tapps-agents 3.6.0__py3-none-any.whl → 3.6.1__py3-none-any.whl

tapps_agents/core/visual_feedback.py

@@ -1,644 +1,644 @@
-"""
-Visual Feedback System for UI/UX Iterative Refinement
-
-Provides visual analysis, comparison, and pattern learning for UI generation.
-"""
-
-import logging
-from dataclasses import asdict, dataclass, field
-from datetime import UTC, datetime
-from enum import Enum
-from typing import Any
-
-from .hardware_profiler import HardwareProfile, HardwareProfiler
-
-logger = logging.getLogger(__name__)
-
-
-class RenderingMode(Enum):
-    """Rendering mode based on hardware capabilities."""
-
-    LIGHTWEIGHT = "lightweight"  # Minimal analysis for NUC
-    STANDARD = "standard"  # Balanced analysis for development machines
-    FULL = "full"  # Full analysis for workstations
-
-
-class VisualElementType(Enum):
-    """Types of visual elements."""
-
-    BUTTON = "button"
-    TEXT = "text"
-    IMAGE = "image"
-    INPUT = "input"
-    CONTAINER = "container"
-    NAVIGATION = "navigation"
-    HEADER = "header"
-    FOOTER = "footer"
-    SIDEBAR = "sidebar"
-    MODAL = "modal"
-    CARD = "card"
-    LIST = "list"
-    GRID = "grid"
-
-
-@dataclass
-class VisualElement:
-    """Represents a visual element in the UI."""
-
-    element_type: VisualElementType
-    position: tuple[float, float]  # (x, y)
-    size: tuple[float, float]  # (width, height)
-    text: str | None = None
-    color: str | None = None
-    background_color: str | None = None
-    font_size: float | None = None
-    z_index: int | None = None
-    metadata: dict[str, Any] = field(default_factory=dict)
-
-
-@dataclass
-class LayoutMetrics:
-    """Layout analysis metrics."""
-
-    spacing_consistency: float  # 0.0 to 1.0
-    alignment_score: float  # 0.0 to 1.0
-    visual_hierarchy: float  # 0.0 to 1.0
-    whitespace_balance: float  # 0.0 to 1.0
-    grid_consistency: float  # 0.0 to 1.0
-    issues: list[str] = field(default_factory=list)
-
-
-@dataclass
-class AccessibilityMetrics:
-    """Accessibility analysis metrics."""
-
-    color_contrast_score: float  # 0.0 to 1.0
-    keyboard_navigable: bool
-    screen_reader_compatible: bool
-    aria_labels_present: bool
-    focus_indicators_present: bool
-    issues: list[str] = field(default_factory=list)
-
-
-@dataclass
-class VisualFeedback:
-    """Visual feedback data structure."""
-
-    timestamp: datetime
-    iteration: int
-    screenshot_path: str | None = None
-    layout_metrics: LayoutMetrics | None = None
-    accessibility_metrics: AccessibilityMetrics | None = None
-    visual_elements: list[VisualElement] = field(default_factory=list)
-    user_interactions: list[dict[str, Any]] = field(default_factory=list)
-    quality_score: float = 0.0  # 0.0 to 1.0
-    issues: list[str] = field(default_factory=list)
-    suggestions: list[str] = field(default_factory=list)
-    metadata: dict[str, Any] = field(default_factory=dict)
-
-    def to_dict(self) -> dict[str, Any]:
-        """Convert to dictionary for serialization."""
-        data = asdict(self)
-        data["timestamp"] = self.timestamp.isoformat()
-        data["visual_elements"] = [
-            {**asdict(elem), "element_type": elem.element_type.value}
-            for elem in self.visual_elements
-        ]
-        return data
-
-
-class VisualFeedbackCollector:
-    """Collects visual feedback from generated UIs."""
-
-    def __init__(self, hardware_profile: HardwareProfile | None = None):
-        """
-        Initialize visual feedback collector.
-
-        Args:
-            hardware_profile: Hardware profile for optimization
-        """
-        self.hardware_profile = hardware_profile or HardwareProfiler().detect_profile()
-        self.rendering_mode = self._get_rendering_mode()
-        self.feedback_history: list[VisualFeedback] = []
-
-    def _get_rendering_mode(self) -> RenderingMode:
-        """Rendering mode. Workstation-like default (hardware taxonomy removed)."""
-        return RenderingMode.FULL
-
-    def collect_feedback(
-        self,
-        iteration: int,
-        screenshot_path: str | None = None,
-        visual_elements: list[VisualElement] | None = None,
-        user_interactions: list[dict[str, Any]] | None = None,
-        metadata: dict[str, Any] | None = None,
-    ) -> VisualFeedback:
-        """
-        Collect visual feedback for an iteration.
-
-        Args:
-            iteration: Iteration number
-            screenshot_path: Path to screenshot
-            visual_elements: List of visual elements
-            user_interactions: List of user interactions
-            metadata: Additional metadata
-
-        Returns:
-            VisualFeedback object
-        """
-        feedback = VisualFeedback(
-            timestamp=datetime.now(UTC),
-            iteration=iteration,
-            screenshot_path=screenshot_path,
-            visual_elements=visual_elements or [],
-            user_interactions=user_interactions or [],
-            metadata=metadata or {},
-        )
-
-        self.feedback_history.append(feedback)
-        return feedback
-
-    def get_feedback_history(self, limit: int | None = None) -> list[VisualFeedback]:
-        """
-        Get feedback history.
