moai-adk 0.35.1__py3-none-any.whl

moai_adk/core/user_behavior_analytics.py

@@ -0,0 +1,851 @@
+"""
+User Behavior Analytics System
+
+Analyzes user interactions, patterns, and preferences to provide insights
+for system optimization and user experience improvement.
+
+Key Features:
+- User interaction tracking and pattern analysis
+- Command usage frequency and preference analysis
+- Session behavior analysis
+- User preference learning and adaptation
+- Collaboration pattern analysis
+- Productivity metrics and insights
+- User experience optimization recommendations
+"""
+
+import json
+import logging
+import statistics
+import uuid
+from collections import Counter, defaultdict, deque
+from dataclasses import asdict, dataclass, field
+from datetime import datetime, timedelta
+from enum import Enum
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Set
+
+# Set up logging
+logger = logging.getLogger(__name__)
+
+
+class UserActionType(Enum):
+    """Types of user actions tracked"""
+
+    COMMAND_EXECUTION = "command_execution"
+    TOOL_USAGE = "tool_usage"
+    FILE_OPERATION = "file_operation"
+    ERROR_OCCURRED = "error_occurred"
+    HELP_REQUESTED = "help_requested"
+    SESSION_START = "session_start"
+    SESSION_END = "session_end"
+    TASK_COMPLETED = "task_completed"
+    PHASE_TRANSITION = "phase_transition"
+
+
+class SessionState(Enum):
+    """User session states"""
+
+    ACTIVE = "active"
+    IDLE = "idle"
+    FOCUSED = "focused"
+    STRUGGLING = "struggling"
+    PRODUCTIVE = "productive"
+    BREAK = "break"
+
+
+@dataclass
+class UserAction:
+    """Single user action event"""
+
+    timestamp: datetime
+    action_type: UserActionType
+    user_id: str
+    session_id: str
+    action_data: Dict[str, Any]
+    context: Dict[str, Any] = field(default_factory=dict)
+    duration_ms: Optional[float] = None
+    success: bool = True
+    tags: Set[str] = field(default_factory=set)
+
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary for serialization"""
+        return {
+            "timestamp": self.timestamp.isoformat(),
+            "action_type": self.action_type.value,
+            "user_id": self.user_id,
+            "session_id": self.session_id,
+            "action_data": self.action_data,
+            "context": self.context,
+            "duration_ms": self.duration_ms,
+            "success": self.success,
+            "tags": list(self.tags),
+        }
+
+
+@dataclass
+class UserSession:
+    """User session information"""
+
+    session_id: str
+    user_id: str
+    start_time: datetime
+    end_time: Optional[datetime] = None
+    actions: List[UserAction] = field(default_factory=list)
+    state: SessionState = SessionState.ACTIVE
+    productivity_score: float = 0.0
+    total_commands: int = 0
+    total_errors: int = 0
+    total_duration_ms: float = 0.0
+    working_directory: str = ""
+    git_branch: str = ""
+    modified_files: Set[str] = field(default_factory=set)
+    tools_used: Set[str] = field(default_factory=set)
+
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary for serialization"""
+        return {
+            "session_id": self.session_id,
+            "user_id": self.user_id,
+            "start_time": self.start_time.isoformat(),
+            "end_time": self.end_time.isoformat() if self.end_time else None,
+            "state": self.state.value,
+            "productivity_score": self.productivity_score,
+            "total_commands": self.total_commands,
+            "total_errors": self.total_errors,
+            "total_duration_ms": self.total_duration_ms,
+            "working_directory": self.working_directory,
+            "git_branch": self.git_branch,
+            "modified_files": list(self.modified_files),
+            "tools_used": list(self.tools_used),
+        }
+
+
+@dataclass
+class UserPreferences:
+    """Learned user preferences and patterns"""
+
+    user_id: str
+    preferred_commands: Dict[str, int] = field(default_factory=dict)
+    preferred_tools: Dict[str, int] = field(default_factory=dict)
+    working_hours: Dict[str, int] = field(default_factory=dict)
+    peak_productivity_times: List[str] = field(default_factory=list)
+    common_workflows: List[List[str]] = field(default_factory=list)
+    error_patterns: List[str] = field(default_factory=list)
+    success_patterns: List[str] = field(default_factory=list)
+    collaboration_patterns: Dict[str, int] = field(default_factory=dict)
+    last_updated: datetime = field(default_factory=datetime.now)
+
+
+class UserBehaviorAnalytics:
