OpenLinkage

🔗 构建健康连接的智能底座 A comprehensive health management platform with multi-agent AI framework and multiple client applications.

---

🚀 Overview / 简介

OpenLinkage is an open-source health management ecosystem that combines:

• 🧠 Multi-Agent AI Framework - Intelligent health assistants powered by AI
• 📱 Multiple Client Applications - Desktop, mobile, web, and WeChat mini-program
• 💾 AI Memory Engine - Short-term and long-term health data intelligence
• 🔒 Privacy-First Architecture - Local data processing and encryption
• 📊 Health Data Integration - Unified platform for all health metrics

---

📁 Project Structure

OpenLinkage/
├── openlinkage/           # Python Multi-Agent AI Framework
│   ├── agents.py          # Health Butler, Nutrition, Exercise, Medication agents
│   └── app.py             # FastAPI application
├── apps/                  # Client Applications
│   ├── desktop/           # Electron desktop app (灵犀健康)
│   ├── mobile/            # Mobile app (iOS/Android via Capacitor)
│   ├── miniprogram/       # WeChat mini program
│   └── web/               # Web applications (doctor/user portals)
├── packages/              # Shared Packages
│   ├── backend/           # Express API server
│   ├── core/              # Core utilities and types
│   └── health-data/       # WellAlly Health data management system
├── docs/                  # Documentation
└── tests/                 # Unit tests

---

🧩 Multi-Agent Framework

The Python-based multi-agent system provides intelligent health analysis:

User
└── Linkage Core
    ├── Health Butler Agent    # Overall health coordination
    ├── Nutrition Agent        # Dietary recommendations
    ├── Exercise Agent         # Fitness planning
    ├── Medication Agent       # Drug safety checks
    ├── Security Agent         # Privacy protection
    └── Memory Engine          # Context-aware intelligence

Quick Start - AI Framework

# Setup Python environment
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
pip install -r requirements.txt

# Run the FastAPI server
python run.py

# Run tests
python -m unittest discover -s tests -v

Visit http://localhost:8000/docs for API documentation.

Example API Request

curl -X POST http://localhost:8000/analyze \
  -H "Content-Type: application/json" \
  -d '{
        "user_id": "demo-user",
        "symptoms": ["fatigue"],
        "goals": ["weight management"],
        "lifestyle_notes": "prefers evening workouts"
      }'

---

📱 Client Applications

Desktop App (灵犀健康)

• Location: apps/desktop
• Tech Stack: Electron, React, tRPC, Express, SQLite
• Features: Full-featured health management with offline support

Mobile App

• Location: apps/mobile
• Tech Stack: Vue 3, Capacitor, Vant UI
• Platforms: iOS, Android

WeChat Mini Program

• Location: apps/miniprogram
• Platform: WeChat ecosystem

Web Applications

• Doctor Portal: apps/web/doctor
• User Portal: apps/web/portal

Quick Start - Client Apps

# Desktop App
cd apps/desktop
pnpm install
pnpm dev

# Mobile App
cd apps/mobile
npm install
npm run dev

# Backend API
cd packages/backend
npm install
npm run dev

---

📦 Shared Packages

Backend API Server

• Location: packages/backend
• Tech: Express, TypeScript, MySQL, Redis
• Features: Authentication, health data CRUD, API endpoints

Core Utilities

• Location: packages/core
• Contents: AI integration, OAuth, notifications, shared types

Health Data System

• Location: packages/health-data
• Features: 28+ health analysis modules powered by Claude AI

---

⚙️ Tech Stack

AI Framework

• Backend: Python, FastAPI, LangGraph
• AI Integration: OpenAI GPT, Qwen, Claude
• Vector DB: pgvector, Qdrant

Client Applications

• Desktop: Electron, React, tRPC, SQLite
• Mobile: Vue 3, Capacitor
• Web: HTML/CSS/JavaScript, Node.js
• Mini Program: WeChat framework

Infrastructure

• Databases: MySQL, SQLite, Redis
• Privacy: Decentralized encryption
• Build Tools: Vite, esbuild, TypeScript

---

📚 Documentation

• Migration Guide - Project reorganization details
• Architecture Overview - System design
• Project Structure - Directory layout

Each application and package has its own README with detailed setup instructions.

---

🔒 Privacy & Security

• Local-first data processing
• End-to-end encryption
• Privacy-preserving AI analysis
• GDPR compliance considerations
• User data ownership

---

🤝 Contributing

We welcome contributions! This project is open source under dual licensing:

• AI Framework: Apache 2.0
• Client Applications: MIT

Please submit issues, feature requests, or pull requests on GitHub.

---

📄 License

• Multi-Agent Framework: Apache 2.0
• Client Applications & Packages: MIT

---

🌟 Features

✅ Multi-agent AI health analysis ✅ Cross-platform client applications ✅ Privacy-first architecture ✅ Comprehensive health data tracking ✅ AI-powered insights and recommendations ✅ Offline-capable desktop and mobile apps ✅ WeChat integration for Chinese users ✅ Open API for third-party integrations

---

© 2025 Linkage Team & OpenLinkage Contributors. All rights reserved.