-
-        Args:
-            limit: Maximum number of feedback items to return
-
-        Returns:
-            List of VisualFeedback objects
-        """
-        if limit:
-            return self.feedback_history[-limit:]
-        return self.feedback_history.copy()
-
-    def clear_history(self):
-        """Clear feedback history."""
-        self.feedback_history.clear()
-
-
-class VisualAnalyzer:
-    """Analyzes visual elements and layout."""
-
-    def __init__(self, hardware_profile: HardwareProfile | None = None):
-        """
-        Initialize visual analyzer.
-
-        Args:
-            hardware_profile: Hardware profile for optimization
-        """
-        self.hardware_profile = hardware_profile or HardwareProfiler().detect_profile()
-        self.rendering_mode = self._get_rendering_mode()
-
-    def _get_rendering_mode(self) -> RenderingMode:
-        """Rendering mode. Workstation-like default (hardware taxonomy removed)."""
-        return RenderingMode.FULL
-
-    def analyze_layout(
-        self,
-        visual_elements: list[VisualElement],
-        design_spec: dict[str, Any] | None = None,
-    ) -> LayoutMetrics:
-        """
-        Analyze layout quality.
-
-        Args:
-            visual_elements: List of visual elements
-            design_spec: Optional design specification
-
-        Returns:
-            LayoutMetrics object
-        """
-        if not visual_elements:
-            return LayoutMetrics(
-                spacing_consistency=0.0,
-                alignment_score=0.0,
-                visual_hierarchy=0.0,
-                whitespace_balance=0.0,
-                grid_consistency=0.0,
-                issues=["No visual elements found"],
-            )
-
-        # Lightweight mode: Basic analysis only
-        if self.rendering_mode == RenderingMode.LIGHTWEIGHT:
-            return self._analyze_layout_lightweight(visual_elements)
-
-        # Standard/Full mode: Detailed analysis
-        return self._analyze_layout_detailed(visual_elements, design_spec)
-
-    def _analyze_layout_lightweight(
-        self, elements: list[VisualElement]
-    ) -> LayoutMetrics:
-        """Lightweight layout analysis for NUC devices."""
-        if len(elements) < 2:
-            return LayoutMetrics(
-                spacing_consistency=0.5,
-                alignment_score=0.5,
-                visual_hierarchy=0.5,
-                whitespace_balance=0.5,
-                grid_consistency=0.5,
-                issues=["Insufficient elements for analysis"],
-            )
-
-        # Basic spacing check
-        spacing_scores = []
-        for i in range(len(elements) - 1):
-            elem1 = elements[i]
-            elem2 = elements[i + 1]
-            distance = abs(elem1.position[1] - elem2.position[1])
-            spacing_scores.append(1.0 if distance > 0 else 0.0)
-
-        spacing_consistency = (
-            sum(spacing_scores) / len(spacing_scores) if spacing_scores else 0.5
-        )
-
-        return LayoutMetrics(
-            spacing_consistency=spacing_consistency,
-            alignment_score=0.7,  # Simplified
-            visual_hierarchy=0.6,  # Simplified
-            whitespace_balance=0.6,  # Simplified
-            grid_consistency=0.7,  # Simplified
-            issues=[],
-        )
-
-    def _analyze_layout_detailed(
-        self, elements: list[VisualElement], design_spec: dict[str, Any] | None
-    ) -> LayoutMetrics:
-        """Detailed layout analysis for workstations."""
-        issues = []
-
-        # Spacing consistency
-        spacing_scores = []
-        for i in range(len(elements) - 1):
-            elem1 = elements[i]
-            elem2 = elements[i + 1]
-            if abs(elem1.position[0] - elem2.position[0]) < 10:  # Same column
-                distance = abs(elem1.position[1] + elem1.size[1] - elem2.position[1])
-                if distance < 5:
-                    spacing_scores.append(0.0)
-                    issues.append(f"Elements {i} and {i+1} are too close")
-                elif distance > 100:
-                    spacing_scores.append(0.5)
-                    issues.append(f"Elements {i} and {i+1} have excessive spacing")
-                else:
-                    spacing_scores.append(1.0)
-
-        spacing_consistency = (
-            sum(spacing_scores) / len(spacing_scores) if spacing_scores else 0.7
-        )
-
-        # Alignment score
-        x_positions = [elem.position[0] for elem in elements]
-        y_positions = [elem.position[1] for elem in elements]
-
-        x_alignment = (
-            1.0 - (max(x_positions) - min(x_positions)) / 1000.0 if x_positions else 0.5
-        )
-        y_alignment = (
-            1.0 - (max(y_positions) - min(y_positions)) / 1000.0 if y_positions else 0.5
-        )
-        alignment_score = (x_alignment + y_alignment) / 2.0
-
-        # Visual hierarchy (based on size and z-index)
-        sizes = [elem.size[0] * elem.size[1] for elem in elements]
-        if sizes:
-            max_size = max(sizes)
-            hierarchy_scores = [size / max_size for size in sizes]
-            visual_hierarchy = sum(hierarchy_scores) / len(hierarchy_scores)
-        else:
-            visual_hierarchy = 0.5
-
-        # Whitespace balance (simplified)
-        total_area = sum(elem.size[0] * elem.size[1] for elem in elements)
-        whitespace_balance = min(1.0, total_area / 100000.0)  # Simplified calculation
-
-        # Grid consistency
-        grid_scores = []
-        for i in range(len(elements) - 1):
-            for j in range(i + 1, len(elements)):
-                elem1 = elements[i]
-                elem2 = elements[j]
-                # Check if aligned on grid
-                x_diff = abs(elem1.position[0] - elem2.position[0])
-                y_diff = abs(elem1.position[1] - elem2.position[1])
-                if x_diff < 10 or y_diff < 10:
-                    grid_scores.append(1.0)
-                else:
-                    grid_scores.append(0.5)
-
-        grid_consistency = sum(grid_scores) / len(grid_scores) if grid_scores else 0.7
-
-        return LayoutMetrics(
-            spacing_consistency=max(0.0, min(1.0, spacing_consistency)),
-            alignment_score=max(0.0, min(1.0, alignment_score)),
-            visual_hierarchy=max(0.0, min(1.0, visual_hierarchy)),
-            whitespace_balance=max(0.0, min(1.0, whitespace_balance)),
-            grid_consistency=max(0.0, min(1.0, grid_consistency)),
-            issues=issues[:10],  # Limit issues
-        )
-
-    def analyze_accessibility(
-        self, visual_elements: list[VisualElement], html_content: str | None = None
-    ) -> AccessibilityMetrics:
-        """
-        Analyze accessibility features.