+    """Main user behavior analytics system"""
+
+    def __init__(self, storage_path: Optional[Path] = None):
+        self.storage_path = storage_path or Path.cwd() / ".moai" / "analytics" / "user_behavior"
+        self.storage_path.mkdir(parents=True, exist_ok=True)
+
+        # Data storage
+        self.user_sessions: Dict[str, UserSession] = {}
+        self.active_sessions: Dict[str, UserSession] = {}
+        self.user_preferences: Dict[str, UserPreferences] = {}
+        self.action_history: deque = deque(maxlen=10000)
+
+        # Analysis caches
+        self._pattern_cache: Dict[str, Any] = {}
+        self._insight_cache: Dict[str, Any] = {}
+        self._last_cache_update = datetime.now()
+
+        # Load existing data
+        self._load_data()
+
+    def track_action(
+        self,
+        action_type: UserActionType,
+        user_id: str,
+        session_id: str,
+        action_data: Dict[str, Any],
+        context: Optional[Dict[str, Any]] = None,
+        duration_ms: Optional[float] = None,
+        success: bool = True,
+    ) -> None:
+        """Track a user action"""
+        action = UserAction(
+            timestamp=datetime.now(),
+            action_type=action_type,
+            user_id=user_id,
+            session_id=session_id,
+            action_data=action_data,
+            context=context or {},
+            duration_ms=duration_ms,
+            success=success,
+            tags=self._extract_action_tags(action_type, action_data),
+        )
+
+        # Store action
+        self.action_history.append(action)
+
+        # Update session if active
+        if session_id in self.active_sessions:
+            session = self.active_sessions[session_id]
+            session.actions.append(action)
+
+            # Update session metrics
+            if action_type == UserActionType.COMMAND_EXECUTION:
+                session.total_commands += 1
+                if not success:
+                    session.total_errors += 1
+
+                # Track tools used
+                if "tool" in action_data:
+                    session.tools_used.add(action_data["tool"])
+
+                # Track file modifications
+                if "files" in action_data:
+                    session.modified_files.update(action_data["files"])
+
+            if duration_ms:
+                session.total_duration_ms += duration_ms
+
+            # Update session state
+            session.state = self._analyze_session_state(session)
+
+        # Update user preferences
+        self._update_user_preferences(user_id, action)
+
+        # Clear caches periodically
+        if (datetime.now() - self._last_cache_update).seconds > 300:
+            self._pattern_cache.clear()
+            self._insight_cache.clear()
+            self._last_cache_update = datetime.now()
+
+    def start_session(self, user_id: str, working_directory: str = "", git_branch: str = "") -> str:
+        """Start tracking a new user session"""
+        session_id = str(uuid.uuid4())
+
+        session = UserSession(
+            session_id=session_id,
+            user_id=user_id,
+            start_time=datetime.now(),
+            working_directory=working_directory,
+            git_branch=git_branch,
+            state=SessionState.ACTIVE,
+        )
+
+        self.active_sessions[session_id] = session
+        self.user_sessions[session_id] = session
+
+        # Track session start action
+        self.track_action(
+            UserActionType.SESSION_START,
+            user_id,
+            session_id,
+            {"working_directory": working_directory, "git_branch": git_branch},
+        )
+
+        return session_id
+
+    def end_session(self, session_id: str) -> Optional[UserSession]:
+        """End a user session"""
+        if session_id not in self.active_sessions:
+            return None
+
+        session = self.active_sessions[session_id]
+        session.end_time = datetime.now()
+
+        # Calculate final productivity score
+        session.productivity_score = self._calculate_productivity_score(session)
+
+        # Move from active to completed sessions
+        del self.active_sessions[session_id]
+
+        # Track session end action
+        self.track_action(
+            UserActionType.SESSION_END,
+            session.user_id,
+            session_id,
+            {
+                "duration_ms": session.total_duration_ms,
+                "commands": session.total_commands,
+                "errors": session.total_errors,
