- Python 3.10+ recommended
- Ollama installed and running
- Git installed (optional but recommended)
- Start Ollama (usually it auto-starts a server at http://localhost:11434)
- Pull the model: ollama pull gemma3:7b
From the project root:
python -m venv .venv
# mac/linux
source .venv/bin/activate
# windows (powershell)
# .venv\Scripts\Activate.ps1
pip install -r requirements.txtgit init
git add .
git commit -m "init"
git config user.name "Your Name"
git config user.email "you@example.com"python run.pyOpen: http://127.0.0.1:5000
- Chapters live in ./manuscript as Markdown files (e.g., ch01.md).
- AI metadata lives in ./meta/chapters/.json (auto-created).
- Index files: ./meta/chapter_index.json and ./meta/character_index.json (auto-created).
- Autosave runs in the browser every N seconds.
- When you stop typing for
inactivity.seconds_before_ai, the backend runs:- Ollama analysis (summary/characters/beats/act structure)
- Git add/commit with cooldown (
git.min_seconds_between_commits)