# AI Content Processing ## Content Validation - Check if content is AI-generated before processing - Implement proper content type detection - Use traits over static classes for AI processing ## Class Name Processing - Support indexed class names with dot notation (e.g., `'eb-feature-list-title.0'`) - Implement flexible class name targeting - Handle dynamic class name patterns ## AI Workflow Phases - Use phased approach: backend session/pack download → parallel AI content generation → customizer integration - Implement proper phase transitions and error handling - Maintain state consistency across workflow phases ## Component Lifecycle - Cancel all promises/intervals/timeouts when `aiconversation` component unmounts - Prevent memory leaks in AI processing components - Implement proper cleanup in `useEffect` cleanup functions ## Image Mapping - Process both `img src` attributes and CSS `background-image` properties - Ensure comprehensive image replacement functionality - Handle various image reference patterns ## OpenAI API Integration - OpenAI Responses API uses different response structure than Chat Completions API - Function calls are in `output[]` array with `type: 'function_call'` - Handle different response formats appropriately ## AI Process Retrieval - Implement priority logic that uses `api_key` when available - Fall back to `user_id` only when `api_key` is not provided/empty/no match - Maintain proper fallback hierarchy for process identification