@the-bearded-bear/claude-craft 5.0.0 → 5.1.0-next.6023ffb

package/Dev/i18n/de/Common/commands/recette.md

@@ -0,0 +1,244 @@
+---
+description: Automatisierte Akzeptanztests mit Claude in Chrome
+argument-hint: --scope=<story|epic|sprint|task> --id=<target-id> [--resume|--record-gif|--dry-run]
+---
+
+# QA Recette - Automatisierte Akzeptanztests
+
+Fuehrt automatisierte Akzeptanztests (Recette) auf Webanwendungen mit Claude in Chrome fuer Browser-Automatisierung durch. Dieses System implementiert die **Goldene Regel**: Ein behobener Fehler darf NIE wieder auftreten.
+
+## Argumente
+
+**$ARGUMENTS**
+
+- `--scope=<type>`: Testumfang (story, epic, sprint, task)
+- `--id=<target-id>`: Ziel-Identifikator (z.B. US-001, EPIC-01, Sprint-3)
+- `--resume=<session-id>`: Von einer vorherigen Sitzung fortsetzen
+- `--record-gif`: GIF der Ausfuehrung aufzeichnen
+- `--dry-run`: Plan generieren ohne Tests auszufuehren
+- `--base-url=<url>`: Basis-URL ueberschreiben
+
+## Hauptfunktionen
+
+| Funktion | Beschreibung |
+|----------|--------------|
+| **Umfassende Plaene** | Generiert umfassende Testplaene aus Akzeptanzkriterien |
+| **Browser-Automatisierung** | Verwendet Claude in Chrome fuer echte Browser-Tests |
+| **Sitzungswiederherstellung** | Checkpoint-basierte Fortsetzung fuer unterbrochene Sitzungen |
+| **Goldene Regel** | Automatische Regressionstestgenerierung fuer alle Fehler |
+| **Lebende Dokumentation** | Pflegt Testdokumentation mit Rueckverfolgbarkeit |
+| **Regressionserkennung** | Vergleicht Laeufe zur Erkennung von Regressionen |
+
+## Voraussetzungen
+
+1. **Claude in Chrome Extension**: Version 1.0.36 oder hoeher
+2. **Chrome Browser**: Geoeffnet mit aktiver Extension
+3. **Claude Code**: Gestartet mit `--chrome` Flag oder `/chrome` Befehl
+
+```bash
+# Claude Code mit Chrome-Unterstuetzung starten
+claude --chrome
+
+# Oder Chrome in bestehender Sitzung aktivieren
+/chrome
+```
+
+## Prozess
+
+### 1. Verifizierung
+
+Der Befehl prueft zuerst, ob Chrome MCP verfuegbar ist:
+
+```
+┌─────────────────────────────────────────┐
+│  1. check_chrome_mcp()                  │
+│     - MCP claude-in-chrome vorhanden?   │
+│     - Extension verbunden?              │
+│     - Website-Berechtigungen OK?        │
+└─────────────────────────────────────────┘
+```
+
+### 2. Testplan-Generierung
+
+Generiert einen umfassenden Testplan, der abdeckt:
+
+| Kategorie | Beschreibung |
+|-----------|--------------|
+| `acceptance_criteria_validation` | Tests fuer jedes AC |
+| `edge_cases` | Randbedingungen |
+| `error_scenarios` | Fehlerbehandlung |
+| `ui_ux_verification` | UI/UX-Konsistenz |
+| `performance_checks` | Ladezeiten |
+| `security_basics` | XSS, CSRF, Injection |
+
+### 3. Testausfuehrung
+
+Jeder Test wird ueber Chrome ausgefuehrt:
+
+```
+Test TC-001
+├── Schritt 1: navigate → /login
+├── Schritt 2: type → #email = "user@test.com"
+├── Schritt 3: click → button[type='submit']
+└── Assertions
+    ├── url_matches → ^.*/dashboard$
+    └── element_visible → .welcome-message
+```
+
+### 4. Fehler → Test → Regression
+
+Wenn ein Fehler erkannt wird:
+
+```
+1. Fehler waehrend Recette erkannt
+         │
+         ▼
+2. Klassifizierung (visual, interaction, validation, logic, security, API)
+         │
+         ▼
+3. Tests nach Typ generieren:
+   - Logic/Validation → Unit-Test
+   - API/Service → Funktionstest
+   - Benutzerfluss → Behat-Feature
+         │
+         ▼
+4. Zum Regressionsregister mit @regression Tag hinzufuegen
+         │
+         ▼
+5. Bug beheben (TDD-Workflow)
+         │
+         ▼
+6. Verifizieren: alle Regressionstests bestehen
+```
+
+## Schnellstart-Beispiele
+
+```bash
+# Eine bestimmte Story testen
+/qa:recette --scope=story --id=US-001
+
+# Alle Stories eines Sprints testen
+/qa:recette --scope=sprint --id=Sprint-3
+
+# Dry Run um Testplan zu sehen
+/qa:recette --scope=story --id=US-001 --dry-run
+
+# Unterbrochene Sitzung fortsetzen
+/qa:recette --scope=story --id=US-001 --resume=REC-20260130-143022
