documente/prompts/4-content-generation.md

# Prompt Sprint 4 - Content Generation

## 🎯 Sprint 4: Content Generation

**Iniziato**: 2026-04-06  
**Stato**: 🟡 In Progress  
**Assegnato**: @sprint-lead

---

## 📋 Obiettivo

Implementare la generazione di contenuti multi-formato da parte di NotebookLM. Supportare audio (podcast), video, slide, infografiche, quiz, flashcard, report, mappe mentali e tabelle dati.

---

## 🏗️ Architettura

### Pattern (stesso di Sprint 1-3)

```
API Layer (FastAPI Routes)
    ↓
Service Layer (ArtifactService)
    ↓
External Layer (notebooklm-py client)
```

### Endpoints da implementare (9 totali)

1. **POST /api/v1/notebooks/{id}/generate/audio** - Generare podcast
2. **POST /api/v1/notebooks/{id}/generate/video** - Generare video
3. **POST /api/v1/notebooks/{id}/generate/slide-deck** - Generare slide
4. **POST /api/v1/notebooks/{id}/generate/infographic** - Generare infografica
5. **POST /api/v1/notebooks/{id}/generate/quiz** - Generare quiz
6. **POST /api/v1/notebooks/{id}/generate/flashcards** - Generare flashcard
7. **POST /api/v1/notebooks/{id}/generate/report** - Generare report
8. **POST /api/v1/notebooks/{id}/generate/mind-map** - Generare mappa mentale
9. **POST /api/v1/notebooks/{id}/generate/data-table** - Generare tabella

### Endpoints gestione artifacts

10. **GET /api/v1/notebooks/{id}/artifacts** - Listare artifacts
11. **GET /api/v1/artifacts/{id}/status** - Controllare stato
12. **GET /api/v1/artifacts/{id}/download** - Scaricare artifact

---

## 📊 Task Breakdown Sprint 4

### Fase 1: Specifiche
- [ ] SPEC-007: Analisi requisiti Content Generation
- [ ] Definire parametri per ogni tipo di contenuto
- [ ] Definire stati artifact (pending, processing, completed, failed)

### Fase 2: API Design
- [ ] API-006: Modelli Pydantic per ogni tipo di generazione
- [ ] Documentazione endpoints

### Fase 3: Implementazione
- [ ] DEV-015: ArtifactService
- [ ] DEV-016: Tutti gli endpoint POST /generate/*
- [ ] DEV-017: GET /artifacts
- [ ] DEV-018: GET /artifacts/{id}/status

### Fase 4: Testing
- [ ] TEST-008: Unit tests ArtifactService
- [ ] TEST-009: Integration tests generation API

---

## 🔧 Implementazione

### ArtifactService Methods

```python
class ArtifactService:
    async def generate_audio(notebook_id, instructions, format, length, language)
    async def generate_video(notebook_id, instructions, style, language)
    async def generate_slide_deck(notebook_id, format, length)
    async def generate_infographic(notebook_id, orientation, detail, style)
    async def generate_quiz(notebook_id, difficulty, quantity)
    async def generate_flashcards(notebook_id, difficulty, quantity)
    async def generate_report(notebook_id, format)
    async def generate_mind_map(notebook_id)
    async def generate_data_table(notebook_id, description)
    
    async def list_artifacts(notebook_id)
    async def get_status(artifact_id)
    async def download_artifact(artifact_id)
```

### Modelli

```python
# Request models
class AudioGenerationRequest(BaseModel):
    instructions: str
    format: str  # deep-dive, brief, critique, debate
    length: str  # short, default, long
    language: str

class VideoGenerationRequest(BaseModel):
    instructions: str
    style: str  # whiteboard, classic, anime, etc.
    language: str

class QuizGenerationRequest(BaseModel):
    difficulty: str  # easy, medium, hard
    quantity: str  # fewer, standard, more

# Response models
class Artifact(BaseModel):
    id: UUID
    notebook_id: UUID
    type: str  # audio, video, quiz, etc.
    title: str
    status: str  # pending, processing, completed, failed
    created_at: datetime
    completed_at: datetime | None
    download_url: str | None
```

---

## 🎨 Content Types

### Audio (Podcast)
- Formats: deep-dive, brief, critique, debate
- Length: short, default, long
- Languages: en, it, es, fr, de

### Video
- Styles: whiteboard, classic, anime, kawaii, watercolor, etc.
- Languages: multi-language support

### Slide Deck
- Formats: detailed, presenter
- Length: default, short

### Infographic
- Orientation: landscape, portrait, square
- Detail: concise, standard, detailed
- Styles: professional, editorial, scientific, etc.

### Quiz / Flashcards
- Difficulty: easy, medium, hard
- Quantity: fewer, standard, more

### Mind Map
- Instant generation (no async)

### Data Table
- Custom description for data extraction

---

## 🚀 Prossimi Passi

1. @sprint-lead: Attivare @api-designer per API-006
2. @api-designer: Definire tutti i modelli generation
3. @tdd-developer: Iniziare implementazione ArtifactService

---

**Dipende da**: Sprint 3 (Chat) ✅  
**Blocca**: Sprint 5 (Webhooks) 🔴

**Nota**: Questo è lo sprint più complesso con 12 endpoint da implementare!