-
-        Args:
-            visual_elements: List of visual elements
-            html_content: Optional HTML content for deeper analysis
-
-        Returns:
-            AccessibilityMetrics object
-        """
-        issues = []
-
-        # Lightweight mode: Basic checks only
-        if self.rendering_mode == RenderingMode.LIGHTWEIGHT:
-            return AccessibilityMetrics(
-                color_contrast_score=0.7,  # Simplified
-                keyboard_navigable=True,  # Assume yes
-                screen_reader_compatible=True,  # Assume yes
-                aria_labels_present=False,  # Unknown
-                focus_indicators_present=False,  # Unknown
-                issues=["Lightweight mode: Limited accessibility analysis"],
-            )
-
-        # Standard/Full mode: Detailed analysis
-        color_contrast_score = 0.8  # Simplified - would use actual contrast calculation
-        keyboard_navigable = True
-        screen_reader_compatible = True
-
-        # Check for ARIA labels in HTML
-        aria_labels_present = False
-        if html_content:
-            aria_labels_present = (
-                "aria-label" in html_content or "aria-labelledby" in html_content
-            )
-
-        # Check for focus indicators
-        focus_indicators_present = False
-        if html_content:
-            focus_indicators_present = (
-                ":focus" in html_content or "focus-visible" in html_content
-            )
-
-        if not aria_labels_present:
-            issues.append("Missing ARIA labels for screen readers")
-        if not focus_indicators_present:
-            issues.append("Missing focus indicators for keyboard navigation")
-
-        return AccessibilityMetrics(
-            color_contrast_score=max(0.0, min(1.0, color_contrast_score)),
-            keyboard_navigable=keyboard_navigable,
-            screen_reader_compatible=screen_reader_compatible,
-            aria_labels_present=aria_labels_present,
-            focus_indicators_present=focus_indicators_present,
-            issues=issues,
-        )
-
-    def calculate_quality_score(
-        self,
-        layout_metrics: LayoutMetrics,
-        accessibility_metrics: AccessibilityMetrics,
-        weights: dict[str, float] | None = None,
-    ) -> float:
-        """
-        Calculate overall quality score.
-
-        Args:
-            layout_metrics: Layout analysis metrics
-            accessibility_metrics: Accessibility analysis metrics
-            weights: Optional weights for different metrics
-
-        Returns:
-            Quality score from 0.0 to 1.0
-        """
-        if weights is None:
-            weights = {"layout": 0.6, "accessibility": 0.4}
-
-        layout_score = (
-            layout_metrics.spacing_consistency * 0.2
-            + layout_metrics.alignment_score * 0.2
-            + layout_metrics.visual_hierarchy * 0.2
-            + layout_metrics.whitespace_balance * 0.2
-            + layout_metrics.grid_consistency * 0.2
-        )
-
-        accessibility_score = (
-            accessibility_metrics.color_contrast_score * 0.3
-            + (1.0 if accessibility_metrics.keyboard_navigable else 0.0) * 0.2
-            + (1.0 if accessibility_metrics.screen_reader_compatible else 0.0) * 0.2
-            + (1.0 if accessibility_metrics.aria_labels_present else 0.0) * 0.15
-            + (1.0 if accessibility_metrics.focus_indicators_present else 0.0) * 0.15
-        )
-
-        quality_score = layout_score * weights.get(
-            "layout", 0.6
-        ) + accessibility_score * weights.get("accessibility", 0.4)
-
-        return max(0.0, min(1.0, quality_score))
-
-
-class UIComparator:
-    """Compares UI iterations for improvements."""
-
-    def __init__(self):
-        """Initialize UI comparator."""
-        pass
-
-    def compare_iterations(
-        self, previous: VisualFeedback, current: VisualFeedback
-    ) -> dict[str, Any]:
-        """
-        Compare two iterations to identify improvements.
