@umituz/react-native-ai-generation-content 1.17.229 → 1.17.231

package/src/domains/content-moderation/README.md

@@ -0,0 +1,363 @@
+# Content Moderation Domain
+
+Content moderation and filtering system for AI-generated content.
+
+## Overview
+
+The Content Moderation domain provides comprehensive content moderation capabilities for AI-generated content. It helps ensure that generated content meets safety guidelines and community standards.
+
+## Features
+
+- **Text Moderation**: Filter inappropriate text content
+- **Image Moderation**: Detect and filter inappropriate images
+- **Video Moderation**: Moderate video content
+- **Voice Moderation**: Filter audio content
+- **Configurable Rules**: Customize moderation rules and policies
+- **Policy Violation Detection**: Detect and report policy violations
+
+## Installation
+
+This domain is part of `@umituz/react-native-ai-generation-content`.
+
+```bash
+npm install @umituz/react-native-ai-generation-content
+```
+
+## Basic Usage
+
+### Using Moderation Wrapper
+
+```tsx
+import { ModerationWrapper } from '@umituz/react-native-ai-generation-content';
+
+// Wrap generation with moderation
+const wrapper = new ModerationWrapper({
+  enabled: true,
+  onViolation: (result) => {
+    Alert.alert('Content Warning', result.warning);
+  },
+});
+
+// Generate with moderation
+const result = await wrapper.execute({
+  type: 'text-to-image',
+  input: { prompt: 'Your prompt here' },
+  generate: async (input) => {
+    return await generateImage(input);
+  },
+});
+
+if (result.isModerated) {
+  console.log('Content was flagged:', result.violations);
+}
+```
+
+### Direct Moderation
+
+```tsx
+import { moderateText, moderateImage } from '@umituz/react-native-ai-generation-content';
+
+// Moderate text
+const textResult = await moderateText({
+  text: 'Some text to moderate',
+  rules: ['profanity', 'hate-speech', 'violence'],
+});
+
+if (textResult.isSafe) {
+  console.log('Text is safe');
+} else {
+  console.log('Violations:', textResult.violations);
+}
+
+// Moderate image
+const imageResult = await moderateImage({
+  imageBase64: 'base64...',
+  rules: ['nudity', 'violence', 'gore'],
+});
+
+if (imageResult.isSafe) {
+  console.log('Image is safe');
+} else {
+  console.log('Violations:', imageResult.violations);
+}
+```
+
+## Configuration Options
+
+### Moderation Config
+
+```tsx
+interface ModerationConfig {
+  enabled: boolean; // Enable/disable moderation
+  rules?: ModerationRule[]; // Custom moderation rules
+  threshold?: number; // Confidence threshold (0-1)
+  onViolation?: (result: ModerationResult) => void; // Violation callback
+  allowOverride?: boolean; // Allow users to override warnings
+}
+```
+
+### Moderation Rules
+
+```tsx
+interface ModerationRule {
+  id: string;
+  category: 'profanity' | 'hate-speech' | 'violence' | 'sexual' | 'self-harm' | 'other';
+  severity: 'low' | 'medium' | 'high';
+  enabled: boolean;
+  action: 'flag' | 'block' | 'warn';
+}
+```
+
+## Moderation Types
+
+### Text Moderation
+
+```tsx
+import { moderateText } from '@umituz/react-native-ai-generation-content';
+
+const result = await moderateText({
+  text: 'Your text content',
+  rules: ['profanity', 'hate-speech', 'harassment'],
+});
+
+// Result
+interface TextModerationResult {
+  isSafe: boolean;
+  confidence: number;
+  violations: {
+    category: string;
+    severity: 'low' | 'medium' | 'high';
+    confidence: number;
+    position?: { start: number; end: number };
+  }[];
+  filteredText?: string; // Text with violations removed
+}
+```
+
+### Image Moderation
+
+```tsx
+import { moderateImage } from '@umituz/react-native-ai-generation-content';
+
+const result = await moderateImage({
+  imageBase64: 'base64...',
+  rules: ['nudity', 'violence', 'gore', 'racy'],
+});
+
+// Result
+interface ImageModerationResult {
+  isSafe: boolean;
+  confidence: number;
+  violations: {
+    category: string;
+    severity: 'low' | 'medium' | 'high';
+    confidence: number;
+    boundingBox?: { x: number; y: number; width: number; height: number };
+  }[];
+}
+```
+
+### Video Moderation
+
+```tsx
+import { moderateVideo } from '@umituz/react-native-ai-generation-content';
+
+const result = await moderateVideo({
+  videoUrl: 'https://...',
+  rules: ['nudity', 'violence', 'gore'],
+  frameInterval: 5, // Check every 5th frame
+});
+
+// Result
+interface VideoModerationResult {
+  isSafe: boolean;
+  confidence: number;
+  violations: {
+    category: string;
+    severity: 'low' | 'medium' | 'high';
+    timestamp: number; // Time in seconds
+    confidence: number;
+  }[];
+}
+```
+
