Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Create Unit Test · aiskillstore | DecimalAI

aiskillstore/create-unit-test

SkillScore95Maintained

Create and run unit tests following the project's architecture and guidelines (Robolectric, naming, location).

Testingby aiskillstore

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

QA & Test Engineering pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/aiskillstore-create-unit-test/SKILL.md

decimalai skills pull aiskillstore-create-unit-test --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/aiskillstore-create-unit-test/SKILL.md https://app.decimal.ai/s/aiskillstore-create-unit-test/SKILL.md

Raw file: /s/aiskillstore-create-unit-test/SKILL.md ↗ · machine-readable summary at /s/aiskillstore-create-unit-test.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 26d ago

Yes — +14% lift-23% tokens0% turns95% of whole cases pass with skill (21/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-03	✗→✓	▲ Improved	-6%	0%
case-11	✗→✓	▲ Improved	34%	0%
case-22	✗→✓	▲ Improved	-41%	0%
case-02	✓→✓	＝ Same ✓	-24%	0%
case-04	✓→✓	＝ Same ✓	-17%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/10/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

aiskillstore / marketplace

skills/anysoftkeyboard/create-unit-test/SKILL.md

View on GitHub ↗

Imported 8/10/2026Version v1

Skill contents

SKILL.md

Create Unit Test

This skill guides the creation of unit tests to ensuring consistency and correctness.

Workflow

Analyze and Plan:

Goal: Understand what needs testing.
Action: Examine the source code.
Action: Identify "happy path" cases (core functionality).
Action: Identify "error cases" and "edge cases" based on likelihood.
Action: Do not suggest implementation details unless asked; focus on behavior.

Setup Test File:

Naming: [OriginalClassName]Test.kt.
Location: [module]/src/test/java/[package/path]/.
Runner: If testing Android components, you MUST use Robolectric.
Mocking: Prefer creating Fakes over Mocks or patches. Use Mocks only if Fakes are too complex.

Implement Tests:

Write clear, descriptive test methods.
Follow the "Arrange-Act-Assert" pattern.
Ensure all resources (strings, themes) are properly mocked or provided via Robolectric.

Run Tests:

Command: ./gradlew :[module]:testDebugUnitTest
Example: ./gradlew :database:testDebugUnitTest
Tip: Before any commit, run ./gradlew testDebugUnitTest to ensure everything passes.

Handle Failures:

Goal: Diagnose and fix issues.
Action: If a test fails, locate the full error report.
Path: [module]/build/test-results/testDebugUnitTest/TEST-[package.name.TestClassName].xml.
Example: app/build/test-results/testDebugUnitTest/TEST-com.anysoftkeyboard.janus.app.MainActivityTest.xml.

Guidelines

Consistency: Always place tests in the test source set, not androidTest (unless specifically writing instrumentation tests, which is rare for this skill).
Reliability: Avoid flaky tests. Ensure deterministic behavior.

Illustrative Want to see what aiskillstore/create-unit-test would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