-
-        Args:
-            previous: Previous iteration feedback
-            current: Current iteration feedback
-
-        Returns:
-            Comparison results
-        """
-        improvements = []
-        regressions = []
-        unchanged = []
-
-        # Compare quality scores
-        if current.quality_score > previous.quality_score:
-            improvements.append(
-                f"Quality score improved from {previous.quality_score:.2f} to {current.quality_score:.2f}"
-            )
-        elif current.quality_score < previous.quality_score:
-            regressions.append(
-                f"Quality score decreased from {previous.quality_score:.2f} to {current.quality_score:.2f}"
-            )
-        else:
-            unchanged.append("Quality score unchanged")
-
-        # Compare layout metrics
-        if previous.layout_metrics and current.layout_metrics:
-            prev_layout = previous.layout_metrics
-            curr_layout = current.layout_metrics
-
-            if curr_layout.spacing_consistency > prev_layout.spacing_consistency:
-                improvements.append("Spacing consistency improved")
-            elif curr_layout.spacing_consistency < prev_layout.spacing_consistency:
-                regressions.append("Spacing consistency decreased")
-
-            if curr_layout.alignment_score > prev_layout.alignment_score:
-                improvements.append("Alignment score improved")
-            elif curr_layout.alignment_score < prev_layout.alignment_score:
-                regressions.append("Alignment score decreased")
-
-            if curr_layout.visual_hierarchy > prev_layout.visual_hierarchy:
-                improvements.append("Visual hierarchy improved")
-            elif curr_layout.visual_hierarchy < prev_layout.visual_hierarchy:
-                regressions.append("Visual hierarchy decreased")
-
-        # Compare accessibility metrics
-        if previous.accessibility_metrics and current.accessibility_metrics:
-            prev_acc = previous.accessibility_metrics
-            curr_acc = current.accessibility_metrics
-
-            if curr_acc.color_contrast_score > prev_acc.color_contrast_score:
-                improvements.append("Color contrast improved")
-            elif curr_acc.color_contrast_score < prev_acc.color_contrast_score:
-                regressions.append("Color contrast decreased")
-
-            if curr_acc.keyboard_navigable and not prev_acc.keyboard_navigable:
-                improvements.append("Keyboard navigation added")
-            elif not curr_acc.keyboard_navigable and prev_acc.keyboard_navigable:
-                regressions.append("Keyboard navigation removed")
-
-        # Compare issues
-        if len(current.issues) < len(previous.issues):
-            improvements.append(
-                f"Reduced issues from {len(previous.issues)} to {len(current.issues)}"
-            )
-        elif len(current.issues) > len(previous.issues):
-            regressions.append(
-                f"Increased issues from {len(previous.issues)} to {len(current.issues)}"
-            )
-
-        return {
-            "improvements": improvements,
-            "regressions": regressions,
-            "unchanged": unchanged,
-            "quality_delta": current.quality_score - previous.quality_score,
-            "iteration_delta": current.iteration - previous.iteration,
-        }
-
-    def get_improvement_trend(
-        self, feedback_history: list[VisualFeedback]
-    ) -> dict[str, Any]:
-        """
-        Analyze improvement trend across multiple iterations.
-
-        Args:
-            feedback_history: List of feedback items
-
-        Returns:
-            Trend analysis
-        """
-        if len(feedback_history) < 2:
-            return {
-                "trend": "insufficient_data",
-                "average_improvement": 0.0,
-                "iterations": len(feedback_history),
-            }
-
-        quality_scores = [fb.quality_score for fb in feedback_history]
-        improvements = []
-
-        for i in range(1, len(quality_scores)):
-            delta = quality_scores[i] - quality_scores[i - 1]
-            improvements.append(delta)
-
-        average_improvement = (
-            sum(improvements) / len(improvements) if improvements else 0.0
-        )
-
-        if average_improvement > 0.05:
-            trend = "improving"
-        elif average_improvement < -0.05:
-            trend = "declining"
-        else:
-            trend = "stable"
-
-        return {
-            "trend": trend,
-            "average_improvement": average_improvement,
-            "iterations": len(feedback_history),
-            "quality_scores": quality_scores,
-            "final_quality": quality_scores[-1] if quality_scores else 0.0,
-        }
-
-
-class VisualPatternLearner:
-    """Learns visual patterns from feedback."""
-
-    def __init__(self):
-        """Initialize visual pattern learner."""
-        self.patterns: dict[str, Any] = {}
-        self.feedback_history: list[VisualFeedback] = []
-
-    def learn_from_feedback(self, feedback: VisualFeedback):
-        """
-        Learn patterns from feedback.
-
-        Args:
-            feedback: Visual feedback to learn from
-        """
-        self.feedback_history.append(feedback)
-
-        # Extract patterns from layout metrics
-        if feedback.layout_metrics:
-            layout = feedback.layout_metrics
-            if layout.spacing_consistency > 0.8:
-                self.patterns["high_spacing_consistency"] = (
-                    self.patterns.get("high_spacing_consistency", 0) + 1
-                )
-            if layout.alignment_score > 0.8:
-                self.patterns["high_alignment"] = (
-                    self.patterns.get("high_alignment", 0) + 1
-                )
-
-        # Extract patterns from accessibility
-        if feedback.accessibility_metrics:
-            acc = feedback.accessibility_metrics
-            if acc.color_contrast_score > 0.8:
-                self.patterns["good_contrast"] = (
-                    self.patterns.get("good_contrast", 0) + 1
-                )
-            if acc.keyboard_navigable:
-                self.patterns["keyboard_navigable"] = (
-                    self.patterns.get("keyboard_navigable", 0) + 1
-                )
-
-    def get_recommendations(self) -> list[str]:
-        """
-        Get recommendations based on learned patterns.
-
-        Returns:
-            List of recommendations
-        """
-        recommendations = []
-
-        if self.patterns.get("high_spacing_consistency", 0) > 5:
-            recommendations.append("Maintain consistent spacing patterns")
-
-        if self.patterns.get("high_alignment", 0) > 5:
-            recommendations.append("Continue using aligned layouts")
-
-        if self.patterns.get("good_contrast", 0) > 5:
-            recommendations.append("Maintain good color contrast")
-
-        if self.patterns.get("keyboard_navigable", 0) < 3:
-            recommendations.append("Improve keyboard navigation support")
-
-        return recommendations
-
-    def clear_patterns(self):
-        """Clear learned patterns."""