+### Voice Moderation
+
+```tsx
+import { moderateVoice } from '@umituz/react-native-ai-generation-content';
+
+const result = await moderateVoice({
+  audioUrl: 'https://...',
+  rules: ['profanity', 'hate-speech', 'violence'],
+  language: 'en',
+});
+
+// Result
+interface VoiceModerationResult {
+  isSafe: boolean;
+  confidence: number;
+  transcript?: string; // Transcribed text
+  violations: {
+    category: string;
+    severity: 'low' | 'medium' | 'high';
+    timestamp: number;
+    confidence: number;
+    text?: string; // Transcribed text segment
+  }[];
+}
+```
+
+## Using with AI Features
+
+### Wrap Generation with Moderation
+
+```tsx
+import { generationWrapper, ModerationWrapper } from '@umituz/react-native-ai-generation-content';
+
+// Setup moderation
+const moderation = new ModerationWrapper({
+  enabled: true,
+  rules: [
+    { id: 'no-nudity', category: 'sexual', severity: 'high', enabled: true, action: 'block' },
+    { id: 'no-violence', category: 'violence', severity: 'medium', enabled: true, action: 'warn' },
+  ],
+  onViolation: (result) => {
+    Alert.alert('Content Flagged', result.warning);
+  },
+});
+
+// Wrap generation
+const wrappedGenerate = moderation.wrap({
+  type: 'text-to-image',
+  generate: async (input) => {
+    return await generationWrapper.executeImageFeature({
+      featureType: 'text-to-image',
+      inputData: input,
+    });
+  },
+});
+
+// Use wrapped generation
+const result = await wrappedGenerate({
+  prompt: 'Your prompt here',
+});
+
+if (result.isModerated) {
+  // Content was flagged
+  console.log('Violations:', result.violations);
+} else {
+  // Content is safe
+  console.log('Generated image:', result.imageUrl);
+}
+```
+
+### Custom Moderation Rules
+
+```tsx
+import { ModerationWrapper } from '@umituz/react-native-ai-generation-content';
+
+const customRules = [
+  {
+    id: 'custom-1',
+    category: 'other' as const,
+    severity: 'medium' as const,
+    enabled: true,
+    action: 'warn' as const,
+    check: async (content: string) => {
+      // Custom check logic
+      return {
+        isViolation: content.includes('forbidden-word'),
+        confidence: 1.0,
+      };
+    },
+  },
+];
+
+const moderation = new ModerationWrapper({
+  enabled: true,
+  rules: customRules,
+});
+```
+
+## Component Integration
+
+### ModerationSummary Component
+
+```tsx
+import { ModerationSummary } from '@umituz/react-native-ai-generation-content';
+
+<ModerationSummary
+  violations={moderationResult.violations}
+  onDismiss={() => console.log('Dismissed')}
+  onOverride={() => console.log('Overridden')}
+/>
+```
+
+### Show Warning Helper
+
+```tsx
+import { showContentModerationWarning } from '@umituz/react-native-ai-generation-content';
+
+const result = await moderateText({ text: '...' });
+
+if (!result.isSafe) {
+  showContentModerationWarning({
+    violations: result.violations,
+    onConfirm: () => {
+      // User acknowledged and wants to proceed
+    },
+    onCancel: () => {
+      // User cancelled
+    },
+  });
+}
+```
+
+## Best Practices
+
+1. **Enable Early**: Enable moderation from the start of development
+2. **Custom Rules**: Customize rules based on your use case
+3. **Thresholds**: Adjust confidence thresholds based on your needs
+4. **User Feedback**: Allow users to provide feedback on moderation
+5. **Transparency**: Be transparent about moderation with users
+
+## Configuration Examples
+
+### Strict Moderation
+
+```tsx
+const strictConfig: ModerationConfig = {
+  enabled: true,
+  threshold: 0.3, // Lower threshold = more strict
+  rules: [
+    { id: 'profanity', category: 'profanity', severity: 'low', enabled: true, action: 'block' },
+    { id: 'hate-speech', category: 'hate-speech', severity: 'high', enabled: true, action: 'block' },
+    { id: 'violence', category: 'violence', severity: 'low', enabled: true, action: 'block' },
+    { id: 'sexual', category: 'sexual', severity: 'low', enabled: true, action: 'block' },
+  ],
+};
+```
+
+### Lenient Moderation
+
+```tsx
+const lenientConfig: ModerationConfig = {
+  enabled: true,
+  threshold: 0.8, // Higher threshold = more lenient
+  allowOverride: true,
+  rules: [
+    { id: 'hate-speech', category: 'hate-speech', severity: 'high', enabled: true, action: 'block' },
+    { id: 'sexual', category: 'sexual', severity: 'high', enabled: true, action: 'warn' },
+  ],
+};
+```
+
+## Related Features
+
+- [Prompts](../prompts) - AI prompt management
+- [Creations](../creations) - Manage AI-generated creations
+- [Face Detection](../face-detection) - Face detection API
+
+## License
+
+MIT