, [https://www.modcoderpack.com/](https://www.modcoderpack.com/)). * Unduh versi MCP yang sesuai dengan versi Minecraft yang ingin Anda modifikasi. 2. **Ekstrak MCP:** * Ekstrak file ZIP MCP ke direktori pilihan Anda. Misalnya, `C:\mcp`. 3. **Konfigurasi MCP:** * Buka direktori MCP yang diekstrak. * Edit file `conf/mcp.cfg` dengan editor teks. * Pastikan pengaturan berikut sudah benar: * `MinecraftVersion`: Versi Minecraft yang Anda gunakan. * `MCPVersion`: Versi MCP yang Anda gunakan. * `ClientPath`: Lokasi file `minecraft.jar` klien Minecraft Anda. * `ServerPath`: Lokasi file `minecraft_server.jar` server Minecraft Anda. **Bagian 3: Membangun Server MCP** 1. **Buka Command Prompt atau Terminal:** * Buka Command Prompt (Windows) atau Terminal (macOS/Linux). 2. **Navigasi ke Direktori MCP:** * Gunakan perintah `cd` untuk menavigasi ke direktori MCP yang diekstrak. Misalnya: ```bash cd C:\mcp ``` 3. **Decompile Minecraft:** * Jalankan perintah berikut untuk mendekompilasi kode Minecraft: ```bash ./decompile.sh (Linux/macOS) decompile.bat (Windows) ``` * Proses ini akan memakan waktu cukup lama. 4. **Apply Patches (Opsional):** * Jika Anda memiliki patch yang ingin diterapkan, letakkan di direktori `patches` dan jalankan perintah berikut: ```bash ./patch.sh (Linux/macOS) patch.bat (Windows) ``` 5. **Recompile Minecraft:** * Jalankan perintah berikut untuk mengkompilasi ulang kode Minecraft: ```bash ./recompile.sh (Linux/macOS) recompile.bat (Windows) ``` 6. **Repack Minecraft:** * Jalankan perintah berikut untuk mengemas ulang kode Minecraft: ```bash ./reobfuscate.sh (Linux/macOS) reobfuscate.bat (Windows) ``` **Bagian 4: Menggunakan VS Code sebagai Klien MCP** 1. **Buka Direktori `src` di VS Code:** * Di VS Code, buka folder `src` di dalam direktori MCP. Ini adalah tempat Anda akan menulis kode mod Anda. 2. **Buat Proyek Java:** * VS Code akan mendeteksi bahwa ini adalah proyek Java dan akan meminta Anda untuk mengkonfigurasi proyek. Ikuti petunjuk untuk mengkonfigurasi proyek Java. 3. **Tambahkan Library MCP ke Proyek:** * Tambahkan library MCP (biasanya terletak di direktori `lib`) ke proyek Java Anda. Ini memungkinkan Anda untuk menggunakan kelas dan metode Minecraft di kode mod Anda. 4. **Tulis Kode Mod Anda:** * Buat kelas Java baru di direktori `src` dan mulai menulis kode mod Anda. Anda dapat menggunakan kelas dan metode Minecraft yang didekompilasi untuk memodifikasi perilaku game. 5. **Bangun Mod Anda:** * Gunakan fitur build VS Code untuk mengkompilasi kode mod Anda. 6. **Uji Mod Anda:** * Salin file JAR mod Anda ke direktori `mods` di instalasi Minecraft Anda. * Jalankan Minecraft dan uji mod Anda. **Tips Tambahan:** * **Dokumentasi MCP:** Baca dokumentasi MCP untuk informasi lebih lanjut tentang cara menggunakan MCP. * **Forum dan Komunitas:** Bergabunglah dengan forum dan komunitas MCP untuk mendapatkan bantuan dan berbagi pengetahuan. * **Debugging:** Gunakan fitur debugging VS Code untuk men-debug kode mod Anda. * **Version Control:** Gunakan Git untuk mengelola kode mod Anda dan melacak perubahan. **Catatan Penting:** * Proses ini kompleks dan membutuhkan pemahaman dasar tentang Java dan Minecraft. * Pastikan Anda memiliki cadangan file Minecraft Anda sebelum memulai proses ini. * Berhati-hatilah saat memodifikasi kode Minecraft, karena dapat menyebabkan masalah stabilitas. Semoga tutorial ini membantu! Jika Anda memiliki pertanyaan lebih lanjut, jangan ragu untuk bertanya.
MCPKG - Model Context Protocol Knowledge Graph
Sebuah server MCP (Model Context Protocol) yang mengekspos primitif untuk berinteraksi dengan Knowledge Graph.
Python MCP Korea Weather Service
MCP server that provides Korean weather information using grid coordinates and the Korea Meteorological Administration API, allowing users to query current weather conditions and forecasts for specific locations in Korea.
pfSense MCP Server
A production-grade server that enables natural language interaction with pfSense firewalls through Claude Desktop and other GenAI applications, supporting multiple access levels and functional categories.
GPTers Search MCP Server
Ini adalah server MCP tempat Anda dapat mencari pengetahuan dari komunitas belajar AI GPTers.
Weather MCP Server
Enables users to get real-time weather data for any city and generate broadcast-style weather news reports. Uses the OpenMeteo API to provide current weather conditions including temperature, humidity, and weather descriptions.
MCP SSH Server with Streamable HTTP
Enables remote server control via SSH through Claude, supporting command execution, file transfers, and multi-server management with both standard and streamable HTTP protocols.
mcp-server-zep-cloud
mcp-server-zep-cloud
Aladin Book Search MCP Server
MCP server that allows searching and retrieving book information from Aladin's book store API, including book details, bestseller lists, and category-based searches.