Director.AI

Faceless video creation platform powered by real AI CLI integration. Built with React, TypeScript, Express, MongoDB, and Socket.IO.

Quick Setup Tutorial

Follow these steps to get Director.AI running on your local machine.

1. Clone the Repository

git clone https://huggingface.co/algorembrant/1st-hackaton
cd 1st-hackaton

2. Install MongoDB and Compass

Download MongoDB Community Server: Visit the MongoDB Download Center and download the MSI installer for Windows.
Install: Run the installer and ensure "Install MongoDB as a Service" is checked.
MongoDB Compass: During installation, check the box to install MongoDB Compass (the GUI for managing your data).
Verify: Open MongoDB Compass and connect to mongodb://localhost:27017.

3. Environment Configuration

Create a .env file in the root directory by copying the example:
```
cp .env.example .env
```
Open .env and fill in the following:
- MONGODB_URI: mongodb://localhost:27017/director-ai
- PORT: 5000
- JWT_SECRET: Any random long string
- REDIS_URL: redis://localhost:6379 (If using Redis features)
or simply just tell you AI agent to configure things for you.

4. Install Dependencies

npm install

5. Running the Application

Server: cd server && npm run dev
Client: cd client && npm run dev
Both: From the root, run npm run dev.

Overview

Director.AI allows users to input scripts, customize voices, music, visual styles, and assets, then generate optimized faceless videos for social platforms (TikTok, Reels, Shorts, YouTube).

The application features a novel real-time AI Terminal that directly bridges the webapp to the Google Antigravity CLI, streaming output via WebSocket—no simulations, no mocks.

Key Features

AI Terminal: Real-time CLI bridge via WebSocket—type prompts in the webapp, get live output from the Antigravity CLI.
5-Step Video Wizard: Script, Voice, Music, Style, Assets → Generate.
Multi-Platform Export: 9:16 (TikTok/Reels/Shorts), 16:9 (YouTube), 1:1 (Instagram).
JWT Authentication: Secure register/login with bcrypt hashing.
Luxury Dark Theme: Gold accents, Playfair Display + Montserrat, glassmorphism, Framer Motion animations.
Style Presets: Save and reuse brand configurations.
File Upload: Multer with 100MB limit and file type filtering.
Admin Dashboard: Queue monitoring and stats overview.

System Architecture

graph TD
    A["User Browser"] -->|HTTP + WebSocket| B["Express Server :5000"]
    B -->|Mongoose| C[MongoDB]
    B -->|Socket.IO| D["WebSocket Layer"]
    D -->|child_process.spawn| E["Antigravity CLI"]
    E -->|stdout/stderr stream| D
    D -->|Real-time output| A
    B -->|Multer| F["File Storage /uploads"]
    B -->|"Generated"| G["/generated"]
    
    subgraph Frontend [":5173"]
        H["React + TypeScript"]
        I["Redux Toolkit"]
        J["Framer Motion"]
        K["Tailwind CSS"]
        L["AI Terminal Component"]
    end
    
    subgraph Backend [":5000"]
        M["JWT Auth"]
        N["REST API Routes"]
        O["CLI Bridge Service"]
        P["Rate Limiting + Helmet"]
    end
    
    A --> H
    H --> I
    H --> L
    L -->|Socket.IO Client| D

Project Structure

1st-hackaton/
├── .env.example
├── .gitignore
├── package.json
├── prompt.md
├── README.md
├── server/
│   ├── src/
│   │   ├── index.ts        # Main Entry
│   │   ├── config/         # Database & App Config
│   │   ├── models/         # Mongoose Schemas
│   │   ├── routes/         # API Endpoints
│   │   ├── services/       # Business Logic
│   │   └── utils/          # Helpers
│   └── tsconfig.json
└── client/
    ├── src/
    │   ├── components/     # UI Building Blocks
    │   ├── pages/          # Main Views
    │   ├── services/       # API Integration
    │   └── store/          # Redux State Management
    ├── vite.config.ts
    └── tailwind.config.js

Hugging Face Deployment

This project is optimized for deployment on Hugging Face.

Repository: https://huggingface.co/algorembrant/1st-hackaton
Author Profile: https://huggingface.co/algorembrant

Large assets or binary files should be handled via Xet Storage or Git LFS to ensure optimal performance on Hugging Face.

Author

Rembrant Oyangoren Albeos

Hugging Face: @algorembrant

License

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support