MCP Servers

模型上下文协议服务器、框架、SDK 和模板的综合目录。

MCP server by Super-MCP-App

创建于 3/21/2026
更新于 about 6 hours ago
Repository documentation and setup instructions
Super-MCP-App Logo

✨ Super-MCP-App ✨

Your Intelligent, Native Mobile AI Assistant

Contributors Forks Stars Issues License

A fast, sleek, and highly capable mobile AI chat application built with React Native (Expo) and Next.js, powered by Meta LLaMA 3.1 and Supabase.


🌟 Key Features

  • 💬 Real-Time Conversational AI – Speak with an intelligent agent powered by the LLaMA 3.1 8B model via NVIDIA NIM.
  • 📱 Native Mobile Experience – Built on React Native & Expo for smooth, native-like iOS and Android performance.
  • 🎨 Modern Material 3 Design – Beautifully constructed interfaces with @react-native-paper utilizing customized color tokens.
  • 📝 Markdown Support – Full rich-text rendering of AI responses, including syntax-highlighted code blocks (react-native-markdown-display).
  • 🔄 Swipe-to-Delete – Fluid, gesture-based conversation management right from your chat list (Powered by react-native-gesture-handler).
  • 🔐 Secure & Serverless Backend – Next.js acts as an API gateway safely routing AI requests, backed by Supabase for authentication and persistent chat storage.

🛠️ Technology Stack

App (Frontend)

  • Framework: React Native & Expo
  • Navigation: React Navigation (Bottom Tabs & Stack Navigators)
  • UI/Styling: React Native Paper
  • Gestures: React Native Gesture Handler
  • Markdown: React Native Markdown Display

API & Database (Backend)

  • API Framework: Next.js 16 (App Router)
  • Database & Auth: Supabase (PostgreSQL with Row-Level Security)
  • AI Engine: NVIDIA API (OpenAI-compatible endpoints running meta/llama-3.1-8b-instruct)

🏗️ System Architecture

The project follows a modern, decoupled architecture powered by the Model Context Protocol (MCP), connecting external tool providers securely to the AI model without exposing credentials to the mobile client.

graph TD
    %% Frontend Layer
    subgraph Mobile_Client ["📱 Mobile App (React Native/Expo)"]
        UI["React Native UI"]
        State["State & API Layers"]
    end

    %% Backend Layer (Next.js API)
    subgraph API_Gateway ["⚙️ Next.js Serverless Backend"]
        AuthAPI["/api/auth"]
        ChatAPI["/api/conversations"]
        
        subgraph MCP_Architecture ["🔌 Model Context Protocol (MCP) Integration"]
            MessageAPI["/api/messages"]
            MCPClient["MCP Client Engine"]
            Tools["Tool & Resource Registry"]
        end
    end

    %% External Infrastructure
    subgraph External_Services ["🌐 Core Infrastructure & AI"]
        SupaDB[("Supabase (PostgreSQL)")]
        SupaAuth{"Supabase Auth"}
        Nvidia(("NVIDIA NIM (LLaMA 3.1)"))
    end

    %% MCP Providers Layer
    subgraph MCP_Providers ["🛠️ MCP Servers (Providers)"]
        FigmaMCP["Figma MCP Server"]
        CanvaMCP["Canva MCP Server"]
        CustomMCP["Custom External Data MCPs"]
    end

    %% Mobile to API Routing
    State <--> |"REST"| AuthAPI
    State <--> |"REST"| ChatAPI
    State <--> |"REST"| MessageAPI

    %% Backend to Infra Routing
    AuthAPI <--> |"Authentication"| SupaAuth
    ChatAPI <--> |"Storage & Retrieval"| SupaDB
    MessageAPI <--> |"RLS Authenticated Reads"| SupaDB
    
    %% The MCP Flow Model
    MessageAPI --> MCPClient
    MCP_Providers <--> |"Standardized Tool/Resource Context (MCP)"| MCPClient
    MCPClient --> Tools
    Tools <--> |"Injected Tools & Function Calls"| Nvidia

🚀 Getting Started

Follow these steps to get the app running locally on your machine and simulator.

Prerequisites

Before you begin, ensure you have the following installed and configured:

1. Setup Backend

The backend utilizes Next.js App Router as a secure API Gateway.

# Navigate to the backend directory
cd backend

# Install dependencies
npm install

Create a .env file in the backend/ directory matching your configuration details:

NEXT_PUBLIC_SUPABASE_URL=your_supabase_url
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_supabase_anon_key
SUPABASE_SERVICE_ROLE_KEY=your_supabase_service_role_key

# NVIDIA AI Configuration
NVIDIA_API_KEY=your_nvidia_nim_api_key
NVIDIA_BASE_URL=https://integrate.api.nvidia.com/v1

# Local Network App Configuration
NEXT_PUBLIC_API_URL=http://127.0.0.1:3000

Start the API development server:

npm run dev

2. Setup Mobile App

In the root project directory, install the Expo packages:

# Return to the root folder
cd ..

# Install app dependencies
npm install

# Start the Expo bundler
npx expo start

From the Expo terminal, press i to open the iOS Simulator or scan the QR code using the Expo Go app on your physical device.


📦 Project Structure

Super-MCP-App/
├── src/                # Front-end React Native App
│   ├── navigation/     # Tab and Stack Navigators
│   ├── screens/        # React Native Screens (Home, Chat, Profile, etc.)
│   ├── services/       # API Fetch Handlers & Supabase configs
│   └── theme/          # UI themer (Colors, Typography)
├── backend/            # Next.js Backend API
│   ├── app/api/        # Serverless Routes (Messages, Conversations, Auth)
│   └── lib/            # NVIDIA SDK handlers and Supabase Admin
└── README.md           

🤝 Contributing

We welcome contributions from the community! Whether it's adding new features, fixing bugs, or improving documentation, your help makes Super-MCP-App better for everyone.

Please read our Contributing Guidelines to get started with setting up your development environment and submitting Pull Requests.


🐛 Issues and Bug Reports

Found a bug or have a great idea for a new feature? We'd love to hear about it! Please use our Issue Tracker to report bugs or request features. We have templates prepared to help you provide all the necessary information.


Designed for fluidity, powered by modern AI.

Made with ❤️ by the Super-MCP-App Open Source Organization.
快速设置
此服务器的安装指南

安装包 (如果需要)

npx @modelcontextprotocol/server-super-mcp-app

Cursor 配置 (mcp.json)

{ "mcpServers": { "super-mcp-app-super-mcp-app": { "command": "npx", "args": [ "super-mcp-app-super-mcp-app" ] } } }