+                "productivity_score": session.productivity_score,
+            },
+        )
+
+        # Save data periodically
+        self._save_data()
+
+        return session
+
+    def get_user_patterns(self, user_id: str, days: int = 30) -> Dict[str, Any]:
+        """Get user behavior patterns and insights"""
+        cache_key = f"patterns_{user_id}_{days}"
+
+        if cache_key in self._pattern_cache:
+            return self._pattern_cache[cache_key]
+
+        cutoff_date = datetime.now() - timedelta(days=days)
+
+        # Get user sessions in the time range
+        user_sessions = [
+            session
+            for session in self.user_sessions.values()
+            if session.user_id == user_id and session.start_time >= cutoff_date
+        ]
+
+        # Analyze patterns
+        patterns = {
+            "session_count": len(user_sessions),
+            "avg_session_duration": 0.0,
+            "avg_productivity_score": 0.0,
+            "most_used_commands": {},
+            "most_used_tools": {},
+            "peak_productivity_hours": [],
+            "common_workflows": [],
+            "error_patterns": [],
+            "working_preferences": {},
+        }
+
+        if user_sessions:
+            # Session metrics
+            durations = [s.total_duration_ms for s in user_sessions if s.total_duration_ms > 0]
+            productivity_scores = [s.productivity_score for s in user_sessions if s.productivity_score > 0]
+
+            if durations:
+                patterns["avg_session_duration"] = statistics.mean(durations)
+            if productivity_scores:
+                patterns["avg_productivity_score"] = statistics.mean(productivity_scores)
+
+            # Command usage
+            all_commands = []
+            all_tools = []
+
+            for session in user_sessions:
+                for action in session.actions:
+                    if action.action_type == UserActionType.COMMAND_EXECUTION:
+                        if "command" in action.action_data:
+                            all_commands.append(action.action_data["command"])
+                        if "tool" in action.action_data:
+                            all_tools.append(action.action_data["tool"])
+
+            patterns["most_used_commands"] = dict(Counter(all_commands).most_common(10))
+            patterns["most_used_tools"] = dict(Counter(all_tools).most_common(10))
+
+            # Peak productivity hours
+            hour_productivity = defaultdict(list)
+
+            for session in user_sessions:
+                hour_productivity[session.start_time.hour].append(session.productivity_score)
+
+            avg_hourly_productivity = {
+                hour: statistics.mean(scores) if scores else 0 for hour, scores in hour_productivity.items()
+            }
+
+            # Top 3 most productive hours
+            top_hours = sorted(avg_hourly_productivity.items(), key=lambda x: x[1], reverse=True)[:3]
+            patterns["peak_productivity_hours"] = [f"{hour:02d}:00" for hour, _ in top_hours]
+
+        # Cache results
+        self._pattern_cache[cache_key] = patterns
+
+        return patterns
+
+    def get_user_insights(self, user_id: str, days: int = 7) -> Dict[str, Any]:
+        """Get actionable insights and recommendations for a user"""
+        cache_key = f"insights_{user_id}_{days}"
+
+        if cache_key in self._insight_cache:
+            return self._insight_cache[cache_key]
+
+        patterns = self.get_user_patterns(user_id, days)
+
+        insights: Dict[str, Any] = {
+            "productivity_insights": [],
+            "efficiency_recommendations": [],
+            "collaboration_insights": [],
+            "learning_opportunities": [],
+            "tool_recommendations": [],
+            "workflow_optimizations": [],
+        }
+
+        # Productivity insights
+        if patterns["avg_productivity_score"] > 80:
+            insights["productivity_insights"].append(
+                "Excellent productivity score! User consistently performs at high level."
+            )
+        elif patterns["avg_productivity_score"] < 50:
+            insights["productivity_insights"].append(
+                "Productivity score could be improved. Consider workflow optimization."
+            )
+
+        # Session duration insights
+        if patterns["avg_session_duration"] > 1800000:  # > 30 minutes
+            insights["productivity_insights"].append(
+                "Long session durations detected. Consider taking regular breaks for sustained productivity."
