Testing Setup logo

Testing Setup

OrganizationPopular
android
testing-setup

Analyze and create a testing strategy for native Android apps - install testing libraries, set up test infrastructure, create harnesses for unit tests, UI tests, screenshot tests, and end-to-end tests.

Overview

Publisherandroid
Repositoryskills
Skill nametesting-setup
Stars
7.4K
Forks
484
Bundled files
3
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by android on GitHub. Read the source before you install it.

Installation

Install the Testing Setup AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/android/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/testing/testing-setup .claude/skills/testing-setup
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Testing Setup in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Testing Setup on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Testing Setup is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Step 1: analyze the current testing setup

To understand the testing setup of an existing project, look for these dependencies in the libs.versions.toml file, or build files:

  1. Dependency Injection framework used. Examples: Hilt, Koin, Anvil, vanilla Dagger...
  2. Unit (local) testing framework this project uses, Example JUnit4, JUnit5...
  3. Mocking framework (if any) used for unit tests, and for Instrumented and UI tests. Examples: Mockito, Mockk...
  4. Robolectric. It can be used in 3 ways:
    1. Used in unit tests to have fakes for platform entities
    2. To run behavior UI tests without a device or emulator. For example, used to run Espresso or Compose tests.
    3. To do screenshot testing with Roborazzi
  5. Is the app 100% Compose, Views or hybrid?
  6. Behavior UI tests:
    1. Compose Tests (androidx.compose.ui:ui-test-*)
    2. Espresso Tests for Views. Might use wrappers like Kaspresso. Dependencies: androidx.test.espresso:espresso-core, androidx.test:runner, androidx.test:rules.
  7. Screenshot tests can be:
    1. Instrumented (device-based). For example, using Dropshots.
    2. Based on Robolectric, so they run locally. For example, using Roborazzi.
    3. Based on LayoutLib, so they run locally. For example, Paparazzi, or the Compose Preview Screenshot Testing tool.
  8. End to end tests (also known as Release Candidate tests) always run on device and use high-level frameworks such as UIAutomator, Appium, or Robotium.
  9. Generate a Markdown report with the analysis

Step 2: Set up Dependency Injection frameworks for testing

If there is no Dependency Injection framework, install one: if it's a multiplatform application, ask the user whether they want to install Koin, or kotlin-inject. If it's not multiplatform, install Hilt.

Install the testing dependencies (for example com.google.dagger:hilt-compiler that should be applied with a kspAndroidTest configuration).

[!IMPORTANT] Important: Always consult the documentation of the applicable framework to learn about testing (for example: Hilt testing guide, Koin Instrumented tests).

For instrumented tests, create and configure (by adding testInstrumentationRunner to the build gradle files) a new test runner and apply the testing rules required by the framework (for example in Hilt, annotate your test classes with @HiltAndroidTest and apply the HiltAndroidRule). Other frameworks use other mechanisms, consult their documentation.

Step 3: Install frameworks

Unless otherwise specified, respect the current stack of testing frameworks.

If there are no testing frameworks, and the user didn't specify any preference, install the following:

  • JUnit4 for local and instrumented tests
  • Jacoco for test coverage
  • For UI tests: if the project has views, Espresso. If it's fully Compose, use the Compose Testing APIs.
  • Robolectric to run UI Tests
  • Compose Preview Screenshot Testing tool for screenshot tests - check setup documentation and follow it strictly.
  • Dropshots for device screenshot tests
  • If a mocking framework is necessary, install Mockk (io.mockk:mockk). Do not install it unless it is clearly necessary.

If instrumented screenshot tests are requested, install Dropshots.

If end-to-end testing is requested, install UI Automator.

Step 4: Refactor and create fakes for testing

Refactor for unit tests

In the next sections you'll be asked to create tests. If you have dependencies on Android framework classes, or entities that are not part of the codebase:

  • First, use a fake. If it doesn't exist, create an interface for the class and a "Default" implementation with the existing code. Add the Fake version to the test sourceset (test or androidTest).

  • If not possible to use a fake (example: no access to the class or interface), mock the dependencies.

