Multimodal AI Agent

A node-based workflow editor for creating AI-powered workflows with multimodal capabilities.

Version Status

Current Version: v1.0.0-stable

This is a known good commit with:

• Working MongoDB integration (Docker, local, and in-memory fallback)
• Fully functional canvas with panning and zooming
• Working authentication system
• All multimodal features operational

Features

• Node-based canvas interface for creating AI workflows
• Support for text, image, audio, and video content
• Integration with OpenAI APIs (GPT-4o, GPT-4.1, GPT-Image)
• Workflow testing and execution
• Document view for consolidated outputs
• Collaboration features

Prerequisites

• Docker
• Docker Compose
• OpenAI API Key

Getting Started

Using Docker (Recommended)

1. Clone the repository:

   git clone https://github.com/yourusername/multimodal-ai-agent.git
   cd multimodal-ai-agent

2. Create a .env file with your OpenAI API key:

   # Server configuration
   PORT=8732
   NODE_ENV=development
   
   # MongoDB connection
   MONGODB_URI=mongodb://mongodb:27017/multimodal-ai-agent
   
   # OpenAI API
   OPENAI_API_KEY=your_openai_api_key_here

3. Start the application with Docker Compose:

   docker-compose up -d

4. Access the application at http://localhost:8732

Without Docker (Local Development)

1. Install MongoDB locally
2. Install Node.js and npm
3. Clone the repository
4. Install dependencies:

   npm install

5. Update the .env file to use your local MongoDB:

   MONGODB_URI=mongodb://localhost:27017/multimodal-ai-agent
   

6. Build CSS and start the application:

   npm run dev

Development

CSS Development

This project uses Tailwind CSS v3 with a proper build process to eliminate browser build warnings and improve performance:

• Development: npm run dev (builds CSS and starts server)
• CSS Watch: npm run dev:css (watches for CSS changes in separate terminal)
• Production Build: npm run build:css:prod

See TAILWIND_SETUP.md for detailed information about the CSS setup and custom classes.

Project Structure

• client/: Frontend code
• controllers/: API controllers
• models/: Database models
• routes/: API routes
• server.js: Main server file
• docker-compose.yml: Docker Compose configuration
• Dockerfile: Docker configuration for the application

Available Scripts

• npm start: Start the server (production)
• npm run dev: Build CSS and start server with nodemon for development
• npm run dev:css: Watch CSS changes (run in separate terminal)
• npm run build:css:prod: Build minified CSS for production
• npm test: Run tests
• npm run lint: Run ESLint
• npm run docker:up: Start with Docker Compose

License

This project is licensed under the MIT License.