-        self.patterns.clear()
-        self.feedback_history.clear()
+"""
+Visual Feedback System for UI/UX Iterative Refinement
+
+Provides visual analysis, comparison, and pattern learning for UI generation.
+"""
+
+import logging
+from dataclasses import asdict, dataclass, field
+from datetime import UTC, datetime
+from enum import Enum
+from typing import Any
+
+from .hardware_profiler import HardwareProfile, HardwareProfiler
+
+logger = logging.getLogger(__name__)
+
+
+class RenderingMode(Enum):
+    """Rendering mode based on hardware capabilities."""
+
+    LIGHTWEIGHT = "lightweight"  # Minimal analysis for NUC
+    STANDARD = "standard"  # Balanced analysis for development machines
+    FULL = "full"  # Full analysis for workstations
+
+
+class VisualElementType(Enum):
+    """Types of visual elements."""
+
+    BUTTON = "button"
+    TEXT = "text"
+    IMAGE = "image"
+    INPUT = "input"
+    CONTAINER = "container"
+    NAVIGATION = "navigation"
+    HEADER = "header"
+    FOOTER = "footer"
+    SIDEBAR = "sidebar"
+    MODAL = "modal"
+    CARD = "card"
+    LIST = "list"
+    GRID = "grid"
+
+
+@dataclass
+class VisualElement:
+    """Represents a visual element in the UI."""
+
+    element_type: VisualElementType
+    position: tuple[float, float]  # (x, y)
+    size: tuple[float, float]  # (width, height)
+    text: str | None = None
+    color: str | None = None
+    background_color: str | None = None
+    font_size: float | None = None
+    z_index: int | None = None
+    metadata: dict[str, Any] = field(default_factory=dict)
+
+
+@dataclass
+class LayoutMetrics:
+    """Layout analysis metrics."""
+
+    spacing_consistency: float  # 0.0 to 1.0
+    alignment_score: float  # 0.0 to 1.0
+    visual_hierarchy: float  # 0.0 to 1.0
+    whitespace_balance: float  # 0.0 to 1.0
+    grid_consistency: float  # 0.0 to 1.0
+    issues: list[str] = field(default_factory=list)
+
+
+@dataclass
+class AccessibilityMetrics:
+    """Accessibility analysis metrics."""
+
+    color_contrast_score: float  # 0.0 to 1.0
+    keyboard_navigable: bool
+    screen_reader_compatible: bool
+    aria_labels_present: bool
+    focus_indicators_present: bool
+    issues: list[str] = field(default_factory=list)
+
+
+@dataclass
+class VisualFeedback:
+    """Visual feedback data structure."""
+
+    timestamp: datetime
+    iteration: int
+    screenshot_path: str | None = None
+    layout_metrics: LayoutMetrics | None = None
+    accessibility_metrics: AccessibilityMetrics | None = None
+    visual_elements: list[VisualElement] = field(default_factory=list)
+    user_interactions: list[dict[str, Any]] = field(default_factory=list)
+    quality_score: float = 0.0  # 0.0 to 1.0
+    issues: list[str] = field(default_factory=list)
+    suggestions: list[str] = field(default_factory=list)
+    metadata: dict[str, Any] = field(default_factory=dict)
+
+    def to_dict(self) -> dict[str, Any]:
+        """Convert to dictionary for serialization."""
+        data = asdict(self)
+        data["timestamp"] = self.timestamp.isoformat()
+        data["visual_elements"] = [
+            {**asdict(elem), "element_type": elem.element_type.value}
+            for elem in self.visual_elements
+        ]
+        return data
+
+
+class VisualFeedbackCollector:
+    """Collects visual feedback from generated UIs."""
+
+    def __init__(self, hardware_profile: HardwareProfile | None = None):
+        """
+        Initialize visual feedback collector.
+
+        Args:
+            hardware_profile: Hardware profile for optimization
+        """
+        self.hardware_profile = hardware_profile or HardwareProfiler().detect_profile()
+        self.rendering_mode = self._get_rendering_mode()
+        self.feedback_history: list[VisualFeedback] = []
+
+    def _get_rendering_mode(self) -> RenderingMode:
+        """Rendering mode. Workstation-like default (hardware taxonomy removed)."""
+        return RenderingMode.FULL
+
+    def collect_feedback(
+        self,
+        iteration: int,
+        screenshot_path: str | None = None,
+        visual_elements: list[VisualElement] | None = None,
+        user_interactions: list[dict[str, Any]] | None = None,
+        metadata: dict[str, Any] | None = None,
+    ) -> VisualFeedback:
+        """
+        Collect visual feedback for an iteration.
+
+        Args:
+            iteration: Iteration number
+            screenshot_path: Path to screenshot
+            visual_elements: List of visual elements
+            user_interactions: List of user interactions
+            metadata: Additional metadata
+
+        Returns:
+            VisualFeedback object
+        """
+        feedback = VisualFeedback(
+            timestamp=datetime.now(UTC),
+            iteration=iteration,
+            screenshot_path=screenshot_path,
+            visual_elements=visual_elements or [],
+            user_interactions=user_interactions or [],
+            metadata=metadata or {},
+        )
+
+        self.feedback_history.append(feedback)
+        return feedback
+
+    def get_feedback_history(self, limit: int | None = None) -> list[VisualFeedback]:
+        """
+        Get feedback history.
+
+        Args:
+            limit: Maximum number of feedback items to return
+
+        Returns:
+            List of VisualFeedback objects
+        """
+        if limit:
+            return self.feedback_history[-limit:]
+        return self.feedback_history.copy()
+
+    def clear_history(self):
+        """Clear feedback history."""
