Discover Awesome MCP Servers

SureChEMBL MCP Server

A comprehensive Model Context Protocol (MCP) server for accessing the SureChEMBL chemical patent database.

🚀 MCPverse

MCPverse adalah alat pencarian dan penemuan untuk server MCP (Model Context Provider) yang dihosting di GitHub. Alat ini memberi peringkat dan menampilkan repositori dengan konfigurasi klien dalam UI yang bersih dan interaktif.

INTERNAL TEST ONLY - will be ready soon

Here are a few ways to set up a public MCP (Modelica Communication Protocol) server to run Modelica simulations, along with considerations for each: **1. Using OpenModelica (Recommended for Open Source and Flexibility)** * **OpenModelica Compiler (OMC) with MCP:** OpenModelica has built-in support for MCP. You can configure OMC to act as an MCP server. * **Steps:** 1. **Install OpenModelica:** Download and install the latest version of OpenModelica from [https://openmodelica.org/](https://openmodelica.org/). 2. **Start OMC in MCP Mode:** You'll typically do this from the command line. The exact command depends on your operating system and OpenModelica installation, but it will look something like this: ```bash omc +mcp ``` This starts the OpenModelica compiler in MCP mode, listening for connections on a default port (usually 5000). You can specify a different port using the `-port` option: ```bash omc +mcp -port=1234 ``` 3. **Firewall Configuration:** Crucially, you need to open the port you've specified (e.g., 5000 or 1234) in your server's firewall to allow external connections. This is a critical security consideration. 4. **Public IP Address:** Your server needs a public IP address that clients can use to connect. If your server is behind a router, you'll need to configure port forwarding on the router to forward traffic on the chosen port to your server's internal IP address. 5. **Security Considerations:** This is the most important part. Running an open MCP server is *inherently insecure* if not properly managed. Anyone who can connect to your server can execute arbitrary Modelica code, which can potentially be used to compromise your system. **Do not expose an MCP server directly to the internet without implementing strong security measures.** * **Authentication:** OpenModelica's MCP doesn't have built-in authentication. You'll need to implement your own authentication layer *outside* of OpenModelica. This could involve: * Using a reverse proxy (like Nginx or Apache) to handle authentication before forwarding requests to OMC. * Writing a custom script that intercepts MCP requests, authenticates the user, and then forwards the request to OMC if authentication is successful. * **Sandboxing:** Ideally, you should run OMC in a sandboxed environment (e.g., using Docker) to limit the damage that a malicious user could do. * **Resource Limits:** Set limits on the resources (CPU, memory, disk space) that OMC can use to prevent denial-of-service attacks. * **Rate Limiting:** Limit the number of requests that can be made to the server in a given time period. * **Regular Updates:** Keep OpenModelica updated to the latest version to patch security vulnerabilities. * **Example Client Code (Python):** ```python import socket def send_mcp_command(host, port, command): with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.connect((host, port)) s.sendall(command.encode()) data = s.recv(1024) # Adjust buffer size as needed return data.decode() if __name__ == "__main__": host = "your_server_ip" # Replace with your server's IP address port = 5000 # Replace with the port you're using command = "loadModel(Modelica.Blocks.Examples.PID_Controller);" response = send_mcp_command(host, port, command) print(f"Received: {response}") ``` * **Advantages:** * Open Source: Free to use and modify. * Flexible: You have full control over the configuration and security. * Active Community: Good support and documentation. * **Disadvantages:** * Security: Requires significant effort to secure properly. This is *not* a task for beginners. * Complexity: Setting up and managing the server can be complex. **2. Using Dymola (Commercial Option)** * **Dymola with MCP:** Dymola also supports MCP. The setup