verify-android-screen

Capture Android screens to PNG and delegate analysis to a sub-agent.

7|Updated Apr 15, 2026
One-click install
npx skills add https://github.com/himattm/skills --skill verify-android-screen
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verify-android-screen
Source: https://github.com/himattm/skills/tree/main/plugins/android/skills/verify-android-screen
Command: npx skills add https://github.com/himattm/skills --skill verify-android-screen

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Verifies Android UI state by capturing real screenshots when a JSON layout tree cannot answer the question, covering WebView content, animations, visual fidelity, and appearance-based element checks where resource IDs are unknown or unreliable.

Core Features & Use Cases

  • Capture and analyze screenshots to validate UI state in mobile apps.
  • Delegate image interpretation to a specialized sub-agent to minimize main-thread workload.
  • Use cases include confirming visual state after transitions, ensuring pixels match design, and locating elements by appearance.

Quick Start

Capture the current Android screen to a PNG, then delegate analysis to a sub-agent.

Frequently Asked Questions about verify-android-screen

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

FAQPage Schema
How do I verify Android UI state when resource IDs are unknown or unreliable?

You can verify Android UI state by capturing real screenshots to a PNG file. This approach works when a JSON layout tree cannot answer the question, covering WebView content, animations, and appearance-based element checks.

Can I use screenshots to validate visual fidelity and animations in Android automation?

Yes, capturing screenshots validates visual fidelity and animations in Android automation. You can confirm visual state after transitions, ensure pixels match design, and locate elements by appearance using captured screen images.

How does visual verification handle screenshot analysis without overloading the main thread?

Visual verification delegates image interpretation to a specialized sub-agent to minimize main-thread workload. The sub-agent receives a self-contained prompt with the image path and validation criteria, then returns a concise YES/NO answer and justification.

What's the best way to locate Android UI elements by appearance rather than resource IDs?

The best way to locate elements by appearance is capturing the Android screen to a PNG. This bypasses unreliable resource IDs by delegating visual analysis to a sub-agent that evaluates the image and returns a YES/NO answer.

When should I not use JSON layout trees for Android UI testing?

You should avoid JSON layout trees for Android UI testing when dealing with WebView content, animations, visual fidelity checks, or appearance-based element identification. In these cases, screenshot-based visual verification provides more accurate results.

Does Android screenshot verification work for checking WebView content?

Yes, Android screenshot verification works for checking WebView content. Capturing the screen to a PNG handles scenarios where layout trees fail, allowing a sub-agent to analyze the rendered WebView pixels directly.