OpenParable
Skip to content
← Back to skills

Mtp Test Running

Testing & Quality
mtp-test-running

Mtp Test Running is an Agent Skill for testing & quality published in the MaterialDesignInXAML/MaterialDesignInXamlToolkit repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Run MaterialDesignInXamlToolkit tests correctly with dotnet test, TUnit, and Microsoft Testing Platform. Use when validating changes, choosing a safe test scope, or filtering to the exact affected tests.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/MaterialDesignInXAML/MaterialDesignInXamlToolkit/blob/HEAD/.github/skills/mtp-test-running/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/mtp-test-running, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Test running with MTP

Use this skill when you need to run or suggest test commands in this repository.

Repository-specific runner setup

  1. Test projects use UseMicrosoftTestingPlatformRunner=true.
  2. Test projects also set TestingPlatformDotnetTestSupport=true.
  3. global.json does not opt into test.runner = Microsoft.Testing.Platform.
  4. That means this repo currently uses dotnet test in the legacy CLI mode while delegating execution to MTP-backed test applications.
  5. Runner-specific arguments must therefore be passed after an extra --.
  6. Do not rely on top-level dotnet test --filter ... for these projects.

Strong preferences

  1. Start with the smallest non-UI test scope that covers the change.
  2. Avoid MaterialDesignThemes.UITests unless the change is inherently runtime WPF behavior.
  3. Only run UI tests after the user explicitly asks or grants permission.
  4. When UI tests are necessary, run only the affected tests, never the whole UI suite.

TUnit filter syntax

TUnit uses --treenode-filter.

Format:

/<Assembly>/<Namespace>/<Class>/<Test>

  • Use * as a wildcard within a segment.
  • Use (A)|(B) inside a segment for OR.
  • Prefer exact class and test names over namespace-wide filters.
  • Add --minimum-expected-tests <n> so an empty selection fails loudly.

Command patterns

Example targeted non-UI test run

dotnet test tests\MaterialDesignThemes.Wpf.Tests\MaterialDesignThemes.Wpf.Tests.csproj -c Release --no-build -- --treenode-filter "/*/*/AutomationPropertiesNameConverterTests/*" --minimum-expected-tests 1 --no-ansi --no-progress

Example targeted UI validation

dotnet test tests\MaterialDesignThemes.UITests\MaterialDesignThemes.UITests.csproj -c Release --no-build -- --treenode-filter "/*/*/DecimalUpDownTests/(NumericButtons_WithMaximum_DisablesPlusButton)|(NumericButtons_WithMinimum_DisablesMinusButton)" --minimum-expected-tests 2 --no-ansi --no-progress

Verify a UI filter before running

tests\MaterialDesignThemes.UITests\bin\Release\net10.0-windows\MaterialDesignThemes.UITests.exe --list-tests --no-ansi --treenode-filter "/*/*/DecimalUpDownTests/(NumericButtons_WithMaximum_DisablesPlusButton)|(NumericButtons_WithMinimum_DisablesMinusButton)"

Mistakes to avoid

  • Using top-level dotnet test --filter ... for these MTP-backed projects.
  • Passing runner-specific flags before the extra --.
  • Running broad UI suites when one or two targeted tests are enough.
  • Forgetting --minimum-expected-tests, which can hide a bad filter.

Useful references