is similar to OpenModelica, but the configuration details are specific to Dymola. Refer to the Dymola documentation for instructions. * **Advantages:** * Commercial Support: You get support from Dassault Systèmes. * Mature Tool: Dymola is a well-established Modelica simulation environment. * **Disadvantages:** * Cost: Dymola is a commercial product and requires a license. * Security: Still requires careful security considerations, although Dymola might have some built-in security features (check the documentation). **3. Using a Cloud-Based Solution (Potentially Easier, but Still Requires Security)** * **Deploy OpenModelica or Dymola on a Cloud Server (AWS, Azure, Google Cloud):** You can rent a virtual machine (VM) on a cloud platform and install OpenModelica or Dymola on it. This gives you more control over the server environment. * **Advantages:** * Scalability: Cloud platforms offer scalability, so you can easily increase resources if needed. * Infrastructure Management: The cloud provider handles much of the infrastructure management. * **Disadvantages:** * Cost: Cloud resources can be expensive. * Complexity: Still requires setting up and configuring the server. * Security: You are still responsible for securing the server and the MCP endpoint. Cloud providers offer security tools, but you need to configure them correctly. **Important Security Considerations (Repeated for Emphasis):** * **Never expose an MCP server directly to the internet without strong security measures.** * **Implement authentication.** * **Use sandboxing (e.g., Docker).** * **Set resource limits.** * **Implement rate limiting.** * **Keep your software updated.** * **Consider using a reverse proxy with authentication.** **In summary, setting up a public MCP server is a complex task that requires careful planning and execution, especially regarding security. If you are not experienced with server administration and security, it is strongly recommended to seek help from a qualified professional.** Here's the translation to Indonesian: **Cara Menyiapkan Server MCP Publik untuk Menjalankan Simulasi Modelica** Berikut adalah beberapa cara untuk menyiapkan server MCP (Modelica Communication Protocol) publik untuk menjalankan simulasi Modelica, beserta pertimbangan untuk masing-masing cara: **1. Menggunakan OpenModelica (Direkomendasikan untuk Sumber Terbuka dan Fleksibilitas)** * **OpenModelica Compiler (OMC) dengan MCP:** OpenModelica memiliki dukungan bawaan untuk MCP. Anda dapat mengonfigurasi OMC untuk bertindak sebagai server MCP. * **Langkah-langkah:** 1. **Instal OpenModelica:** Unduh dan instal versi terbaru OpenModelica dari [https://openmodelica.org/](https://openmodelica.org/). 2. **Jalankan OMC dalam Mode MCP:** Anda biasanya melakukan ini dari baris perintah. Perintah yang tepat tergantung pada sistem operasi dan instalasi OpenModelica Anda, tetapi akan terlihat seperti ini: ```bash omc +mcp ``` Ini memulai kompilator OpenModelica dalam mode MCP, mendengarkan koneksi pada port default (biasanya 5000). Anda dapat menentukan port yang berbeda menggunakan opsi `-port`: ```bash omc +mcp -port=1234 ``` 3. **Konfigurasi Firewall:** Sangat penting, Anda perlu membuka port yang telah Anda tentukan (misalnya, 5000 atau 1234) di firewall server Anda untuk mengizinkan koneksi eksternal. Ini adalah pertimbangan keamanan yang kritis. 4. **Alamat IP Publik:** Server Anda membutuhkan alamat IP publik yang dapat digunakan klien untuk terhubung. Jika server Anda berada di belakang router, Anda perlu mengonfigurasi penerusan port (port forwarding) pada router untuk meneruskan lalu lintas pada port yang dipilih ke alamat IP internal server Anda. 5. **Pertimbangan Keamanan:** Ini adalah bagian terpenting. Menjalankan server MCP terbuka *secara inheren tidak aman* jika tidak dikelola dengan benar. Siapa pun yang dapat terhubung ke server Anda dapat menjalankan kode Modelica arbitrer, yang berpotensi digunakan untuk membahayakan sistem Anda. **Jangan mengekspos server MCP langsung ke internet tanpa menerapkan langkah-langkah keamanan yang kuat.