+            )
+
+        # Error pattern analysis
+        recent_actions = [
+            action
+            for action in self.action_history
+            if action.user_id == user_id and action.timestamp >= datetime.now() - timedelta(days=days)
+        ]
+
+        error_actions = [
+            action
+            for action in recent_actions
+            if action.action_type == UserActionType.ERROR_OCCURRED or not action.success
+        ]
+
+        if len(error_actions) > len(recent_actions) * 0.1:  # > 10% error rate
+            insights["efficiency_recommendations"].append(
+                "High error rate detected. Consider additional training or tool familiarization."
+            )
+
+        # Tool usage insights
+        if patterns["most_used_tools"]:
+            top_tool = max(patterns["most_used_tools"].items(), key=lambda x: x[1])
+            insights["tool_recommendations"].append(f"Most frequently used tool: {top_tool[0]} ({top_tool[1]} uses)")
+
+        # Command pattern insights
+        if patterns["most_used_commands"]:
+            commands = list(patterns["most_used_commands"].keys())
+            if "/moai:" in " ".join(commands[:3]):  # Top 3 commands
+                insights["workflow_optimizations"].append(
+                    "Heavy MoAI command usage detected. Consider learning keyboard shortcuts for faster workflow."
+                )
+
+        # Peak productivity insights
+        if patterns["peak_productivity_hours"]:
+            peak_hours = ", ".join(patterns["peak_productivity_hours"])
+            insights["efficiency_recommendations"].append(
+                f"Peak productivity hours: {peak_hours}. Schedule important work during these times."
+            )
+
+        # Learning opportunities
+        if patterns["session_count"] < 5:
+            insights["learning_opportunities"].append(
+                "Consider more frequent sessions to build momentum and consistency."
+            )
+
+        # Cache results
+        self._insight_cache[cache_key] = insights
+
+        return insights
+
+    def get_team_analytics(self, days: int = 30) -> Dict[str, Any]:
+        """Get team-wide analytics and collaboration patterns"""
+        cutoff_date = datetime.now() - timedelta(days=days)
+
+        # Get all sessions in time range
+        recent_sessions = [session for session in self.user_sessions.values() if session.start_time >= cutoff_date]
+
+        # Aggregate metrics
+        team_metrics = {
+            "total_sessions": len(recent_sessions),
+            "unique_users": len(set(s.user_id for s in recent_sessions)),
+            "avg_session_duration": 0.0,
+            "avg_productivity_score": 0.0,
+            "most_active_users": {},
+            "most_used_tools": {},
+            "collaboration_patterns": {},
+            "productivity_distribution": {},
+            "peak_hours": {},
+        }
+
+        if recent_sessions:
+            # Calculate averages
+            durations = [s.total_duration_ms for s in recent_sessions if s.total_duration_ms > 0]
+            scores = [s.productivity_score for s in recent_sessions if s.productivity_score > 0]
+
+            if durations:
+                team_metrics["avg_session_duration"] = statistics.mean(durations)
+            if scores:
+                team_metrics["avg_productivity_score"] = statistics.mean(scores)
+
+            # Most active users
+            user_session_counts = Counter(s.user_id for s in recent_sessions)
+            team_metrics["most_active_users"] = dict(user_session_counts.most_common(10))
+
+            # Most used tools across team
+            all_tools: List[str] = []
+            for session in recent_sessions:
+                all_tools.extend(session.tools_used)
+
+            team_metrics["most_used_tools"] = dict(Counter(all_tools).most_common(10))
+
+            # Productivity distribution
+            productivity_ranges = {
+                "0-20": 0,
+                "21-40": 0,
+                "41-60": 0,
+                "61-80": 0,
+                "81-100": 0,
+            }
+
+            for score in scores:
+                if score <= 20:
+                    productivity_ranges["0-20"] += 1
+                elif score <= 40:
+                    productivity_ranges["21-40"] += 1
+                elif score <= 60:
+                    productivity_ranges["41-60"] += 1
+                elif score <= 80:
+                    productivity_ranges["61-80"] += 1
+                else:
+                    productivity_ranges["81-100"] += 1
+