+        self.feedback_history.clear()
+
+
+class VisualAnalyzer:
+    """Analyzes visual elements and layout."""
+
+    def __init__(self, hardware_profile: HardwareProfile | None = None):
+        """
+        Initialize visual analyzer.
+
+        Args:
+            hardware_profile: Hardware profile for optimization
+        """
+        self.hardware_profile = hardware_profile or HardwareProfiler().detect_profile()
+        self.rendering_mode = self._get_rendering_mode()
+
+    def _get_rendering_mode(self) -> RenderingMode:
+        """Rendering mode. Workstation-like default (hardware taxonomy removed)."""
+        return RenderingMode.FULL
+
+    def analyze_layout(
+        self,
+        visual_elements: list[VisualElement],
+        design_spec: dict[str, Any] | None = None,
+    ) -> LayoutMetrics:
+        """
+        Analyze layout quality.
+
+        Args:
+            visual_elements: List of visual elements
+            design_spec: Optional design specification
+
+        Returns:
+            LayoutMetrics object
+        """
+        if not visual_elements:
+            return LayoutMetrics(
+                spacing_consistency=0.0,
+                alignment_score=0.0,
+                visual_hierarchy=0.0,
+                whitespace_balance=0.0,
+                grid_consistency=0.0,
+                issues=["No visual elements found"],
+            )
+
+        # Lightweight mode: Basic analysis only
+        if self.rendering_mode == RenderingMode.LIGHTWEIGHT:
+            return self._analyze_layout_lightweight(visual_elements)
+
+        # Standard/Full mode: Detailed analysis
+        return self._analyze_layout_detailed(visual_elements, design_spec)
+
+    def _analyze_layout_lightweight(
+        self, elements: list[VisualElement]
+    ) -> LayoutMetrics:
+        """Lightweight layout analysis for NUC devices."""
+        if len(elements) < 2:
+            return LayoutMetrics(
+                spacing_consistency=0.5,
+                alignment_score=0.5,
+                visual_hierarchy=0.5,
+                whitespace_balance=0.5,
+                grid_consistency=0.5,
+                issues=["Insufficient elements for analysis"],
+            )
+
+        # Basic spacing check
+        spacing_scores = []
+        for i in range(len(elements) - 1):
+            elem1 = elements[i]
+            elem2 = elements[i + 1]
+            distance = abs(elem1.position[1] - elem2.position[1])
+            spacing_scores.append(1.0 if distance > 0 else 0.0)
+
+        spacing_consistency = (
+            sum(spacing_scores) / len(spacing_scores) if spacing_scores else 0.5
+        )
+
+        return LayoutMetrics(
+            spacing_consistency=spacing_consistency,
+            alignment_score=0.7,  # Simplified
+            visual_hierarchy=0.6,  # Simplified
+            whitespace_balance=0.6,  # Simplified
+            grid_consistency=0.7,  # Simplified
+            issues=[],
+        )
+
+    def _analyze_layout_detailed(
+        self, elements: list[VisualElement], design_spec: dict[str, Any] | None
+    ) -> LayoutMetrics:
+        """Detailed layout analysis for workstations."""
+        issues = []
+
+        # Spacing consistency
+        spacing_scores = []
+        for i in range(len(elements) - 1):
+            elem1 = elements[i]
+            elem2 = elements[i + 1]
+            if abs(elem1.position[0] - elem2.position[0]) < 10:  # Same column
+                distance = abs(elem1.position[1] + elem1.size[1] - elem2.position[1])
+                if distance < 5:
+                    spacing_scores.append(0.0)
+                    issues.append(f"Elements {i} and {i+1} are too close")
+                elif distance > 100:
+                    spacing_scores.append(0.5)
+                    issues.append(f"Elements {i} and {i+1} have excessive spacing")
+                else:
+                    spacing_scores.append(1.0)
+
+        spacing_consistency = (
+            sum(spacing_scores) / len(spacing_scores) if spacing_scores else 0.7
+        )
+
+        # Alignment score
+        x_positions = [elem.position[0] for elem in elements]
+        y_positions = [elem.position[1] for elem in elements]
+
+        x_alignment = (
+            1.0 - (max(x_positions) - min(x_positions)) / 1000.0 if x_positions else 0.5
+        )
+        y_alignment = (
+            1.0 - (max(y_positions) - min(y_positions)) / 1000.0 if y_positions else 0.5
+        )
+        alignment_score = (x_alignment + y_alignment) / 2.0
+
+        # Visual hierarchy (based on size and z-index)
+        sizes = [elem.size[0] * elem.size[1] for elem in elements]
+        if sizes:
+            max_size = max(sizes)
+            hierarchy_scores = [size / max_size for size in sizes]
+            visual_hierarchy = sum(hierarchy_scores) / len(hierarchy_scores)
+        else:
+            visual_hierarchy = 0.5
+
+        # Whitespace balance (simplified)
+        total_area = sum(elem.size[0] * elem.size[1] for elem in elements)
+        whitespace_balance = min(1.0, total_area / 100000.0)  # Simplified calculation
+
+        # Grid consistency
+        grid_scores = []
+        for i in range(len(elements) - 1):
+            for j in range(i + 1, len(elements)):
+                elem1 = elements[i]
+                elem2 = elements[j]
+                # Check if aligned on grid
+                x_diff = abs(elem1.position[0] - elem2.position[0])
+                y_diff = abs(elem1.position[1] - elem2.position[1])
+                if x_diff < 10 or y_diff < 10:
+                    grid_scores.append(1.0)
+                else:
+                    grid_scores.append(0.5)
+
+        grid_consistency = sum(grid_scores) / len(grid_scores) if grid_scores else 0.7
+
+        return LayoutMetrics(
+            spacing_consistency=max(0.0, min(1.0, spacing_consistency)),
+            alignment_score=max(0.0, min(1.0, alignment_score)),
+            visual_hierarchy=max(0.0, min(1.0, visual_hierarchy)),
+            whitespace_balance=max(0.0, min(1.0, whitespace_balance)),
+            grid_consistency=max(0.0, min(1.0, grid_consistency)),
+            issues=issues[:10],  # Limit issues
+        )
+
+    def analyze_accessibility(
+        self, visual_elements: list[VisualElement], html_content: str | None = None
+    ) -> AccessibilityMetrics:
+        """
+        Analyze accessibility features.