** * **Autentikasi:** MCP OpenModelica tidak memiliki autentikasi bawaan. Anda perlu menerapkan lapisan autentikasi Anda sendiri *di luar* OpenModelica. Ini bisa melibatkan: * Menggunakan reverse proxy (seperti Nginx atau Apache) untuk menangani autentikasi sebelum meneruskan permintaan ke OMC. * Menulis skrip khusus yang mencegat permintaan MCP, mengautentikasi pengguna, dan kemudian meneruskan permintaan ke OMC jika autentikasi berhasil. * **Sandboxing:** Idealnya, Anda harus menjalankan OMC di lingkungan sandbox (misalnya, menggunakan Docker) untuk membatasi kerusakan yang dapat dilakukan oleh pengguna jahat. * **Batas Sumber Daya:** Tetapkan batasan pada sumber daya (CPU, memori, ruang disk) yang dapat digunakan OMC untuk mencegah serangan penolakan layanan (denial-of-service). * **Pembatasan Laju (Rate Limiting):** Batasi jumlah permintaan yang dapat dibuat ke server dalam jangka waktu tertentu. * **Pembaruan Reguler:** Jaga agar OpenModelica tetap diperbarui ke versi terbaru untuk menambal kerentanan keamanan. * **Contoh Kode Klien (Python):** ```python import socket def send_mcp_command(host, port, command): with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.connect((host, port)) s.sendall(command.encode()) data = s.recv(1024) # Sesuaikan ukuran buffer sesuai kebutuhan return data.decode() if __name__ == "__main__": host = "your_server_ip" # Ganti dengan alamat IP server Anda port = 5000 # Ganti dengan port yang Anda gunakan command = "loadModel(Modelica.Blocks.Examples.PID_Controller);" response = send_mcp_command(host, port, command) print(f"Diterima: {response}") ``` * **Keuntungan:** * Sumber Terbuka: Gratis untuk digunakan dan dimodifikasi. * Fleksibel: Anda memiliki kendali penuh atas konfigurasi dan keamanan. * Komunitas Aktif: Dukungan dan dokumentasi yang baik. * **Kekurangan:** * Keamanan: Membutuhkan upaya signifikan untuk mengamankan dengan benar. Ini *bukan* tugas untuk pemula. * Kompleksitas: Menyiapkan dan mengelola server bisa jadi rumit. **2. Menggunakan Dymola (Opsi Komersial)** * **Dymola dengan MCP:** Dymola juga mendukung MCP. Pengaturannya mirip dengan OpenModelica, tetapi detail konfigurasinya khusus untuk Dymola. Lihat dokumentasi Dymola untuk instruksi. * **Keuntungan:** * Dukungan Komersial: Anda mendapatkan dukungan dari Dassault Systèmes. * Alat yang Matang: Dymola adalah lingkungan simulasi Modelica yang mapan. * **Kekurangan:** * Biaya: Dymola adalah produk komersial dan membutuhkan lisensi. * Keamanan: Masih membutuhkan pertimbangan keamanan yang cermat, meskipun Dymola mungkin memiliki beberapa fitur keamanan bawaan (periksa dokumentasinya). **3. Menggunakan Solusi Berbasis Cloud (Potensial Lebih Mudah, tetapi Tetap Membutuhkan Keamanan)** * **Deploy OpenModelica atau Dymola di Server Cloud (AWS, Azure, Google Cloud):** Anda dapat menyewa mesin virtual (VM) di platform cloud dan menginstal OpenModelica atau Dymola di atasnya. Ini memberi Anda lebih banyak kendali atas lingkungan server. * **Keuntungan:** * Skalabilitas: Platform cloud menawarkan skalabilitas, sehingga Anda dapat dengan mudah meningkatkan sumber daya jika diperlukan. * Manajemen Infrastruktur: Penyedia cloud menangani sebagian besar manajemen infrastruktur. * **Kekurangan:** * Biaya: Sumber daya cloud bisa mahal. * Kompleksitas: Masih membutuhkan pengaturan dan konfigurasi server. * Keamanan: Anda masih bertanggung jawab untuk mengamankan server dan endpoint MCP. Penyedia cloud menawarkan alat keamanan, tetapi Anda perlu mengonfigurasinya dengan benar. **Pertimbangan Keamanan Penting (Diulang untuk Penekanan):** * **Jangan pernah mengekspos server MCP langsung ke internet tanpa langkah-langkah keamanan yang kuat.** * **Terapkan autentikasi.** * **Gunakan sandboxing (misalnya, Docker).** * **Tetapkan batas sumber daya.** * **Terapkan pembatasan laju (rate limiting).** * **Jaga agar perangkat lunak Anda tetap diperbarui.** * **Pertimbangkan untuk menggunakan reverse proxy dengan autentikasi.** **Singkatnya, menyiapkan server MCP publik adalah tugas yang kompleks yang membutuhkan perencanaan dan pelaksanaan yang cermat, terutama mengenai keamanan. Jika Anda tidak berpengalaman dengan administrasi server dan keamanan, sangat disarankan untuk mencari bantuan dari profesional yang berkualifikasi.**