+
+# Ausfuehrung als GIF aufzeichnen
+/qa:recette --scope=story --id=US-001 --record-gif
+```
+
+## Sitzungswiederherstellung
+
+Sitzungen werden nach jedem Test gespeichert:
+
+```yaml
+# .recette/sessions/{session-id}/state.yaml
+session:
+  id: "REC-20260130-143022"
+  status: "paused"
+
+progress:
+  current_test_index: 5
+  tests:
+    total: 15
+    passed: 4
+    failed: 1
+    pending: 10
+
+recovery:
+  resumable: true
+  resume_from:
+    test_id: "TC-005"
+    step_index: 0
+```
+
+Zum Fortsetzen:
+
+```bash
+/qa:recette --scope=story --id=US-001 --resume=REC-20260130-143022
+```
+
+## Regressionsregister
+
+Alle erkannten Fehler werden verfolgt:
+
+```yaml
+# .recette/regression/registry.yaml
+entries:
+  - id: "REG-001"
+    error_id: "ERR-001"
+    source:
+      scope: "story"
+      target_id: "US-001"
+    generated_tests:
+      - type: "unit"
+        path: "tests/Unit/Auth/LoginErrorTest.php"
+      - type: "behat"
+        path: "features/auth/login_error.feature"
+    fix:
+      status: "verified"
+```
+
+## Ausgabestruktur
+
+```
+.recette/
+├── plans/              # Testplaene (YAML)
+│   └── story-US-001-plan.yaml
+├── sessions/           # Sitzungszustaende
+│   └── REC-20260130-143022/
+│       ├── state.yaml
+│       ├── screenshots/
+│       ├── checkpoints/
+│       └── logs/
+├── regression/         # Regressions-Suite
+│   ├── registry.yaml
+│   └── tests/
+│       ├── Unit/
+│       ├── Functional/
+│       └── Behat/
+├── metrics/            # Historische Daten
+│   └── history.jsonl
+└── reports/            # Generierte Berichte
+    └── REC-20260130-143022-report.md
+```
+
+## Verwandte Befehle
+
+| Befehl | Beschreibung |
+|--------|--------------|
+| `/qa:recette-status` | Sitzungsstatus anzeigen |
+| `/qa:recette-regression` | Regressionstests anzeigen |
+| `/qa:recette-report` | Bericht generieren |
+| `/qa:validate` | Story AC validieren |
+| `/qa:automate` | Automatisierte Tests erstellen |
+
+## Chrome-Faehigkeiten
+
+| Kategorie | Aktionen |
+|-----------|----------|
+| **Navigation** | navigate, back, forward, refresh |
+| **Interaktion** | click, type, fill_form, scroll, hover |
+| **Lesen** | DOM-Zustand, Element-Text, Attribute |
+| **Debugging** | Konsolen-Logs, Netzwerkanfragen, Fehler |
+| **Erfassung** | Screenshot, GIF-Aufnahme |
+
+## Fehlermeldungen
+
+| Fehler | Loesung |
+|--------|---------|
+| "MCP nicht erkannt" | `claude --chrome` oder `/chrome` ausfuehren |
+| "Extension nicht verbunden" | Chrome oeffnen, Extension pruefen |
+| "Berechtigung erforderlich" | Extension auf der Domain autorisieren |
+| "Version veraltet" | Chrome Extension auf v1.0.36+ aktualisieren |
+
+## Best Practices
+
+1. **Mit Dry-Run beginnen**: Testplan vor Ausfuehrung pruefen
+2. **Spezifische Scopes verwenden**: Stories einzeln testen fuer bessere Nachverfolgung
+3. **Regressionen pruefen**: `.recette/regression/` nach jedem Lauf konsultieren
+4. **GIF-Aufnahme aktivieren**: Zum Debuggen komplexer Fehler
+5. **Basis-URL pflegen**: Im Plan fuer konsistente Tests konfigurieren

package/Dev/i18n/en/Common/commands/ralph-sprint.md

@@ -108,6 +108,61 @@ The conductor stops when any condition is met:
 | Stop window | 06:00 | Time-based stop (for overnight) |
 | Critical escalation | - | Pauses on critical issues |
 
+## Parallel Processing
+
+When `--parallel N` is specified, the ASC processes stories in parallel waves:
+
+### How It Works
+
+1. **Dependency Graph**: The ASC builds a dependency graph from `blocked_by` relationships
+2. **Wave Processing**: Stories are processed in waves based on dependency satisfaction
+3. **Slot Management**: Up to N concurrent sessions, limited by CPU/memory thresholds
+4. **Sequential Merge**: Git branches are merged sequentially to avoid conflicts
+
+```
+Wave 1: [US-001, US-002, US-003]  ← Independent stories (no blockers)
+           ↓        ↓        ↓
+        Session  Session  Session
+           ↓        ↓        ↓
