Reading Assistant

An AI-powered tool that converts books (PDF, EPUB, or plain text) into structured summaries, and interactive chat.

The goal is to help busy readers extract core ideas of a book quickly, without needing to read it cover to cover.

---

Features

• Multi-format input (PDF, EPUB, plain text)
• Structured summary (core thesis, key insights, actionable takeaways, who it’s for)
• Chapter-by-chapter summaries
• Chat with the book content
• Export summaries to PDF
• Caching to avoid re-processing identical content

---

Tech Stack

Backend

• Language: Python 3.12
• Framework: FastAPI
• ORM: SQLAlchemy
• Database: SQLite (default), configurable to PostgreSQL
• Vector Store: ChromaDB (for RAG)

Frontend

• Framework: Next.js 16 (App Router)
• UI: React 19 + TypeScript + Tailwind CSS
• Auth: Better Auth

AI & Infra

• LLM: OpenRouter API (Gemini 3 Flash)
• Embeddings: OpenAI embeddings (for RAG)
• RAG: LlamaIndex

---

Installation

Prerequisites

• Python 3.12+
• Node.js 24+
• uv Python package manager

---

Backend setup

python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

uv pip install -r requirements.txt

cp .env.example .env
# Edit .env with your API keys and settings

---

Frontend setup

cd client
npm install

cp .env.local.example .env.local
# Edit environment variables

Auth database (required)

Better Auth requires a Postgres database:

1. Create a Postgres instance (e.g. via Neon)
2. Set DATABASE_URL in client/.env.local
3. Run migrations:

npx drizzle-kit push

---

Running locally

# Backend
uvicorn api:app --reload --host 0.0.0.0 --port 8000

# Frontend
cd client
npm run dev

Open: http://localhost:3000

---

API Overview

• POST /api/summarize – Upload a book or text and generate a summary
• GET /api/summary/{id} – Retrieve a summary
• GET /api/summary/{id}/pdf – Download summary as PDF
• POST /api/summary/{id}/save – Save summary to user library
• GET /api/library – List saved summaries
• POST /api/chat – Chat with a summarized book (RAG)

---

Environment Variables

Backend (.env)

OPENROUTER_API_KEY=your_openrouter_key
OPENAI_API_KEY=your_openai_key
LLM_MODEL=google/gemini-3-flash-preview
DATABASE_URL=sqlite:///data/app.db
BETTER_AUTH_URL=http://localhost:3000
LOG_LEVEL=INFO

Frontend (client/.env.local)

NEXT_PUBLIC_API_URL=http://localhost:8000

---

Project Structure

├── api.py                 # FastAPI entrypoint
├── core/                  # Core business logic
│   ├── config.py
│   ├── extractor.py       # PDF / EPUB text extraction
│   ├── summarizer.py      # LLM summarization logic
│   ├── auth.py            # Auth helpers
│   └── pdf_export.py      # PDF generation
├── db/
│   ├── models/            # SQLAlchemy models
│   └── repositories/      # Data access layer
├── routers/               # API route definitions
├── rag/                   # RAG pipeline
├── client/                # Next.js frontend
│   ├── app/
│   ├── components/
│   └── lib/
├── requirements.txt
└── data/                  # SQLite DB + Chroma vectors

---

Contributing

Contributions are welcome! Please open an issue or submit a pull request.

---

License

MIT