skills.rest
shirulot
shirulotCOMMUNITY
Explore all skills by this author →
GitHub

harmonyos-device-automation

Automate HarmonyOS NEXT device interactions via vision-based screen analysis and HDC commands.

Updated Jul 6, 2026
Search Tags:#automation#testing#vision#hdc#harmonyos#midscene
One-click install
npx skills add https://github.com/shirulot/codex-skill --skill harmonyos-device-automation
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: harmonyos-device-automation
Source: https://github.com/shirulot/codex-skill/tree/main/harmonyos-device-automation
Command: npx skills add https://github.com/shirulot/codex-skill --skill harmonyos-device-automation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @midscene/harmony.

What problem does it solve?

This skill eliminates the need for manual interaction with HarmonyOS devices by providing a vision-based automation layer that works without DOM access or accessibility labels.

Core Features & Use Cases

  • Visual Interaction: Performs taps, swipes, and text input based on real-time screen analysis.
  • Cross-Stack Compatibility: Interacts with any visible element regardless of the underlying technology stack.
  • Use Case: Automate end-to-end testing of a HarmonyOS application by launching the app, navigating through menus, and verifying UI states using natural language commands.

Quick Start

Use the harmonyos-device-automation skill to launch the settings app and verify that the wifi toggle is visible on the screen.

Frequently Asked Questions about harmonyos-device-automation

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

FAQPage Schema
How do I automate HarmonyOS NEXT device interactions without DOM access?▼

Automate HarmonyOS NEXT device interactions by using vision-based screen analysis to perform taps, swipes, and text input without requiring DOM access or accessibility labels.

What's the best way to run end-to-end testing for a HarmonyOS application?▼

Run end-to-end testing for a HarmonyOS application by launching the app, navigating menus, and verifying UI states using natural language commands through a vision-based automation layer.

Does HarmonyOS automation require an HDC environment setup?▼

HarmonyOS automation requires a configured HDC environment and Midscene.js integration to execute synchronous screen-analyze-act loops for device interaction.

Can I automate cross-app navigation on HarmonyOS NEXT using visual analysis?▼

Automate cross-app navigation on HarmonyOS NEXT through real-time visual analysis that interacts with any visible element regardless of the underlying technology stack.

Why use vision-driven automation instead of accessibility labels for mobile UI verification?▼

Vision-driven automation eliminates the need for accessibility labels by analyzing the screen visually, allowing cross-stack compatibility for mobile UI verification workflows.

Related Skills

Explore More Software Engineering Skills→
linhaylinhay
284

harmony-next

Automate HarmonyOS NEXT development tasks with offline API references and diagnostics.

Community
Advanced
shirulotshirulot

android-device-automation

Automate Android UI interactions via visual grounding and screenshot analysis.

Community
Advanced
TrainLCDTrainLCD
71

argent-test-ui-flow

Automate end-to-end UI testing and visual regression analysis for iOS and Android mobile apps.

Official
Advanced
smallnestsmallnest
601

burner-phone

Control Android devices via ADB commands with AI vision for screen analysis.

Community
Intermediate