Install any skill in seconds. Free to start, no credit card required.
Get Started Free →How to build, run and verify the .NET sample projects in the Agent Framework repository. Use this when a user wants to verify that the samples still function as expected.
.claude/skills/microsoft-verify-dotnet-samples/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | -10% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -48% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 12% | 0% |
We should only support verifying samples that:
Always report to the user which samples were run and which were not, and why.
Samples should be verified to ensure that they actually work as intended and that their output matches what is expected. For each sample that is run, output should be produced that shows the result and explains the reasoning about what output was expected, what was produced, and why it didn't match what the sample was expected to produce.
Steps to verify a sample:
text [Sample Name] Succeeded
text [Sample Name] Failed Actual Output: [What the sample produced] Expected Output: [Explanation of what was expected and why the actual output didn't match expectations]
Most samples use environment variables to configure settings.
csharpvar endpoint = Environment.GetEnvironmentVariable("AZURE_OPENAI_ENDPOINT") ?? throw new InvalidOperationException("AZURE_OPENAI_ENDPOINT is not set."); var deploymentName = Environment.GetEnvironmentVariable("AZURE_OPENAI_DEPLOYMENT_NAME") ?? "gpt-4o-mini";
To run a sample, the environment variables should be set first. Before running a sample, check whether each environment variable in the sample has a value and then give the user a list of environment variables to set.
You can provide the user some examples of how to set the variables like this:
bashexport AZURE_OPENAI_ENDPOINT="https://my-openai-instance.openai.azure.com/" export AZURE_OPENAI_DEPLOYMENT_NAME="gpt-4o-mini"
To check if a variable has a value use e.g.:
bashecho $AZURE_OPENAI_ENDPOINT
bashcd dotnet/samples/<category>/<sample-dir> dotnet run
For multi-targeted projects (e.g., Durable console apps), specify the framework:
bashdotnet run --framework net10.0
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 6,146 | 8,760 | +43% | 1 | 1 | 0% | 207 | 1,074 | +419% | 0 | 0 | — |
case-02 | fail→fail | 6,090 | 6,847 | +12% | 1 | 1 | 0% | 399 | 1,135 | +184% | 0 | 0 | — |
case-03 | pass→fail | 20,552 | 5,207 | -75% | 1 | 1 | 0% | 3,330 | 1,000 | -70% | 0 | 0 | — |
case-04 | fail→fail | 14,420 | 4,012 | -72% | 1 | 1 | 0% | 3,216 | 866 | -73% | 0 | 0 | — |
case-05 | pass→pass | 14,587 | 7,851 | -46% | 1 | 1 | 0% | 2,907 | 2,319 | -20% | 0 | 0 | — |
case-06 | pass→pass | 15,008 | 11,545 | -23% | 1 | 1 | 0% | 2,238 | 2,874 | +28% | 0 | 0 | — |
case-07 | pass→pass | 2,863 | 2,573 | -10% | 1 | 1 | 0% | 523 | 1,184 | +126% | 0 | 0 | — |
case-08 | pass→pass | 9,584 | 2,624 | -73% | 1 | 1 | 0% | 1,626 | 1,154 | -29% | 0 | 0 | — |
case-09 | fail→pass | 7,826 | 2,780 | -64% | 1 | 1 | 0% | 1,329 | 1,201 | -10% | 0 | 0 | — |
case-10 | pass→pass | 7,197 | 4,601 | -36% | 1 | 1 | 0% | 1,414 | 1,531 | +8% | 0 | 0 | — |
case-11 | fail→fail | 9,712 | 1,838 | -81% | 1 | 1 | 0% | 1,517 | 1,029 | -32% | 0 | 0 | — |
case-12 | fail→pass | 13,811 | 2,834 | -79% | 1 | 1 | 0% | 2,207 | 1,229 | -44% | 0 | 0 | — |
case-13 | fail→pass | 7,621 | 2,369 | -69% | 1 | 1 | 0% | 1,132 | 919 | -19% | 0 | 0 | — |
case-14 | fail→pass | 13,711 | 3,011 | -78% | 1 | 1 | 0% | 2,185 | 1,143 | -48% | 0 | 0 | — |
case-15 | pass→pass | 5,993 | 1,917 | -68% | 1 | 1 | 0% | 1,038 | 1,010 | -3% | 0 | 0 | — |
case-16 | pass→pass | 12,788 | 6,648 | -48% | 1 | 1 | 0% | 2,089 | 1,767 | -15% | 0 | 0 | — |
case-17 | pass→pass | 11,478 | 2,200 | -81% | 1 | 1 | 0% | 1,659 | 955 | -42% | 0 | 0 | — |
case-18 | pass→pass | 11,650 | 2,762 | -76% | 1 | 1 | 0% | 1,916 | 1,224 | -36% | 0 | 0 | — |
case-19 | pass→pass | 5,770 | 2,204 | -62% | 1 | 1 | 0% | 875 | 1,050 | +20% | 0 | 0 | — |
case-20 | fail→pass | 6,814 | 3,076 | -55% | 1 | 1 | 0% | 1,079 | 1,207 | +12% | 0 | 0 | — |
case-21 | pass→pass | 9,404 | 2,888 | -69% | 1 | 1 | 0% | 1,439 | 1,140 | -21% | 0 | 0 | — |
case-22 | pass→pass | 11,847 | 2,093 | -82% | 1 | 1 | 0% | 1,883 | 1,047 | -44% | 0 | 0 | — |
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, and 18 counted toward the lift figure. The other 4 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +18 percentage points is the difference between those two pass rates over the 18 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
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.
Other measured skills in the registry, with their headline benchmark lift.