+
+        Args:
+            visual_elements: List of visual elements
+            html_content: Optional HTML content for deeper analysis
+
+        Returns:
+            AccessibilityMetrics object
+        """
+        issues = []
+
+        # Lightweight mode: Basic checks only
+        if self.rendering_mode == RenderingMode.LIGHTWEIGHT:
+            return AccessibilityMetrics(
+                color_contrast_score=0.7,  # Simplified
+                keyboard_navigable=True,  # Assume yes
+                screen_reader_compatible=True,  # Assume yes
+                aria_labels_present=False,  # Unknown
+                focus_indicators_present=False,  # Unknown
+                issues=["Lightweight mode: Limited accessibility analysis"],
+            )
+
+        # Standard/Full mode: Detailed analysis
+        color_contrast_score = 0.8  # Simplified - would use actual contrast calculation
+        keyboard_navigable = True
+        screen_reader_compatible = True
+
+        # Check for ARIA labels in HTML
+        aria_labels_present = False
+        if html_content:
+            aria_labels_present = (
+                "aria-label" in html_content or "aria-labelledby" in html_content
+            )
+
+        # Check for focus indicators
+        focus_indicators_present = False
+        if html_content:
+            focus_indicators_present = (
+                ":focus" in html_content or "focus-visible" in html_content
+            )
+
+        if not aria_labels_present:
+            issues.append("Missing ARIA labels for screen readers")
+        if not focus_indicators_present:
+            issues.append("Missing focus indicators for keyboard navigation")
+
+        return AccessibilityMetrics(
+            color_contrast_score=max(0.0, min(1.0, color_contrast_score)),
+            keyboard_navigable=keyboard_navigable,
+            screen_reader_compatible=screen_reader_compatible,
+            aria_labels_present=aria_labels_present,
+            focus_indicators_present=focus_indicators_present,
+            issues=issues,
+        )
+
+    def calculate_quality_score(
+        self,
+        layout_metrics: LayoutMetrics,
+        accessibility_metrics: AccessibilityMetrics,
+        weights: dict[str, float] | None = None,
+    ) -> float:
+        """
+        Calculate overall quality score.
+
+        Args:
+            layout_metrics: Layout analysis metrics
+            accessibility_metrics: Accessibility analysis metrics
+            weights: Optional weights for different metrics
+
+        Returns:
+            Quality score from 0.0 to 1.0
+        """
+        if weights is None:
+            weights = {"layout": 0.6, "accessibility": 0.4}
+
+        layout_score = (
+            layout_metrics.spacing_consistency * 0.2
+            + layout_metrics.alignment_score * 0.2
+            + layout_metrics.visual_hierarchy * 0.2
+            + layout_metrics.whitespace_balance * 0.2
+            + layout_metrics.grid_consistency * 0.2
+        )
+
+        accessibility_score = (
+            accessibility_metrics.color_contrast_score * 0.3
+            + (1.0 if accessibility_metrics.keyboard_navigable else 0.0) * 0.2
+            + (1.0 if accessibility_metrics.screen_reader_compatible else 0.0) * 0.2
+            + (1.0 if accessibility_metrics.aria_labels_present else 0.0) * 0.15
+            + (1.0 if accessibility_metrics.focus_indicators_present else 0.0) * 0.15
+        )
+
+        quality_score = layout_score * weights.get(
+            "layout", 0.6
+        ) + accessibility_score * weights.get("accessibility", 0.4)
+
+        return max(0.0, min(1.0, quality_score))
+
+
+class UIComparator:
+    """Compares UI iterations for improvements."""
+
+    def __init__(self):
+        """Initialize UI comparator."""
+        pass
+
+    def compare_iterations(
+        self, previous: VisualFeedback, current: VisualFeedback
+    ) -> dict[str, Any]:
+        """
+        Compare two iterations to identify improvements.
