skills.rest
bidah
bidahCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

on-device-ai

Integrate on-device AI runtimes into React Native apps with ExecuTorch hooks.

28|1|Updated Apr 8, 2026
Search Tags:#ocr#react-native#computer-vision#audio-processing#on-device-ai#executorch#mobile-llm
One-click install
npx skills add https://github.com/bidah/react-native-hifi --skill on-device-ai-bidah
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: on-device-ai
Source: https://github.com/bidah/react-native-hifi/tree/main/skills/software-mansion-best-practices/on-device-ai
Command: npx skills add https://github.com/bidah/react-native-hifi --skill on-device-ai-bidah

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Integrates React Native Executorch so mobile developers can add AI features without relying on cloud services, removing privacy, latency, and offline limitations while keeping models on-device.

Core Features & Use Cases

  • On-device LLMs: Load ExecuTorch-powered language models with hooks like useLLM, configure chat, tool calling, structured outputs, and streaming.
  • Vision, OCR, and audio utilities: Use hooks for classification, detection, segmentation, style transfer, text-to-image, OCR, vertical OCR, speech-to-text, text-to-speech, VAD, and embeddings, all referencing the curated models.
  • Device-aware planning: Guidance on selecting quantized models, managing downloads, handling audio sample rates, and monitoring storage/performance ensures features work across low-end to high-end devices.
  • Scenario: When building a privacy-first assistant that must chat, scan receipts, and transcribe voice locally, follow these references and best practices to stay offline-ready.

Quick Start

Ask the agent to wire up React Native Executorch hooks for an on-device chatbot, vision, and audio assistant so every AI call stays local.

Frequently Asked Questions about on-device-ai

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run LLMs locally in a React Native app without cloud dependencies?▼

Run LLMs locally in a React Native app by integrating React Native Executorch hooks like useLLM, which handle on-device inference for chat, tool calling, and streaming without relying on cloud services.

Can I perform OCR and computer vision tasks entirely on-device in React Native?▼

Yes, you can perform on-device OCR and computer vision in React Native using Executorch hooks for text recognition, classification, detection, and segmentation without sending data to a server.

Does React Native Executorch support audio processing for offline speech-to-text?▼

React Native Executorch supports offline audio processing by providing hooks for speech-to-text, text-to-speech, and voice activity detection, enabling local voice interfaces and transcription.

What is the best way to manage model downloads and device constraints for on-device AI?▼

Manage model downloads and device constraints by selecting quantized ExecuTorch models and monitoring storage and performance, ensuring AI features run efficiently across low-end to high-end devices.

How do I build a privacy-first offline assistant with chat, vision, and audio features?▼

Build a privacy-first offline assistant by wiring up React Native Executorch hooks for on-device LLMs, OCR, and audio inference, keeping all chat, scanning, and transcription tasks completely local.

Related Skills

Explore More Software Engineering Skills→
software-mansionsoftware-mansion
1.7k

react-native-executorch

Integrate on-device AI models into React Native applications using ExecuTorch.

Official
Advanced
XSpoonAiXSpoonAi
17

vector-notes-indexer

Index local notes with vector embeddings for offline semantic search.

Official
Advanced
ZpankzZpankz
7

limitless-cli

Query Limitless Pendant lifelogs and chats via FalkorDBLite semantic search.

Community
Advanced
rnavarychrnavarych
14

domain-iot:edge-computing

Provides guidance for designing IoT edge computing systems with gateway architecture and offline resilience.

Community
Advanced