Best practices for building on-device AI features in React Native using React Native ExecuTorch. Use when the user wants to add AI to a mobile app without cloud dependencies: AI chatbots and assistants, image classification, object detection, text recognition and document parsing (OCR), style transfer, image generation, speech-to-text transcription, text-to-speech synthesis, voice activity detection, semantic search with embeddings, real-time camera AI with VisionCamera, or vision-language image understanding. Also use when the user mentions offline AI, on-device ML, privacy-preserving AI, reducing cloud API costs or latency, running models locally on mobile, or downloading and managing ML models. Covers react-native-executorch hooks (useLLM, useClassification, useObjectDetection, useOCR, useSemanticSegmentation, useInstanceSegmentation, useStyleTransfer, useTextToImage, useImageEmbeddings, useSpeechToText, useTextToSpeech, useVAD, useTextEmbeddings, useExecutorchModule), tool calling, structured output, VLMs
Software Mansion's best practices for production React Native and Expo apps on the New Architecture. MUST USE before writing, reviewing, or debugging ANY code in a React Native or Expo project. If the working directory contains a package.json with react-native, expo, or expo-router as a dependency, this skill applies. Trigger on: any code task in a React Native/Expo project, 'React Native', 'Expo', 'New Architecture', 'Reanimated', 'Gesture Handler', 'react-native-svg', 'ExecuTorch', 'react-native-audio-api', 'react-native-enriched', 'Worklet', 'Fabric', 'TurboModule', 'WebGPU', 'react-native-wgpu', 'TypeGPU', 'GPU shader', 'WGSL', 'svg', 'animation', 'gesture', 'audio', 'rich text', 'AI model', 'multithreading', 'chart', 'vector', 'image filter', 'shared value', 'useSharedValue', 'runOnJS', 'scheduleOnRN', 'thread', 'worklet', or any question involving UI, graphics, native modules, or React Native threading and animation behavior. Also use when a more specific sub-skill matches.