AI Novel Generator

中文文档 | English

AI Novel Generator is an AI-assisted novel creation project. The repository currently contains:

• a FastAPI backend for configuration, novel, volume, upload, and AI workflow APIs
• a Next.js frontend that is being rebuilt
• a Windows desktop launcher that can start the backend and frontend together

Current Status

This project is still under an ongoing refactor and is not yet considered feature-complete.

• interfaces, folder structure, and workflows may continue to change
• some features from earlier iterations may be incomplete or temporarily unavailable
• if you need the previous version, please check the main branch

Requirements

• Python 3
• MongoDB
• Node.js and npm for the frontend

Configuration

Application configuration is loaded from application/config/config.yaml.

On first run, the project will ensure the following files exist:

• application/config/config_default.yaml
• application/config/config.yaml

Before running AI-related features, update the LLM provider settings in application/config/config.yaml, especially:

• api_key
• base_url where applicable
• default_provider and workflow provider mapping
• use_system_proxy if you explicitly want the SDK to inherit your Windows/system proxy. The default is false.

You should also make sure MongoDB is running locally if you use the default database settings.

Configuration can be edited through the frontend settings interface or by directly modifying the YAML files mentioned above.

Install

Install Python dependencies:

pip install -r requirements.txt

Install frontend dependencies:

cd frontend
npm install

Run

Option 1: Windows launcher

The quickest Windows entry point is:

start.bat

This starts the desktop launcher, which can then start:

• backend: http://127.0.0.1:8000
• API docs: http://127.0.0.1:8000/docs
• frontend: http://localhost:3000

Option 2: Run backend manually

python main.py

Enable backend debug logging and raw AI response output:

python main.py --debug

When you use the Windows launcher, you can enable the backend checkbox labeled 后端调试日志 before starting the backend.

Option 3: Run frontend manually

cd frontend
npm run dev

Tests

Example test commands:

python -m tests.test_volumes
python -m tests.test_llm openai

Some tests require local services or valid model credentials.

Notes

This README intentionally documents the current refactor state only. If the current branch does not match the workflow you expect, review the main branch for the earlier implementation.