+        Complete Complete Complete
+
+Wave 2: [US-004, US-005]  ← Depend on US-001/US-002 (now done)
+           ↓        ↓
+        Session  Session
+           ↓        ↓
+        Complete Complete
+```
+
+### Dependency Handling
+
+Stories are only scheduled when their dependencies are satisfied:
+
+```yaml
+# sprint-status.yaml
+stories:
+  US-001:
+    status: ready-for-dev
+    blocked_by: []            # ✓ Can start immediately
+  US-002:
+    status: ready-for-dev
+    blocked_by: []            # ✓ Can start immediately
+  US-003:
+    status: ready-for-dev
+    blocked_by: [US-001]      # ✗ Waits for US-001 completion
+```
+
+### Resource Limits
+
+```yaml
+# ralph-autonomous.yml
+parallel:
+  enabled: true
+  max_concurrent: 3
+  resource_limits:
+    cpu_percent: 80      # Don't spawn if CPU > 80%
+    mem_percent: 80      # Don't spawn if memory > 80%
+```
+
 ## Quick Start Examples
 
 ```bash
@@ -117,6 +172,9 @@ The conductor stops when any condition is met:
 # Parallel processing with 3 sessions
 /common:ralph-sprint "Sprint 3" --parallel 3
 
+# Combined: parallel overnight
+/common:ralph-sprint "Sprint 3" --parallel 3 --overnight
+
 # Supervised mode (confirm each story)
 /common:ralph-sprint "Sprint 3" --supervised
 

package/Dev/i18n/en/Common/commands/recette.md

@@ -0,0 +1,244 @@
+---
+description: Automated acceptance testing with Claude in Chrome
+argument-hint: --scope=<story|epic|sprint|task> --id=<target-id> [--resume|--record-gif|--dry-run]
+---
+
+# QA Recette - Automated Acceptance Testing
+
+Execute automated acceptance tests (recette) on web applications using Claude in Chrome for browser automation. This system implements the **Golden Rule**: A fixed bug should NEVER reappear.
+
+## Arguments
+
+**$ARGUMENTS**
+
+- `--scope=<type>`: Scope of testing (story, epic, sprint, task)
+- `--id=<target-id>`: Target identifier (e.g., US-001, EPIC-01, Sprint-3)
+- `--resume=<session-id>`: Resume from a previous session
+- `--record-gif`: Record GIF of test execution
+- `--dry-run`: Generate plan without executing tests
+- `--base-url=<url>`: Override base URL for testing
+
+## Key Features
+
+| Feature | Description |
+|---------|-------------|
+| **Comprehensive Plans** | Generates exhaustive test plans from acceptance criteria |
+| **Browser Automation** | Uses Claude in Chrome for real browser testing |
+| **Session Recovery** | Checkpoint-based resume for interrupted sessions |
+| **Golden Rule** | Automatic regression test generation for all errors |
+| **Living Documentation** | Maintains test documentation with traceability |
+| **Regression Detection** | Compares runs to detect regressions |
+
+## Prerequisites
+
+1. **Claude in Chrome Extension**: Version 1.0.36 or higher
+2. **Chrome Browser**: Open with the extension active
+3. **Claude Code**: Started with `--chrome` flag or `/chrome` command
+
+```bash
+# Start Claude Code with Chrome support
+claude --chrome
+
+# Or enable Chrome in existing session
+/chrome
+```
+
+## Process
+
+### 1. Verification
+
+The command first verifies Chrome MCP is available:
+
+```
+┌─────────────────────────────────────────┐
+│  1. check_chrome_mcp()                  │
+│     - MCP claude-in-chrome present?     │
+│     - Extension connected?              │
+│     - Site permissions OK?              │
+└─────────────────────────────────────────┘
+```
+
+### 2. Test Plan Generation
+
+Generates comprehensive test plan covering:
+
+| Category | Description |
+|----------|-------------|
+| `acceptance_criteria_validation` | Tests for each AC |
+| `edge_cases` | Boundary conditions |
+| `error_scenarios` | Error handling |
+| `ui_ux_verification` | UI/UX consistency |
+| `performance_checks` | Load times |
+| `security_basics` | XSS, CSRF, injection |
+
+### 3. Test Execution
+
+Each test is executed via Chrome:
+
+```
+Test TC-001
+├── Step 1: navigate → /login
+├── Step 2: type → #email = "user@test.com"
+├── Step 3: click → button[type='submit']
