<div align="center"> <img src="public/icon.png" alt="OpenFlux Logo" width="160" />

⚡ OpenFlux

Open-source AI Agent desktop client — multi-LLM, long-term memory, browser automation & tool orchestration

English | 中文

</div>

---

✨ Features

• 🧠 Multi-Agent Routing — Auto-detects user intent and dispatches to general / coding / automation agents
• 🔌 Multi-LLM Support — Anthropic / OpenAI / DeepSeek / Moonshot / MiniMax / Zhipu / Google / Ollama, switch with one click
• 💾 Long-term Memory — SQLite + vector search (sqlite-vec), with conversation distillation & knowledge accumulation
• 🌐 Browser Automation — Built-in Playwright for web interaction, data scraping & form filling
• 🛠️ MCP Tool Ecosystem — Model Context Protocol compatible, extensible with Excel, PPT and other external tools
• 🗣️ Voice Interaction — Offline speech recognition (Sherpa-ONNX) + Edge TTS synthesis
• 🔒 Sandbox Isolation — Local code hardening / Docker container isolation for safe code execution
• 🖥️ Desktop Control — Mouse & keyboard simulation, window management, desktop automation
• 📡 Remote Access — Connect to Lark, DingTalk and other platforms via OpenFlux Router
• 🏗️ Tauri v2 — Rust backend + TypeScript frontend, small footprint & high performance

🌐 Ecosystem

OpenFlux is the desktop entry point in the Enterprise AI Assistant ecosystem, working together with NexusAI to build a complete AI workflow system:

┌─────────────────────────────────────────────────────────┐
│               NexusAI (Enterprise Platform)             │
│  Agent Design · Visual Workflows · Knowledge Base       │
└────────────────────────┬────────────────────────────────┘
                         │ Workflows / Agent Config / API Key Distribution
              ┌──────────▼──────────┐
              │  OpenFlux Router    │
              │  Integration Hub    │
              │  Lark/DingTalk/API  │
              └──────────┬──────────┘
                         │ WebSocket
              ┌──────────▼──────────┐
              │   OpenFlux Desktop  │  ← You are here
              │  Local Agent Engine │
              │  Private Workflows  │
              └─────────────────────┘

| Component | Role | Value | |-----------|------|-------| | NexusAI | Enterprise AI Platform | Agent/Workflow design, knowledge management, team collaboration | | OpenFlux Router | Integration Hub | Multi-platform bridging (Lark/DingTalk), unified API key distribution, message routing | | OpenFlux (this project) | Desktop Client | Local agent execution, browser automation, private workflows, long-term memory |

OpenFlux can also run standalone — no NexusAI or Router required. Just configure your own API keys and you're good to go.

🏗️ Architecture

┌─────────────────────────────┐
│       Tauri v2 Shell        │  ← Rust process management + native APIs
├─────────────────────────────┤
│   Frontend (TypeScript/HTML)│  ← Chat UI / Settings / File preview
├─────────────────────────────┤
│   Gateway Sidecar (Node.js) │  ← AI engine / Tool calls / Memory system
└─────────────────────────────┘

🚀 Quick Start

Prerequisites

• Node.js >= 20
• pnpm >= 10
• Rust (stable)
• Tauri v2 CLI: cargo install tauri-cli --version "^2"

Installation

# Clone the repository
git clone https://github.com/EDEAI/OpenFlux.git
cd OpenFlux

# Install frontend dependencies
pnpm install

# Install Gateway dependencies
cd gateway && npm install && cd ..

# Build Gateway
# (see scripts/build-gateway.ps1)

Configuration

# Copy the config template
cp openflux.example.yaml openflux.yaml

# Edit openflux.yaml and add your API keys
# At least one LLM provider is required

Development

pnpm tauri dev

Build

pnpm tauri build

⚙️ Configuration

All settings are in openflux.yaml. See openflux.example.yaml for reference:

| Section | Description | |---------|-------------| | providers | LLM provider API keys and endpoints | | llm | Orchestration / execution / embedding / fallback model selection | | memory | Long-term memory toggle, vector dimensions, distillation strategy | | agents | Multi-agent routing and tool permissions | | browser | Browser automation settings | | voice | Speech recognition & synthesis | | sandbox | Code execution isolation | | web | Search (Brave/Perplexity) & web scraping | | mcp | External MCP tool servers |

📁 Project Structure

OpenFlux/
├── src/              # Frontend TypeScript (UI / interaction)
├── src-tauri/        # Rust backend (Tauri Shell)
│   └── src/          # Rust source code
├── gateway/          # Gateway Sidecar (AI engine)
│   └── src/          # TypeScript source code
├── public/           # Static assets
├── resources/        # Model files
├── scripts/          # Build scripts
└── openflux.example.yaml  # Config template

🤝 Contributing

Contributions are welcome! Feel free to open issues and pull requests.

1. Fork this repository
2. Create your feature branch: git checkout -b feature/amazing-feature
3. Commit your changes: git commit -m 'Add amazing feature'
4. Push to the branch: git push origin feature/amazing-feature
5. Open a Pull Request

📄 License

This project is licensed under the MIT License.