⚡ AutoFlow AI: The Intelligent Agentic Ecosystem

AutoFlow AI is a professional, low-code platform designed to build, visualize, and deploy autonomous AI agents. By combining a futuristic visual builder with a robust multi-service architecture, AutoFlow enables anyone to create complex automation workflows powered by Google Gemini and local LLMs (via Ollama).

---

🌟 Key Features

• 🧠 Hybrid AI Intelligence: Seamlessly switch between Google Gemini 1.5 Pro for cloud power and Ollama (Qwen3:8B) for local, private inference.
• 🎨 Professional Visual Builder: An infinite-canvas editor built with React Flow, featuring agent-centric topology and dynamic Bezier connections.
• 🔌 Multi-Channel Connectivity: Native bridge for WhatsApp, with upcoming support for Slack, Gmail, and Salesforce.
• 📊 Real-Time Analytics: Built-in ROI dashboard and audit logs to track agent performance and decision-making history.
• 🛠️ Tech-Mode Customization: Toggle between simple AI generation and advanced manual editing for granular control over tool nodes.
• 📦 Data-Driven Actions: Integrated CSV/Spreadsheet support for inventory lookups, payment logging, and automated broadcasting.

---

🏗️ System Architecture

AutoFlow is built on a high-performance microservice architecture, ensuring scalability and ease of deployment.

graph TB
    subgraph "Frontend Layer (React + Vite)"
        UI[Visual Builder UI]
        State[Zustand Store]
        Flow[React Flow Canvas]
    end

    subgraph "Backend Orchestrator (FastAPI)"
        API[REST API Handlers]
        Executor[Workflow Executor Engine]
        DB[(SQLite / SQLModel)]
    end

    subgraph "AI & Intelligence"
        Gemini[Google Gemini 1.5]
        Ollama[Local Ollama Inference]
    end

    subgraph "Communication Bridge"
        WA_Bridge[WhatsApp Bridge Node.js]
        WA_Web[WhatsApp Web Protocol]
    end

    %% Connections
    UI <--> API
    API <--> Executor
    Executor <--> DB
    Executor <--> Gemini
    Executor <--> Ollama
    API <--> WA_Bridge
    WA_Bridge <--> WA_Web

Loading

---

🔄 Agentic Workflow

The following diagram illustrates how messages are processed through the AutoFlow ecosystem:

sequenceDiagram
    participant User as 👤 Customer (WhatsApp)
    participant Bridge as 🔌 WhatsApp Bridge
    participant Backend as ⚙️ Backend Executor
    participant AI as 🧠 LLM (Gemini/Ollama)
    participant Data as 📊 Inventory/CSV

    User->>Bridge: Sends Inquiry ("Is sugar in stock?")
    Bridge->>Backend: Forward Webhook (JSON)
    Backend->>Backend: Match Active Workflow
    Backend->>Data: Lookup Resource (inventory.csv)
    Data-->>Backend: Return Data (Qty: 50)
    Backend->>AI: Contextual Prompt (Data + Inquiry)
    AI-->>Backend: Generated Response ("Yes, we have 50 units!")
    Backend->>Bridge: Send Message Command
    Bridge-->>User: Delivers WhatsApp Message

Loading

---

🛠️ Technology Stack

Component	Technology
Frontend	React 18, Vite, TailwindCSS, React Flow, Zustand, Recharts
Backend	Python 3.11, FastAPI, SQLModel, Uvicorn
Bridges	Node.js, Express, WhatsApp-Web.js (Puppeteer)
Database	SQLite (Production-ready via SQLModel)
AI/ML	Google Gemini API, Ollama (Local LLM)
Deployment	Docker, Docker Compose

---

📂 Project Structure

AutoFlow/
├── autoflow-frontend/     # React + Vite Visual Builder
│   ├── src/               # UI Components & State Management
│   └── public/            # Static Assets
├── backend/               # FastAPI Orchestrator
│   ├── routers/           # API Endpoints
│   ├── executor.py        # Workflow Traversal Engine
│   └── models.py          # Database Schemas
├── whatsapp-bridge/       # Node.js WhatsApp WebSocket Bridge
│   ├── server.js          # Express Webhook Handler
│   └── session/           # WhatsApp Auth States
├── data/                  # Persistent SQLite DB & CSV Resources
└── docker-compose.yml     # Multi-service Orchestration

---

🚀 Quick Start

1. Prerequisites

• Docker & Docker Compose
• Google AI Studio API Key (Optional, for Gemini support)

2. Environment Setup

Create a .env file in the root directory:

GEMINI_API_KEY=your_key_here
WHATSAPP_PORT=3001
BACKEND_PORT=8000

3. Launch the Ecosystem

docker-compose up --build

• Builder UI: http://localhost:3000
• Backend API: http://localhost:8000
• WhatsApp Bridge: http://localhost:3001

---

🤝 Contributing

We welcome contributions! Please fork the repository and submit a pull request for any features or bug fixes.

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

---

📄 License

Distributed under the MIT License. See LICENSE for more information.

Built with ❤️ by the AutoFlow Team