+└── Assertions
+    ├── url_matches → ^.*/dashboard$
+    └── element_visible → .welcome-message
+```
+
+### 4. Error → Test → Regression
+
+When an error is detected:
+
+```
+1. Error detected during recette
+         │
+         ▼
+2. Classification (visual, interaction, validation, logic, security, API)
+         │
+         ▼
+3. Generate tests based on type:
+   - Logic/Validation → Unit test
+   - API/Service → Functional test
+   - User flow → Behat feature
+         │
+         ▼
+4. Add to regression registry with @regression tag
+         │
+         ▼
+5. Fix the bug (TDD workflow)
+         │
+         ▼
+6. Verify: all regression tests pass
+```
+
+## Quick Start Examples
+
+```bash
+# Test a specific story
+/qa:recette --scope=story --id=US-001
+
+# Test all stories in a sprint
+/qa:recette --scope=sprint --id=Sprint-3
+
+# Dry run to see the test plan
+/qa:recette --scope=story --id=US-001 --dry-run
+
+# Resume an interrupted session
+/qa:recette --scope=story --id=US-001 --resume=REC-20260130-143022
+
+# Record execution as GIF
+/qa:recette --scope=story --id=US-001 --record-gif
+```
+
+## Session Recovery
+
+Sessions are checkpointed after each test:
+
+```yaml
+# .recette/sessions/{session-id}/state.yaml
+session:
+  id: "REC-20260130-143022"
+  status: "paused"
+
+progress:
+  current_test_index: 5
+  tests:
+    total: 15
+    passed: 4
+    failed: 1
+    pending: 10
+
+recovery:
+  resumable: true
+  resume_from:
+    test_id: "TC-005"
+    step_index: 0
+```
+
+To resume:
+
+```bash
+/qa:recette --scope=story --id=US-001 --resume=REC-20260130-143022
+```
+
+## Regression Registry
+
+All detected errors are tracked:
+
+```yaml
+# .recette/regression/registry.yaml
+entries:
+  - id: "REG-001"
+    error_id: "ERR-001"
+    source:
+      scope: "story"
+      target_id: "US-001"
+    generated_tests:
+      - type: "unit"
+        path: "tests/Unit/Auth/LoginErrorTest.php"
+      - type: "behat"
+        path: "features/auth/login_error.feature"
+    fix:
+      status: "verified"
+```
+
+## Output Structure
+
+```
+.recette/
+├── plans/              # Test plans (YAML)
+│   └── story-US-001-plan.yaml
+├── sessions/           # Session states
+│   └── REC-20260130-143022/
+│       ├── state.yaml
+│       ├── screenshots/
+│       ├── checkpoints/
+│       └── logs/
+├── regression/         # Regression suite
+│   ├── registry.yaml
+│   └── tests/
+│       ├── Unit/
+│       ├── Functional/
+│       └── Behat/
+├── metrics/            # Historical data
+│   └── history.jsonl
+└── reports/            # Generated reports
+    └── REC-20260130-143022-report.md
+```
+
+## Related Commands
+
+| Command | Description |
+|---------|-------------|
+| `/qa:recette-status` | Show session status |
+| `/qa:recette-regression` | View regression tests |
+| `/qa:recette-report` | Generate report |
+| `/qa:validate` | Validate story AC |
+| `/qa:automate` | Create automated tests |
+
+## Chrome Capabilities
+
+| Category | Actions |
+|----------|---------|
+| **Navigation** | navigate, back, forward, refresh |
+| **Interaction** | click, type, fill_form, scroll, hover |
+| **Reading** | DOM state, element text, attributes |
+| **Debugging** | Console logs, network requests, errors |
+| **Capture** | Screenshot, record GIF |
+
+## Error Messages
+
+| Error | Solution |
+|-------|----------|
+| "MCP not detected" | Run `claude --chrome` or `/chrome` |
+| "Extension not connected" | Open Chrome, verify extension |
+| "Permission required" | Allow extension on the domain |
+| "Version outdated" | Update Chrome extension to v1.0.36+ |
+
+## Best Practices
+
+1. **Start with dry-run**: Verify the test plan before execution
+2. **Use specific scopes**: Test stories individually for better tracking
+3. **Review regressions**: Check `.recette/regression/` after each run
+4. **Enable GIF recording**: For debugging complex failures
+5. **Maintain base URL**: Configure in plan for consistent testing