run-unit-tests
Provides instructions to run the unit tests for project in a consistent and cost efficient way
When & Why to Use This Skill
This Claude skill provides a streamlined, token-efficient method for executing unit tests via a specialized Gradle wrapper script. Designed specifically for AI agents, it optimizes the feedback loop by filtering out framework noise and delivering concise, high-signal reports on test successes, failures, and compilation errors, thereby reducing operational costs and accelerating the debugging process.
Use Cases
- Automated Test Execution: Seamlessly run full project test suites within AI-driven development workflows to maintain code quality and prevent regressions.
- Targeted Debugging: Execute specific test classes or individual test methods to isolate bugs quickly without wasting tokens on irrelevant log data.
- Module-Level Validation: Run tests for specific project modules (e.g., persistence, API) to verify localized changes in complex architectures.
- Cost-Effective Error Analysis: Utilize concise failure summaries and filtered stack traces to identify the root cause of broken builds without manual log parsing.
| name | run-unit-tests |
|---|---|
| description | Provides instructions to run the unit tests for project in a consistent and cost efficient way |
Run Unit Tests
We provide a wrapper script around Gradle that provides concise, token-efficient output for test execution. This is especially useful for AI agents running tests.
Usage:
# Run all tests in the project
./bin/run_tests.main.kts
# Run tests for a specific module
./bin/run_tests.main.kts -m persistence
# Run a specific test
./bin/run_tests.main.kts -t "MongoDBSaveGameRepositoryTest.should save and retrieve a save game"
Output Behavior:
- Success (exit 0): Prints
✓ All tests passed successfully - Compilation Error (exit 1): Shows full Gradle output with compilation error details
- Test Failure (exit 1): Shows concise failure information:
- Test class and method name
- Failure message (truncated if too long)
- Relevant stack trace lines (framework noise filtered out)
Example Success:
✓ All tests passed successfully
Example Test Failure:
✗ Build/tests failed
Failed Tests (1):
1. io.dungeons.persistence.mongodb.MongoDBSaveGameRepositoryTest.should save and retrieve a save game()
Message: expected: <wrong-id> but was: <Id(value=a5638bd6-0aa8-474b-8b92-8bc859be044e)>
Stack trace:
org.opentest4j.AssertionFailedError: expected: <wrong-id> but was: <Id(value=a5638bd6...)>
at kotlin.test.junit5.JUnit5Asserter.assertEquals(JUnitSupport.kt:32)
at kotlin.test.AssertionsKt__AssertionsKt.assertEquals(Assertions.kt:63)
Example Compilation Error:
✗ Build/tests failed
Compilation error detected. Gradle output:
> Task :persistence:compileTestKotlin FAILED
e: file:///.../MongoDBSaveGameRepositoryTest.kt:14:18 Unresolved reference 'undefinedVariable'.
FAILURE: Build failed with an exception.
...