Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Get best practices for TUnit unit testing, including data-driven tests
.claude/skills/csharp-tunit/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | — | — |
| case-03 | ✗→✓ | ▲ Improved | — | — |
| case-10 | ✗→✓ | ▲ Improved | — | — |
| case-20 | ✗→✓ | ▲ Improved | — | — |
| case-12 | ✗→✓ | ▲ Improved | — | — |
Your goal is to help me write effective unit tests with TUnit, covering both standard and data-driven testing approaches.
[ProjectName].TestsCalculatorTests for Calculator)dotnet test for running tests[Test] attribute for test methods (not [Fact] like xUnit)MethodName_Scenario_ExpectedBehavior[Before(Test)] for setup and [After(Test)] for teardown[Before(Class)] and [After(Class)] for shared context between tests in a class[Before(Assembly)] and [After(Assembly)] for shared context across test classes[Before(TestSession)] and [After(TestSession)]await Assert.That()[DependsOn] attribute if needed)[Arguments] attribute for inline test data (equivalent to xUnit's [InlineData])[MethodData] for method-based test data (equivalent to xUnit's [MemberData])[ClassData] for class-based test dataITestDataSource[Arguments] attributes can be applied to the same test methodawait Assert.That(value).IsEqualTo(expected) for value equalityawait Assert.That(value).IsSameReferenceAs(expected) for reference equalityawait Assert.That(value).IsTrue() or await Assert.That(value).IsFalse() for boolean conditionsawait Assert.That(collection).Contains(item) or await Assert.That(collection).DoesNotContain(item) for collectionsawait Assert.That(value).Matches(pattern) for regex pattern matchingawait Assert.That(action).Throws<TException>() or await Assert.That(asyncAction).ThrowsAsync<TException>() to test exceptions.And operator: await Assert.That(value).IsNotNull().And.IsEqualTo(expected).Or operator for alternative conditions: await Assert.That(value).IsEqualTo(1).Or.IsEqualTo(2).Within(tolerance) for DateTime and numeric comparisons with tolerance[Repeat(n)] to repeat tests multiple times[Retry(n)] for automatic retry on failure[ParallelLimit<T>] to control parallel execution limits[Skip("reason")] to skip tests conditionally[DependsOn(nameof(OtherTest))] to create test dependencies[Timeout(milliseconds)] to set test timeouts[Category("CategoryName")] for test categorization[DisplayName("Custom Test Name")] for custom test namesTestContext for test diagnostics and information[WindowsOnly] for platform-specific tests[NotInParallel] to disable parallel execution for specific tests[ParallelLimit<T>] with custom limit classes to control concurrency[Repeat(n)] with [ParallelLimit<T>] for load testing scenarios[Fact] with [Test][Theory] with [Test] and use [Arguments] for data[InlineData] with [Arguments][MemberData] with [MethodData]Assert.Equal with await Assert.That(actual).IsEqualTo(expected)Assert.True with await Assert.That(condition).IsTrue()Assert.Throws<T> with await Assert.That(action).Throws<T>()[Before(Test)]/[After(Test)]IClassFixture<T> with [Before(Class)]/[After(Class)]Why TUnit over xUnit?
TUnit offers a modern, fast, and flexible testing experience with advanced features not present in xUnit, such as asynchronous assertions, more refined lifecycle hooks, and improved data-driven testing capabilities. TUnit's fluent assertions provide clearer and more expressive test validation, making it especially suitable for complex .NET projects.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-08 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-16 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-24 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-03 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-19 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-23 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-04 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-10 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-18 | pass→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-20 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-12 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-09 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-11 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-07 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-05 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-06 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-17 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-01 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-02 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-15 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-22 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-13 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-14 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-21 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
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. 24 cases were attempted. The headline lift of +58 percentage points is the difference between those two pass rates over the 24 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.
Other measured skills in the registry, with their headline benchmark lift.