+            team_metrics["productivity_distribution"] = productivity_ranges
+
+        return team_metrics
+
+    def _extract_action_tags(self, action_type: UserActionType, action_data: Dict[str, Any]) -> Set[str]:
+        """Extract relevant tags from action data"""
+        tags = {action_type.value}
+
+        # Extract tool tags
+        if "tool" in action_data:
+            tags.add(f"tool:{action_data['tool']}")
+
+        # Extract command tags
+        if "command" in action_data:
+            command = action_data["command"]
+            if "/moai:" in command:
+                tags.add("moai_command")
+            if "git" in command.lower():
+                tags.add("git_operation")
+
+        # Extract phase tags
+        if "phase" in action_data:
+            tags.add(f"phase:{action_data['phase']}")
+
+        # Extract success/failure tags
+        if "success" in action_data:
+            tags.add("success" if action_data["success"] else "failure")
+
+        return tags
+
+    def _update_user_preferences(self, user_id: str, action: UserAction) -> None:
+        """Update learned user preferences based on actions"""
+        if user_id not in self.user_preferences:
+            self.user_preferences[user_id] = UserPreferences(user_id=user_id)
+
+        prefs = self.user_preferences[user_id]
+
+        # Update command preferences
+        if action.action_type == UserActionType.COMMAND_EXECUTION:
+            command = action.action_data.get("command", "")
+            if command:
+                prefs.preferred_commands[command] = prefs.preferred_commands.get(command, 0) + 1
+
+        # Update tool preferences
+        if "tool" in action.action_data:
+            tool = action.action_data["tool"]
+            prefs.preferred_tools[tool] = prefs.preferred_tools.get(tool, 0) + 1
+
+        # Update working hours
+        hour = action.timestamp.hour
+        prefs.working_hours[str(hour)] = prefs.working_hours.get(str(hour), 0) + 1
+
+        # Update patterns based on success/failure
+        if action.success:
+            pattern_key = f"{action.action_type.value}_{action.action_data}"
+            if pattern_key not in prefs.success_patterns:
+                prefs.success_patterns.append(pattern_key)
+        else:
+            pattern_key = f"{action.action_type.value}_{action.action_data}"
+            if pattern_key not in prefs.error_patterns:
+                prefs.error_patterns.append(pattern_key)
+
+        prefs.last_updated = datetime.now()
+
+    def _analyze_session_state(self, session: UserSession) -> SessionState:
+        """Analyze and determine current session state"""
+        if not session.actions:
+            return SessionState.ACTIVE
+
+        recent_actions = [
+            action
+            for action in session.actions[-10:]  # Last 10 actions
+            if (datetime.now() - action.timestamp).seconds < 300  # Last 5 minutes
+        ]
+
+        if not recent_actions:
+            return SessionState.IDLE
+
+        # Calculate metrics
+        error_rate = sum(1 for action in recent_actions if not action.success) / len(recent_actions)
+
+        durations_with_ms = [a.duration_ms for a in recent_actions if a.duration_ms is not None]
+        avg_duration = statistics.mean(durations_with_ms) if durations_with_ms else 0
+
+        # Determine state
+        if error_rate > 0.5:
+            return SessionState.STRUGGLING
+        elif avg_duration > 10000 and error_rate < 0.1:
+            return SessionState.PRODUCTIVE
+        elif error_rate < 0.1 and len(recent_actions) > 5:
+            return SessionState.FOCUSED
+        else:
+            return SessionState.ACTIVE
+
+    def _calculate_productivity_score(self, session: UserSession) -> float:
+        """Calculate productivity score for a session (0-100)"""
+        if not session.actions:
+            return 0.0
+
+        # Base score from success rate
+        successful_actions = sum(1 for action in session.actions if action.success)
+        success_rate = successful_actions / len(session.actions)
+        base_score = success_rate * 50
+
+        # Duration factor (optimal session duration)
+        duration_hours = session.total_duration_ms / (1000 * 60 * 60)
+        duration_factor = 0
+
+        if 0.5 <= duration_hours <= 4:  # Optimal 30 min to 4 hours
+            duration_factor = 25
+        elif duration_hours > 0.25:  # At least 15 minutes
+            duration_factor = 15
+
+        # Tool diversity factor
+        tool_diversity = min(len(session.tools_used) / 10, 1) * 15
+
+        # File modifications factor (if it's a coding session)
