dotfit-mcp-desktop

# DOTfit Desk — MCP Integration ## 🚀 What is this? **DOTfit Desk — MCP** is a desktop helper toolkit that bridges a local GUI tool with an MCP (Model Context Protocol) server demo. The goal of this project is to showcase how a *user desktop interface* can interact with an MCP backend to fetch results and display them locally. The MCP demo included is based on a lightweight Python server. This repository is **not just a notebook** — it's an app + a server designed to demonstrate a full cycle: 🖥️ Desktop app → ⚙️ MCP server → 📦 result output --- ## 📌 Why this matters Most AI demos are limited to *scripts or Jupyter notebooks*. This repo is structured as a **mini production flow**: - UI/Frontend (HTML/CSS/JS) - Backend server (MCP demo) - IPC / communication pattern - Real runnable application This is a good base for: ✅ AI assistant UIs ✅ Project templates for desktop-to-server communication ✅ Prototyping logic with Python + MCP --- ## 🧠 Features - 💻 **Desktop UI** (HTML/CSS frontend) - 📡 **MCP demo server** written in Python - 🔗 Communication bridge between UI and backend - 🧪 Demo workflow included --- ## 🔧 Requirements Install Python 3.8+: ```bash python --version ``` Install dependencies: ```bash pip install -r requirements.txt ``` --- ## 🏃 How to Run ### 1️⃣ Start the MCP demo server ```bash cd mcp-server-demo python main.py ``` This spins up the backend that the desktop app talks to. --- ### 2️⃣ Open the Desktop UI Open `index.html` in your browser or local desktop app shell. ```bash open index.html ``` --- ### 🔍 How It Works ``` [Desktop UI] <– fetch requests –> [MCP Server] ``` The UI sends structured requests to the local MCP server. The server responds with processed output. This pattern can be extended to: ✔ External APIs ✔ AI models ✔ Local OS interactions --- ## 📁 Repo Structure ``` / ├── mcp-server-demo 🚀 Python MCP backend ├── Final_DOTfitDesk_exe 🧰 Executable build of the desktop UI ├── README.md 📘 You're here ├── assets/ 🖼 UI resources └── index.html 🧩 Desktop entry ``` --- ## 📌 Improvements You Could Add If you want to level this up: ✔ Add a real AI backend (FastAPI + GPT-based logic) ✔ Package the UI into an Electron/desktop app ✔ Dockerize the MCP server ✔ Add tests and CI/CD ✔ Provide real use cases (OCR, Chat UI, tool integration) --- ## 📞 Contributing If you build something cool on top of this: - Create an issue - Submit a friendly pull request - Add your demo link in the README