docs: Add v3.7.0 architecture docs and wizard factory CLI

Added comprehensive documentation for:
- Wizard Factory completion and discovery
- Workflow Factory completion and pattern analysis
- CrewAI integration architecture
- XML enhancement summary and migration status
- Phase 1 completion status

Added wizard factory CLI and coach wizard examples.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <[email protected]>

Author: 2 people

coach_wizards/code_reviewer_README.md

@@ -0,0 +1,60 @@
+# Code Reviewer Wizard
+
+**Domain:** software
+**Type:** Wizard
+**Generated:** Pattern-Compose Methodology
+
+## Overview
+
+Auto-generated wizard using proven patterns from the Empathy Framework.
+
+## Patterns Used
+
+- **Empathy Level**: 0-4 empathy level configuration
+- **User Guidance**: Help text, examples, and prompts
+- **Code Analysis Input**: Standard code analysis input (code, file_path, language)
+- **Risk Assessment**: Level 4 Anticipatory risk analysis
+- **Prediction**: Level 4 Anticipatory prediction of future issues
+- **Config Validation**: Validate wizard configuration on initialization
+
+## Usage
+
+```python
+from coach_wizards.code_reviewer_wizard import CodeReviewerWizard
+
+wizard = CodeReviewerWizard()
+result = await wizard.process(user_input="...")
+```
+
+## API Endpoints
+
+```bash
+# Process with wizard
+POST /api/wizard/code_reviewer/process
+{
+  "input": "your input here",
+  "context": {}
+}
+```
+
+## Testing
+
+```bash
+# Run unit tests
+pytest tests/unit/wizards/test_code_reviewer_wizard.py
+
+# Run with coverage
+pytest tests/unit/wizards/test_code_reviewer_wizard.py --cov
+```
+
+## Next Steps
+
+1. Customize wizard logic as needed
+2. Add domain-specific validation
+3. Extend with additional features
+4. Update tests for custom logic
+
+---
+
+**Generated by:** Empathy Framework - Wizard Factory
+**Methodology:** Pattern-Compose

coach_wizards/code_reviewer_wizard.py

@@ -0,0 +1,180 @@
+"""Code reviewer Coach Wizard.
+
+Auto-generated by Empathy Framework Scaffolding
+Methodology: pattern-compose
+Domain: software
+Patterns: empathy_level, user_guidance, code_analysis_input, risk_assessment, prediction, config_validation
+Generated: 2026-01-05T20:15:45.111917
+"""
+
+import logging
+from typing import Any
+
+from fastapi import APIRouter, HTTPException
+from pydantic import BaseModel, Field
+
+from patterns import get_pattern_registry
+
+logger = logging.getLogger(__name__)
+
+router = APIRouter(prefix="/code_reviewer", tags=["code_reviewer"])
+
+# Pattern registry for recommendations
+registry = get_pattern_registry()
+
+
+# Request/Response Models
+class AnalysisRequest(BaseModel):
+    """Request to analyze code."""
+
+    code: str = Field(..., description="Code to analyze")
+    context: dict[str, Any] = Field(default_factory=dict, description="Additional context")
+    include_risk_assessment: bool = Field(default=True, description="Include risk analysis")
+
+
+class AnalysisResult(BaseModel):
+    """Analysis result."""
+
+    wizard_id: str
+    analysis: dict[str, Any]
+    risk_assessment: dict[str, Any] | None = None
+    predictions: list[dict[str, Any]] = Field(default_factory=list)
+
+
+# In-memory session storage (replace with Redis in production)
+sessions: dict[str, dict[str, Any]] = {}
+
+
+@router.post("/analyze", response_model=AnalysisResult)
+async def analyze_code(request: AnalysisRequest) -> AnalysisResult:
+    """Analyze code and provide recommendations.
+
+    Args:
+        request: Analysis request with code and context
+
+    Returns:
+        Analysis results with recommendations
+
+    Raises:
+        HTTPException: If analysis fails
+    """
+    try:
+        # Create session
+        wizard_id = f"code_reviewer_{len(sessions) + 1}"
+
+        # Perform code analysis
+        analysis = await _analyze_code(request.code, request.context)
+
+        result = AnalysisResult(
+            wizard_id=wizard_id,
+            analysis=analysis,
+        )
+
+        # Perform risk assessment
+        if request.include_risk_assessment:
+            risk_assessment = await _assess_risk(analysis)
+            result.risk_assessment = risk_assessment
+
+        # Generate predictions
+        predictions = await _generate_predictions(analysis)
+        result.predictions = predictions
+
+        # Store session
+        sessions[wizard_id] = {
+            "wizard_id": wizard_id,
+            "code": request.code,
+            "analysis": analysis,
+            "result": result.model_dump(),
+        }
+
+        logger.info(f"Analysis complete for {wizard_id}")
+        return result
+
+    except Exception as e:
+        logger.exception(f"Analysis failed: {e}")
+        raise HTTPException(status_code=500, detail=f"Analysis failed: {str(e)}")
+
+
+@router.get("/{wizard_id}/report", response_model=dict[str, Any])
+async def get_report(wizard_id: str) -> dict[str, Any]:
+    """Get analysis report.
+
+    Args:
+        wizard_id: Wizard session ID
+
+    Returns:
+        Complete analysis report
+
+    Raises:
+        HTTPException: If session not found
+    """
+    if wizard_id not in sessions:
+        raise HTTPException(status_code=404, detail="Session not found")
+
+    session = sessions[wizard_id]
+    return {
+        "wizard_id": wizard_id,
+        "analysis": session["analysis"],
+        "result": session["result"],
+    }
+
+
+# Helper functions
+async def _analyze_code(code: str, context: dict[str, Any]) -> dict[str, Any]:
+    """Analyze code (placeholder - implement actual analysis).
+
+    Args:
+        code: Code to analyze
+        context: Additional context
+
+    Returns:
+        Analysis results
+    """
+    # TODO: Implement actual code analysis
+    return {
+        "lines_of_code": len(code.split("\n")),
+        "complexity": "medium",
+        "issues_found": 0,
+        "context": context,
+    }
+
+
+async def _assess_risk(analysis: dict[str, Any]) -> dict[str, Any]:
+    """Assess risk based on analysis.
+
+    Args:
+        analysis: Code analysis results
+
+    Returns:
+        Risk assessment
+    """
+    # TODO: Implement actual risk assessment
+    return {
+        "alert_level": "LOW",
+        "risk_score": 0.2,
+        "by_risk_level": {
+            "critical": 0,
+            "high": 0,
+            "medium": 0,
+            "low": 0,
+        },
+    }
+
+
+async def _generate_predictions(analysis: dict[str, Any]) -> list[dict[str, Any]]:
+    """Generate predictions about future issues.
+
+    Args:
+        analysis: Code analysis results
+
+    Returns:
+        List of predictions
+    """
+    # TODO: Implement actual predictions
+    return [
+        {
+            "type": "performance",
+            "confidence": 0.8,
+            "description": "May experience performance issues with large datasets",
+        }
+    ]