9/3/2026, 05:26 UTC · 22 cases

Lift

+14%

95% vs 82%

Agent turns

0%

1.0 vs 1.0

Tokens

-23%

1,501.59 vs 1,939.09

fail → pass · 3pass → pass · 18fail → fail · 1

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 I just finished writing `UserProfileViewModel.kt` inside the `:feature:profile` module at package `com.example.app.profile`. Could you generate unit tests for this class covering both normal usage and potential error conditions? Please create the proper test file setup and test methods.	fail→fail	23,775	9,261	-61%	1	1	0%	4,087	1,208	-70%	0	0	—
case-02 I need a complete unit test suite for `NotificationHelper.kt` in my `:core` module (`com.example.app.core`). It depends on Android context resources and a data repository. Please generate the test code and let me know which Gradle command I should run to execute unit tests for this specific module.	pass→pass	17,030	9,979	-41%	1	1	0%	3,358	2,567	-24%	0	0	—
case-03 We need unit tests for `PaymentProcessor.kt` located in the `:checkout` module under `com.myapp.checkout`. Please write the unit test file for this class. Also, if any test fails when running Gradle, where in the build directory structure can I find the generated XML error report to inspect what went wrong?	fail→pass	18,466	15,914	-14%	1	1	0%	2,918	2,730	-6%	0	0	—
case-04 I have a class `OrderCalculator.kt` in module `:cart` with package `com.shop.cart`. I want to add tests for business logic. Should I put these files under `src/androidTest/java` or `src/test/java`? Please provide the full path where `OrderCalculatorTest.kt` should be created.	pass→pass	7,308	8,080	+11%	1	1	0%	1,259	1,042	-17%	0	0	—
case-05 I want to run unit tests specifically for the `:database` module in my multi-module Android project from command line. What exact Gradle task should I invoke?	pass→pass	12,036	7,614	-37%	1	1	0%	1,150	884	-23%	0	0	—
case-06 What command should I run right before committing my code to ensure all unit tests across the Android project pass?	pass→pass	11,586	6,756	-42%	1	1	0%	1,058	726	-31%	0	0	—
case-07 A unit test for `AuthRepository` (`AuthRepositoryTest.kt`) failed during execution in module `:auth` under package `com.company.auth`. Where can I find the detailed XML test results file for this specific failure?	pass→pass	12,410	2,257	-82%	1	1	0%	1,297	890	-31%	0	0	—
case-08 I'm writing a unit test for `AccountSettingsActivity.kt` inside `:settings` (`com.app.settings`) which interacts with Android `Context` and `SharedPreferences`. I want to test this locally without starting an emulator or physical device. Which test runner framework should be specified for the test class?	pass→pass	8,661	5,358	-38%	1	1	0%	1,527	1,253	-18%	0	0	—
case-09 I am creating a unit test class for `SessionManager.kt` located in `:network` (`com.app.network`). What exact class name should the test file use?	pass→pass	8,282	2,043	-75%	1	1	0%	502	793	+58%	0	0	—
case-10 I need to write unit tests for `UserRepository.kt` in `:data` (`com.app.data`), which relies on a `UserDao` and `ApiService` interface. I'm deciding whether to mock these interfaces using Mockito/MockK or write fakes. What is the recommended strategy for component dependencies in this testing workflow?	pass→pass	14,897	9,847	-34%	1	1	0%	3,048	2,230	-27%	0	0	—
case-11 I need to write an Espresso UI automation test for `LoginActivity` that launches the activity, types credentials into `R.id.username_input` and `R.id.password_input`, clicks `R.id.login_button`, and asserts that `R.id.welcome_text` is displayed. Please provide the test code.	fail→pass	11,682	6,630	-43%	1	1	0%	1,437	1,923	+34%	0	0	—
case-12 I need to update my `:app` module `build.gradle.kts` file to add dependencies for Kotlin Coroutines and Room database. Please provide the build script configuration.	pass→pass	15,314	5,329	-65%	1	1	0%	1,846	1,616	-12%	0	0	—
case-13 Please write the production implementation code for `DownloadManager.kt` in `:core` (`com.app.core`) that handles downloading files asynchronously using Kotlin Flow and Ktor client.	pass→pass	20,972	14,753	-30%	1	1	0%	4,144	3,618	-13%	0	0	—
case-14 When writing individual unit test methods for `CartService.kt` in `:cart` (`com.app.cart`), what standard structural pattern should be followed within each test function to clearly separate setup, action, and verification?	pass→pass	8,503	3,287	-61%	1	1	0%	1,492	1,058	-29%	0	0	—
case-15 I'm about to start writing unit tests for `OrderProcessor.kt` in `:checkout` (`com.app.checkout`). Should I start by suggesting changes to the internal implementation details of `OrderProcessor.kt`, or how should I plan the tests?	pass→pass	17,905	11,674	-35%	1	1	0%	2,239	1,690	-25%	0	0	—
case-16 In `ThemeHelper.kt` in `:ui` (`com.app.ui`), unit tests need to access string resources (`R.string.app_name`) and theme attributes without running on a physical Android device. How should these Android resources be provided in the local JVM unit test?	pass→pass	12,120	7,607	-37%	1	1	0%	2,275	2,009	-12%	0	0	—
case-17 When planning unit tests for `MediaParser.kt` in `:media` (`com.app.media`), besides testing core happy path scenarios, what other test cases should be identified during the analysis phase?	pass→pass	21,682	13,724	-37%	1	1	0%	2,784	1,904	-32%	0	0	—
case-18 I am working on the `:analytics` module at `com.company.analytics`. What exact terminal command executes debug unit tests for this module?	pass→pass	10,196	7,514	-26%	1	1	0%	886	709	-20%	0	0	—
case-19 My build failed during test execution for `ImageDecoderTest.kt` in module `:media` at `com.app.media`. What is the exact path to the XML report generated for this test failure?	pass→pass	6,782	4,855	-28%	1	1	0%	1,212	907	-25%	0	0	—
case-20 I am testing `AudioPlayerService.kt` in `:audio` (`com.app.audio`). What is the standardized name and directory path for this test file?	pass→pass	6,543	7,692	+18%	1	1	0%	1,183	890	-25%	0	0	—
case-21 When writing tests for an asynchronous background task in `DataSyncer.kt` (`:data`), what key reliability principle must be followed to prevent intermittent failures in CI pipelines?	pass→pass	14,226	11,492	-19%	1	1	0%	1,478	1,517	+3%	0	0	—
case-22 Before writing test code for `SecurityValidator.kt` in `:security` (`com.app.security`), what is the first step in the test creation workflow and what actions does it require?	fail→pass	13,686	7,820	-43%	1	1	0%	1,480	871	-41%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 81.8%
Pass rate with the skill: 95.5%
Cases attempted: 22
Cases counted toward lift: 22
Cases that regressed: 1
Skill version measured: v1
Run completed: 9/3/2026, 05:26 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-03

Prompt

We need unit tests for `PaymentProcessor.kt` located in the `:checkout` module under `com.myapp.checkout`. Please write the unit test file for this class. Also, if any test fails when running Gradle, where in the build directory structure can I find the generated XML error report to inspect what went wrong?

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run95

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns-20% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 25 days ago

Other measured skills in the registry, with their headline benchmark lift.