+
+        Args:
+            previous: Previous iteration feedback
+            current: Current iteration feedback
+
+        Returns:
+            Comparison results
+        """
+        improvements = []
+        regressions = []
+        unchanged = []
+
+        # Compare quality scores
+        if current.quality_score > previous.quality_score:
+            improvements.append(
+                f"Quality score improved from {previous.quality_score:.2f} to {current.quality_score:.2f}"
+            )
+        elif current.quality_score < previous.quality_score:
+            regressions.append(
+                f"Quality score decreased from {previous.quality_score:.2f} to {current.quality_score:.2f}"
+            )
+        else:
+            unchanged.append("Quality score unchanged")
+
+        # Compare layout metrics
+        if previous.layout_metrics and current.layout_metrics:
+            prev_layout = previous.layout_metrics
+            curr_layout = current.layout_metrics
+
+            if curr_layout.spacing_consistency > prev_layout.spacing_consistency:
+                improvements.append("Spacing consistency improved")
+            elif curr_layout.spacing_consistency < prev_layout.spacing_consistency:
+                regressions.append("Spacing consistency decreased")
+
+            if curr_layout.alignment_score > prev_layout.alignment_score:
+                improvements.append("Alignment score improved")
+            elif curr_layout.alignment_score < prev_layout.alignment_score:
+                regressions.append("Alignment score decreased")
+
+            if curr_layout.visual_hierarchy > prev_layout.visual_hierarchy:
+                improvements.append("Visual hierarchy improved")
+            elif curr_layout.visual_hierarchy < prev_layout.visual_hierarchy:
+                regressions.append("Visual hierarchy decreased")
+
+        # Compare accessibility metrics
+        if previous.accessibility_metrics and current.accessibility_metrics:
+            prev_acc = previous.accessibility_metrics
+            curr_acc = current.accessibility_metrics
+
+            if curr_acc.color_contrast_score > prev_acc.color_contrast_score:
+                improvements.append("Color contrast improved")
+            elif curr_acc.color_contrast_score < prev_acc.color_contrast_score:
+                regressions.append("Color contrast decreased")
+
+            if curr_acc.keyboard_navigable and not prev_acc.keyboard_navigable:
+                improvements.append("Keyboard navigation added")
+            elif not curr_acc.keyboard_navigable and prev_acc.keyboard_navigable:
+                regressions.append("Keyboard navigation removed")
+
+        # Compare issues
+        if len(current.issues) < len(previous.issues):
+            improvements.append(
+                f"Reduced issues from {len(previous.issues)} to {len(current.issues)}"
+            )
+        elif len(current.issues) > len(previous.issues):
+            regressions.append(
+                f"Increased issues from {len(previous.issues)} to {len(current.issues)}"
+            )
+
+        return {
+            "improvements": improvements,
+            "regressions": regressions,
+            "unchanged": unchanged,
+            "quality_delta": current.quality_score - previous.quality_score,
+            "iteration_delta": current.iteration - previous.iteration,
+        }
+
+    def get_improvement_trend(
+        self, feedback_history: list[VisualFeedback]
+    ) -> dict[str, Any]:
+        """
+        Analyze improvement trend across multiple iterations.
+
+        Args:
+            feedback_history: List of feedback items
+
+        Returns:
+            Trend analysis
+        """
+        if len(feedback_history) < 2:
+            return {
+                "trend": "insufficient_data",
+                "average_improvement": 0.0,
+                "iterations": len(feedback_history),
+            }
+
+        quality_scores = [fb.quality_score for fb in feedback_history]
+        improvements = []
+
+        for i in range(1, len(quality_scores)):
+            delta = quality_scores[i] - quality_scores[i - 1]
+            improvements.append(delta)
+
+        average_improvement = (
+            sum(improvements) / len(improvements) if improvements else 0.0
+        )
+
+        if average_improvement > 0.05:
+            trend = "improving"
+        elif average_improvement < -0.05:
+            trend = "declining"
+        else:
+            trend = "stable"
+
+        return {
+            "trend": trend,
+            "average_improvement": average_improvement,
+            "iterations": len(feedback_history),
+            "quality_scores": quality_scores,
+            "final_quality": quality_scores[-1] if quality_scores else 0.0,
+        }
+
+
+class VisualPatternLearner:
+    """Learns visual patterns from feedback."""
+
+    def __init__(self):
+        """Initialize visual pattern learner."""
+        self.patterns: dict[str, Any] = {}
+        self.feedback_history: list[VisualFeedback] = []
+
+    def learn_from_feedback(self, feedback: VisualFeedback):
+        """
+        Learn patterns from feedback.
+
+        Args:
+            feedback: Visual feedback to learn from
+        """
+        self.feedback_history.append(feedback)
+
+        # Extract patterns from layout metrics
+        if feedback.layout_metrics:
+            layout = feedback.layout_metrics
+            if layout.spacing_consistency > 0.8:
+                self.patterns["high_spacing_consistency"] = (
+                    self.patterns.get("high_spacing_consistency", 0) + 1
+                )
+            if layout.alignment_score > 0.8:
+                self.patterns["high_alignment"] = (
+                    self.patterns.get("high_alignment", 0) + 1
+                )
+
+        # Extract patterns from accessibility
+        if feedback.accessibility_metrics:
+            acc = feedback.accessibility_metrics
+            if acc.color_contrast_score > 0.8:
+                self.patterns["good_contrast"] = (
+                    self.patterns.get("good_contrast", 0) + 1
+                )
+            if acc.keyboard_navigable:
+                self.patterns["keyboard_navigable"] = (
+                    self.patterns.get("keyboard_navigable", 0) + 1
+                )
+
+    def get_recommendations(self) -> list[str]:
+        """
+        Get recommendations based on learned patterns.
+
+        Returns:
+            List of recommendations
+        """
+        recommendations = []
+
+        if self.patterns.get("high_spacing_consistency", 0) > 5:
+            recommendations.append("Maintain consistent spacing patterns")
+
+        if self.patterns.get("high_alignment", 0) > 5:
+            recommendations.append("Continue using aligned layouts")
+
+        if self.patterns.get("good_contrast", 0) > 5:
+            recommendations.append("Maintain good color contrast")
+
+        if self.patterns.get("keyboard_navigable", 0) < 3:
+            recommendations.append("Improve keyboard navigation support")
+
+        return recommendations
+
+    def clear_patterns(self):
+        """Clear learned patterns."""
+        self.patterns.clear()
+        self.feedback_history.clear()