+        file_factor: float = 0.0
+        if session.modified_files:
+            file_factor = min(len(session.modified_files) / 5, 1) * 10
+
+        total_score = base_score + duration_factor + tool_diversity + file_factor
+        return min(total_score, 100.0)
+
+    def _load_data(self) -> None:
+        """Load stored analytics data"""
+        try:
+            # Load user preferences
+            prefs_file = self.storage_path / "user_preferences.json"
+            if prefs_file.exists():
+                with open(prefs_file, "r") as f:
+                    prefs_data = json.load(f)
+
+                for user_id, prefs_dict in prefs_data.items():
+                    prefs = UserPreferences(user_id=user_id)
+                    prefs.preferred_commands = prefs_dict.get("preferred_commands", {})
+                    prefs.preferred_tools = prefs_dict.get("preferred_tools", {})
+                    prefs.working_hours = prefs_dict.get("working_hours", {})
+                    prefs.peak_productivity_times = prefs_dict.get("peak_productivity_times", [])
+                    prefs.common_workflows = prefs_dict.get("common_workflows", [])
+                    prefs.error_patterns = prefs_dict.get("error_patterns", [])
+                    prefs.success_patterns = prefs_dict.get("success_patterns", [])
+                    prefs.collaboration_patterns = prefs_dict.get("collaboration_patterns", {})
+
+                    if prefs_dict.get("last_updated"):
+                        prefs.last_updated = datetime.fromisoformat(prefs_dict["last_updated"])
+
+                    self.user_preferences[user_id] = prefs
+
+            logger.info(f"Loaded preferences for {len(self.user_preferences)} users")
+
+        except Exception as e:
+            logger.error(f"Error loading analytics data: {e}")
+
+    def _save_data(self) -> None:
+        """Save analytics data to storage"""
+        try:
+            # Save user preferences
+            prefs_data = {}
+            for user_id, prefs in self.user_preferences.items():
+                prefs_data[user_id] = {
+                    "preferred_commands": prefs.preferred_commands,
+                    "preferred_tools": prefs.preferred_tools,
+                    "working_hours": prefs.working_hours,
+                    "peak_productivity_times": prefs.peak_productivity_times,
+                    "common_workflows": prefs.common_workflows,
+                    "error_patterns": prefs.error_patterns,
+                    "success_patterns": prefs.success_patterns,
+                    "collaboration_patterns": prefs.collaboration_patterns,
+                    "last_updated": prefs.last_updated.isoformat(),
+                }
+
+            prefs_file = self.storage_path / "user_preferences.json"
+            with open(prefs_file, "w") as f:
+                json.dump(prefs_data, f, indent=2)
+
+            logger.info(f"Saved preferences for {len(self.user_preferences)} users")
+
+        except Exception as e:
+            logger.error(f"Error saving analytics data: {e}")
+
+    def export_data(self, output_path: Path, user_id: Optional[str] = None) -> bool:
+        """Export analytics data to file"""
+        try:
+            export_data = {
+                "export_timestamp": datetime.now().isoformat(),
+                "user_id": user_id or "all_users",
+                "data": {},
+            }
+
+            if user_id:
+                # Export single user data
+                patterns = self.get_user_patterns(user_id)
+                insights = self.get_user_insights(user_id)
+                prefs = self.user_preferences.get(user_id)
+
+                export_data["data"] = {
+                    "patterns": patterns,
+                    "insights": insights,
+                    "preferences": asdict(prefs) if prefs else None,
+                }
+            else:
+                # Export team data
+                export_data["data"] = {
+                    "team_analytics": self.get_team_analytics(),
+                    "user_count": len(self.user_preferences),
+                    "session_count": len(self.user_sessions),
+                }
+
+            with open(output_path, "w") as f:
+                json.dump(export_data, f, indent=2)
+
+            return True
+
+        except Exception as e:
+            logger.error(f"Error exporting data: {e}")
+            return False
+
+    def get_realtime_metrics(self, user_id: Optional[str] = None) -> Dict[str, Any]:
+        """Get real-time analytics metrics"""
+        current_time = datetime.now()