Refactor for UI tests

If you need to fake components to make testing easier and faster and more reliable, replace slow and problematic dependencies with fakes. Use runtime fakes using the Dependency Injection framework installed to:

  • Simulate different scenarios with the user (wrong credentials, reset password flow...), with a server (no connection, server down, bad JSON from server...) or with a platform component (insufficient permissions, no disk space, no front camera available)
  • Improve speed and reliability (replacing a database with an in-memory database, replacing a repository with an in-memory fake to avoid hitting the network)

Step 5: Unit testing

Create a task to add or review unit tests in every file that contains business logic (ViewModels, Repositories, database-related classes such as DAOs, etc.). Don't create unit tests for Activities, Compose layouts, or dependency injection configuration files.

Step 6: UI testing

Espresso or Compose UI tests live in the test sourceset because they will be run with Robolectric. If instrumented (emulator or device) tests are requested, put them in the androidTest sourceset.

Step 7: Test databases

If the database is using SQLite (using Room, SQLDelight, etc.), create instrumented tests using an in-memory database to make sure that they work with the SQLite engine on device.

Step 8: Screenshot tests

Irrespective of the framework used, screenshot tests focus on 2 types of tests:

  • Screen-level screenshot tests, where each screen is tested in 9 different sizes, combining compact, medium and expanded widths (400, 610, 900 dp) and heights (400, 500 and 1000 dp).
  • Screen-level variations. Add a mobile (400x500) screenshot of:
    • All the alternative themes, if used.
    • Font scale set to 1.5.
  • Component-level screenshot tests, where each component is tested in different themes and font scales.

Behavior isn't tested with screenshots, but do test different common scenarios if their UIs change a lot depending on the state. For example, test loading screens by injecting a loading state to the UI or simulating it with a fake.

Step 9: UI Behavior tests

Test the UI logic using behavior tests, which ensures that the UIs react as expected when different states are passed, and when user actions are performed.

Compose UI behavior tests

  • Use the ComposeTestRule with a ComponentActivity to access resources such as strings.
  • Always try to match with semantic matchers first. If the matcher is too complicated to write (using more than 3 matchers to find a single element), use testTag.
  • Always verify state restoration

Views (XML) UI behavior tests

Use Espresso to match views and interact with them.

Step 10: Navigation tests

Create a test suite to verify navigation logic. Include:

  • Back handling
  • Deeplinks
  • Special patterns like "exit through home" with multiple backstacks.

Step 11: Simulate different window sizes and settings

For Compose layouts, use DeviceConfigurationOverride described in "UI testing common patterns" to simulate different window sizes, font scales

Step 12: End-to-end tests

Create a low number (about 5% of all tests) of end-to-end tests that cover big user journeys. Use Compose Test APIs or Espresso for that. If you have to access platform features (notifications, system UI...), use UI Automator.

If you need to take screenshots of the app running in a device, use Dropshots. You need a device for screenshot tests when verifying interaction with the system UI (examples: edge-to-edge rendering, notifications, picture-in-picture)

Step 13: Instrumented Screenshot tests

Install the com.dropbox.dropshots plugin in the module and a Dropshots() JUnit Rule. Create a new instrumented screenshot test for one of the app's features.

Step 14: Install jacoco

Install jacoco for local testing code coverage.

  • Add the jacoco plugin to each module that contains tests.

Final touches

  • Ask whether to document the findings of the analysis and the changes applied to the testing strategy. If the user agrees:

    • If there is an AGENTS.md file present in the project, update it with any changes you've made to the testing strategy.

    • If there is no AGENTS.md file, create a new file (docs/testing.md) with a description of the testing strategy, including the commands needed to run every type of test, where the screenshot reference files live, etc. Also create a new AGENTS.md file in the root and create a link to docs/testing.md.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Testing Setup AI skill do?

Analyze and create a testing strategy for native Android apps - install testing libraries, set up test infrastructure, create harnesses for unit tests, UI tests, screenshot tests, and end-to-end tests.

Why use Testing Setup on TypingMind?

Because you install it once and use it with any model. Testing Setup is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Testing Setup in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/android/skills/tree/main/testing/testing-setup. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Testing Setup?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Testing Setup?

As many as you like. As long as a model supports skills, you can use Testing Setup with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Testing Setup AI skill free?

Yes. It is published on GitHub by android under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