# Sagaphone

A web-based, procedural family-saga narrator with TTS — generates random
family trees and narrates their story out loud. Intended as a "fall asleep"
tool: set a timer, and it reads family sagas until time runs out.

See [SPEC.md](SPEC.md) for the full spec, architecture, and design areas.

## Status

Scaffolded, no features implemented yet. `backend/` and `frontend/` both run
and have a passing test suite; actual simulation/narration/TTS logic is next.

## Stack

- Frontend: SvelteKit (`frontend/`) — `bun`
- Backend: FastAPI (`backend/`) — `uv`
- TTS: Piper (server-side)
- Narration: Anthropic Haiku API (server-side)
- Reverse proxy: Caddy
- Deployment target: Zeus, `sagaphone.orbitstack.casa`

## Dev

```bash
# backend
cd backend && uv sync && uv run uvicorn app.main:app --reload

# frontend
cd frontend && bun install && bun run dev
```

Each subproject's own README has more detail (lint/test/typecheck commands).