ChronicleAI - Automated Multi-Modal Content Pipeline

ChronicleAI is a high-performance content orchestration platform designed to ingest long-form text (light novel chapters, scripts, blog posts) and automatically compile serialized visual storyboard scripts, detailed AI art prompts, and chronological video timeline assets using the Gemini Flash model.

Technical Stack

• Frontend: Next.js (App Router, TypeScript) + Tailwind CSS v4 (Glassmorphic Dark UI)
• API Backend: FastAPI (Python) asynchronous worker threads
• Database: Supabase (PostgreSQL) + Prisma ORM v7
• LLM Engine: Google Gemini API (gemini-flash-latest)

---

🛠️ Installation & Setup

Follow these steps to configure the environment and run both servers locally.

Prerequisites

• Node.js (v18.x or newer)
• Conda or Python 3.10+ installed

---

1. Database Configuration (Supabase + Prisma v7)

Prisma v7 manages database configurations via a centralized configuration script prisma.config.ts rather than the schema file directly.

1. Navigate to the frontend/ directory:

   cd frontend

2. Make sure your .env contains the Supabase connection strings:

   # Connect to Postgres via the shared transaction-mode pooler (pgbouncer port 6543)
   DATABASE_URL="postgresql://postgres.ztjeguzdbmmupuysoilw:[YOUR_PASSWORD]@aws-1-ap-south-1.pooler.supabase.com:6543/postgres?pgbouncer=true"
   
   # Connect to Postgres via the direct session-mode pooler (port 5432 - used for schema alterations)
   DIRECT_URL="postgresql://postgres.ztjeguzdbmmupuysoilw:[YOUR_PASSWORD]@aws-1-ap-south-1.pooler.supabase.com:5432/postgres"
   
   # Backend FastAPI connection endpoint
   NEXT_PUBLIC_API_URL="http://localhost:8000"

3. Push the schema directly to your Supabase instance:

   npx prisma db push

4. Generate the strongly-typed Prisma Client:

   npx prisma generate

---

2. API Backend Setup (FastAPI)

1. Navigate to the backend/ directory:

   cd backend

2. Create and activate the conda environment chronicleai:

   conda create -n chronicleai python=3.10 -y
   conda activate chronicleai

3. Install Python dependencies:

   pip install -r requirements.txt

4. Configuration: The backend environment settings are located in app/config.py. It is pre-configured with:
   • Supabase connection string (DATABASE_URL).
   • Gemini API URL & Key ([YOUR_GEMINI_API_KEY]).
5. Run the FastAPI development server:

   python run.py

   The backend API will run on http://localhost:8000. You can inspect the interactive OpenAPI docs at http://localhost:8000/docs.

---

3. Frontend Setup (Next.js + Tailwind v4)

1. Navigate back to the frontend/ directory:

   cd frontend

2. Install npm packages:

   npm install

3. Start the Next.js development server:

   npm run dev

   Open http://localhost:3000 in your web browser to enter the ChronicleAI workspace dashboard.

---

🚀 Features & Workspace Guide

1. Dashboard Feed: Ingest text documents by inputting a title and pasting raw chapters. Use the "Load Sample Script" helper in the ingestion modal to automatically load a beautifully structured test story chapter for immediate evaluation.
2. Storyboard Workspace: Click any active project card to view your workspace. If the pipeline task is running (PENDING/PROCESSING), the client automatically updates scene cards, pacing logs, and timing elements progressively.
3. Multi-Tab Preview Panels:
   • Storyboard Tab: Displays detailed narrative scene grids, visual composition cues, set/character tags, and art prompts with single-click copy actions.
   • Timeline Tab: Displays serialized timelines matching voiceover scripts and narration cue times (e.g. 0.0s - 6.5s) in an interactive feed.
   • Image Prompts Tab: Consolidates all generated art prompts in a streamlined code list for quick batch copying into generator systems.
   • Input Text Tab: Reference page for reading original raw drafts.