Lens

MCP server for visual feedback, video direction, and QA assertions on web pages, enabling AI agents to read, reply, and resolve annotations in real time.

Apple Health MCP Server

A Model Context Protocol server that enables seamless interaction between LLM-based agents and Apple Health data, allowing users to query, analyze, and manage health records through natural language commands.

my-mcp-server

A Model Context Protocol server that allows integration with Claude Desktop by creating and managing custom tools that can be executed through the MCP framework.

wso2-bi-salesforce-mcp-server

Enables AI assistants to set up, run, and manage Ballerina + Salesforce integrations inside a WSO2 Integrator workspace, providing 18 tools for OAuth2 token acquisition, project scaffolding, building, deploying, and stopping the integration service.

memory-bridge

Cross-project memory sharing for Claude Code and Cowork. Stop teaching Claude the same thing twice.

ModelScope Image MCP Server

Enables users to generate high-quality images using ModelScope's Qwen-Image model through natural language prompts. Supports async task processing with both image URL and base64 encoded data output options.

Todo Markdown MCP Server

An MCP server that allows AI assistants to manage todo lists stored in a simple markdown file, supporting creation, reading, updating, and deletion of todo items with persistent IDs.

paper-intelligence

A local-first database and assistant layer for organizing, analyzing, and retrieving research papers efficiently.

milele-prime-mcp

Read-only MCP server that lets a verified client connect their own AI to their trading account to read account data without order execution.

SearchAPI MCP Agent

A bridge between AI assistants and search services that implements A2A protocol and Model Context Protocol, enabling search functionality for web, videos, maps, flights, and hotels.

Microsoft 365 MCP Server

A Model Context Protocol (MCP) server for interacting with Microsoft 365 and Microsoft Office services through the Graph API.

ADB MCP Server

A TypeScript-based bridge between AI models and Android device functionality, enabling interaction with Android devices through ADB commands for tasks like app installation, file transfer, UI analysis, and shell command execution.

AWS Athena MCP Server

Enables AI assistants to execute SQL queries against AWS Athena databases, check query status, retrieve results, and manage saved queries with support for both local and remote deployment via Lambda.

Ambient Code Platform MCP Server

Enables delegation of agentic sessions to Kubernetes-hosted Claude agents running on the Ambient Code Platform. Supports creating, managing, and communicating with remote AI agent sessions through OpenShift authentication.

realestate-mcp

MCP server for real estate agents with 12 tools covering floor plans, virtual tours, property descriptions (EN/FR), price estimation, property comparison, staging suggestions, neighborhood analysis, 3D previews, French DPE energy diagnostic, and live DVF transaction data.

ibmi-mcp-server

An MCP server that enables AI assistants to interact with IBM i AS/400 source members for source code management, compilation, and development workflows.

WeatherXM Pro MCP Server

Provides comprehensive access to real-time weather data, forecasts, historical data, and weather alerts from the WeatherXM decentralized weather network worldwide.

OpenVan MCP Server

OpenVan MCP Server exposes OpenVan.camp public APIs over the Model Context Protocol: fuel prices across 125+ countries, VanSky weather suitability scores (0–100), VanBasket food price index, currency exchange rates, vanlife events (expos, festivals, meetups) and aggregated news stories in 7 languages. 11 read-only tools, no authentication, 120 req/min rate limit, CC BY 4.0 license. Remote Streamab

Splits MCP Server

Enables managing Splits transactions, accounts, signers, and authentication through natural language or AI assistants.

NASA SciX MCP Server

Enables LLMs to search astronomical literature, retrieve paper metadata, analyze citation metrics, and export bibliographic data via the NASA Astrophysics Data System (SciX) API.

Kimi Memory MCP

为Kimi Code CLI提供持久化记忆系统，支持文件学习、智能检索和混合存储，让大模型拥有长效记忆能力。

ai-price-index-mcp

Enables coding agents to query AI model API prices, including historical point-in-time lookups with cited sources, using a bundled dated dataset and requiring no API keys.

MCP Nano Banana

Enables image generation using the Google Gemini API through a simple text prompt. Generated images are saved locally and hosted online via ImgBB for easy access and sharing.

СДАМ ГИА MCP Server

Enables LLMs to search and retrieve exam problems, solutions, and answers from the СДАМ ГИА educational platform across multiple subjects. It supports fuzzy text matching, catalog browsing, and structured data retrieval to assist with academic study and test preparation.

Alexandria MCP

Provides access to 61 public digital libraries through a single unified interface, enabling users to search and retrieve information from academic papers, books, legal records, and more using natural language.

audio-transcription-mcp

Captures and transcribes system audio in real-time using OpenAI Whisper, enabling meeting transcription, content creation, and accessibility through natural language.