View Live Demo »

StoryWeaver AI was built to make visual storytelling fast, fun, and accessible to everyone, regardless of artistic skill. It transforms the slow, traditional process of creating storyboards into a dynamic, real-time brainstorming session.

Users can create a private room, invite friends with a unique code, and take turns writing text prompts. For each prompt, our AI generates a unique, comic-style illustration, building a visual narrative live for everyone in the room. The application leverages an AI-powered context engine to summarize the story so far, ensuring that new panels are thematically and narratively connected to the previous ones.

• Real-Time Collaboration: New panels appear instantly for all users in a room without needing a refresh, powered by Supabase Realtime.
• AI-Powered Illustration: No art skills required! Simply write a prompt, and the AI generates a comic book-style panel.
• AI Story Context: An AI text model generates a summary of previous panels to provide context for new image generations, creating a more coherent story.
• Turn-Based System: A structured, turn-based system ensures everyone gets a chance to contribute to the story.
• User Authentication & Profiles: Secure user sign-up and login, with customizable usernames for a personal touch.
• Shareable Rooms: Create a room and invite friends with a simple, easy-to-share code.
• Interactive Comic View: View your final creation in a fun, flippable "comic book" format.

This project is a full-stack application composed of a Spring Boot backend and a React frontend.

Backend:

• Java 17 & Spring Boot 3
• Spring Security (for JWT authentication)
• PostgreSQL (managed by Supabase)
• JPA / Hibernate
• Maven

Frontend:

• React (Vite)
• Tailwind CSS
• shadcn/ui for components
• Supabase.js for authentication and real-time subscriptions
• React Router for navigation

Services:

• Supabase: Used for database, authentication, and file storage.
• Render: Hosting for the Spring Boot backend.
• Vercel: Hosting for the React frontend.
• Pollinations.ai: The free, public API for both text and image generation.

To get a local copy up and running, follow these simple steps.

• Java JDK 17 or later
• Maven
• Node.js and npm
• A Supabase account (for database, auth, and storage keys)

1. Clone the repository:

   git clone [https://github.com/karannn3011/swaibackend.git](https://github.com/karannn3011/swaibackend.git)
   cd swaibackend

2. Configure Environment Variables: Your application.properties file contains placeholders for secrets. For local development, you can fill them in directly. For production, these should be set as environment variables.

   • spring.datasource.url: Your Supabase Postgres connection string.
   • spring.datasource.username: Your Supabase username.
   • spring.datasource.password: Your Supabase password.
   • api.supabase.url: Your Supabase project URL.
   • api.supabase.key: Your Supabase service_role key.
   • api.supabase.jwt-secret: Your Supabase JWT secret from the API settings.

3. Run the application:

   ./mvnw spring-boot:run

   The backend will start on http://localhost:8080.

1. Clone the repository:

   git clone [https://github.com/your-username/swai-frontend.git](https://github.com/your-username/swai-frontend.git)
   cd swai-frontend

2. Install NPM packages:

   npm install

3. Configure Environment Variables: Create a .env file in the root of your project.

   VITE_SUPABASE_URL=YOUR_SUPABASE_URL
   VITE_SUPABASE_ANON_KEY=YOUR_SUPABASE_ANON_KEY
   VITE_BACKEND_URL=http://localhost:8080
   

4. Run the application:

   npm run dev

   The frontend will start on http://localhost:5173.

This project relies on free-tier services, which come with certain limitations.

• Backend Cold Start: The backend is deployed on Render's free tier. If the service receives no traffic for 15 minutes, it will spin down to save resources. The next request will trigger a "cold start," which can take 30-60 seconds to complete. During this time, the application may appear unresponsive. This is normal for free-tier hosting.

• Pollinations.ai API Traffic: The AI for both text and image generation is powered by the free and public Pollinations.ai API.

  • High Traffic: As a public service, the API can sometimes be slow or overloaded, which may increase the time it takes to generate a panel.
  • Generation Failures: Occasionally, an API request may fail. The backend has fallbacks in place (e.g., using a simpler context model if the summary generation fails), but image generation is a critical step.

• Supabase Realtime: The real-time updates are generally very fast but depend on a persistent WebSocket connection. Unstable network conditions could cause a temporary delay in seeing updates from other users.

This project has a solid foundation, but there's always more to build!

• Panel Management: Allow users to delete or edit the panels they created.
• Export to PDF: Add a feature to download the final storyboard as a shareable PDF document.
• Style Selection: Allow users to choose from different AI art styles (e.g., "Anime," "Fantasy," "Vintage").
• Improved "Story So Far": Display the full text of the story in a scrollable view in addition to the AI summary.
• Social Logins: Integrate Supabase's social providers (Google, GitHub, etc.) for easier sign-up.