+
+        metrics: Dict[str, Any] = {
+            "timestamp": current_time.isoformat(),
+            "active_sessions": len(self.active_sessions),
+            "total_sessions_today": 0,
+            "avg_session_duration": 0.0,
+            "current_productivity_scores": [],
+        }
+
+        today_start = current_time.replace(hour=0, minute=0, second=0, microsecond=0)
+
+        # Filter sessions and calculate metrics
+        today_sessions = [session for session in self.user_sessions.values() if session.start_time >= today_start]
+
+        metrics["total_sessions_today"] = len(today_sessions)
+
+        if today_sessions:
+            durations = [s.total_duration_ms for s in today_sessions if s.total_duration_ms > 0]
+            if durations:
+                metrics["avg_session_duration"] = statistics.mean(durations)
+
+        # Current productivity scores
+        productivity_scores: List[Dict[str, Any]] = []
+        for session in self.active_sessions.values():
+            if user_id is None or session.user_id == user_id:
+                current_score = self._calculate_productivity_score(session)
+                productivity_scores.append(
+                    {
+                        "user_id": session.user_id,
+                        "session_id": session.session_id,
+                        "score": current_score,
+                        "state": session.state.value,
+                    }
+                )
+        metrics["current_productivity_scores"] = productivity_scores
+
+        return metrics
+
+
+# Global instance for easy access
+_user_analytics: Optional[UserBehaviorAnalytics] = None
+
+
+def get_user_analytics() -> UserBehaviorAnalytics:
+    """Get or create global user behavior analytics instance"""
+    global _user_analytics
+    if _user_analytics is None:
+        _user_analytics = UserBehaviorAnalytics()
+    return _user_analytics
+
+
+# Convenience functions
+def track_user_action(
+    action_type: UserActionType,
+    user_id: str,
+    session_id: str,
+    action_data: Dict[str, Any],
+    context: Optional[Dict[str, Any]] = None,
+    duration_ms: Optional[float] = None,
+    success: bool = True,
+) -> None:
+    """Track a user action"""
+    analytics = get_user_analytics()
+    analytics.track_action(action_type, user_id, session_id, action_data, context, duration_ms, success)
+
+
+def start_user_session(user_id: str, working_directory: str = "", git_branch: str = "") -> str:
+    """Start tracking a user session"""
+    analytics = get_user_analytics()
+    return analytics.start_session(user_id, working_directory, git_branch)
+
+
+def end_user_session(session_id: str) -> Optional[UserSession]:
+    """End a user session"""
+    analytics = get_user_analytics()
+    return analytics.end_session(session_id)
+
+
+def get_user_patterns(user_id: str, days: int = 30) -> Dict[str, Any]:
+    """Get user behavior patterns"""
+    analytics = get_user_analytics()
+    return analytics.get_user_patterns(user_id, days)
+
+
+def get_user_insights(user_id: str, days: int = 7) -> Dict[str, Any]:
+    """Get user insights and recommendations"""
+    analytics = get_user_analytics()
+    return analytics.get_user_insights(user_id, days)
+
+
+if __name__ == "__main__":
+    # Example usage
+    print("Testing User Behavior Analytics...")
+
+    analytics = UserBehaviorAnalytics()
+
+    # Simulate user session
+    user_id = "test_user"
+    session_id = analytics.start_session(user_id, "/Users/goos/MoAI/MoAI-ADK", "main")
+
+    # Track some actions
+    analytics.track_action(
+        UserActionType.COMMAND_EXECUTION,
+        user_id,
+        session_id,
+        {"command": "/moai:1-plan test feature", "tool": "spec_builder"},
+        duration_ms=1500,
+        success=True,
+    )
+
+    analytics.track_action(
+        UserActionType.TOOL_USAGE,
+        user_id,
+        session_id,
+        {"tool": "git", "operation": "commit"},
+        duration_ms=800,
+        success=True,
+    )
+
+    # End session
+    ended_session = analytics.end_session(session_id)
+
+    if ended_session:
+        print("Session completed:")
+        print(f"  Duration: {ended_session.total_duration_ms / 1000:.1f} seconds")
+        print(f"  Commands: {ended_session.total_commands}")
+        print(f"  Productivity Score: {ended_session.productivity_score}")
+        print(f"  Tools Used: {list(ended_session.tools_used)}")
+
+    # Get patterns
+    patterns = analytics.get_user_patterns(user_id, days=1)
+    print(f"User patterns: {patterns}")
+
+    print("User Behavior Analytics test completed!")