Android Ui Verification

If you're testing React Native or native Android apps and need to automate UI verification without leaving Claude, this skill uses ADB commands to tap, swipe, input text, and capture screenshots on an emulator. It walks through a calibration and inspection workflow using uiautomator dumps to find element bounds, then verifies results through both screenshots and logcat output. The approach is methodical, centered taps are calculated from XML bounds, and everything lands in an artifacts folder. It's genuinely useful if you're iterating on UI changes and want autonomous debugging without manual tapping, though you'll need the emulator already running and ADB in your PATH. The documentation is thorough about timing animations and when to stop versus blind-tap.

skills.sh installs: 119
GitHub stars: 39.4K
last updated: Jun 3, 2026

View on GitHub

Install

$ npx -y skills add sickn33/antigravity-awesome-skills --skill android_ui_verification --agent claude-code

Installs into .claude/skills of the current project.

Files

Android UI Verification Skill

This skill provides a systematic approach to testing React Native applications on an Android emulator using ADB commands. It allows for autonomous interaction, state verification, and visual regression checking.

When to Use

Verifying UI changes in React Native or Native Android apps.
Autonomous debugging of layout issues or interaction bugs.
Ensuring feature functionality when manual testing is too slow.
Capturing automated screenshots for PR documentation.

🛠 Prerequisites

Android Emulator running.
adb installed and in PATH.
Application in debug mode for logcat access.

🚀 Workflow

1. Device Calibration

Select the authorized emulator serial with adb devices -l; pass -s SERIAL to every command when multiple devices exist. Before interacting, inspect current display geometry and UI bounds to ensure tap coordinates are accurate.

adb shell wm size

Note: Size overrides, density, rotation and insets affect input coordinates. Prefer fresh UI bounds on the actual selected display rather than deriving coordinates solely from physical size.

SKILL.md on GitHub