.Net: Bug: Arguments provided in the KernelArguments collection are not visible when prompt is handled

[DmitryLukyanov]

Describe the bug
Looking at this documentation, specifically in code block here. We may see how to define {{$repository}} argument in the prompt and how to provide a value:

    Arguments =
        new KernelArguments(new AzureOpenAIPromptExecutionSettings() { FunctionChoiceBehavior = FunctionChoiceBehavior.Auto() })
        {
            { "repository", "microsoft/semantic-kernel" }
        }

however when running this example and ask model a question about this argument like:

what repository you can query?

the models output is:

I can query a public repository with the name specified as {{$repository}}. Please provide the repository name so I can assist you further.

I feel like there is a misconception here. Any help would be appreciated.

To Reproduce
Steps to reproduce the behavior:

1. Go to 'https://learn.microsoft.com/en-us/semantic-kernel/frameworks/agent/examples/example-chat-agent?pivots=programming-language-csharp#final'
2. Copy the code from the code snipped into an empty console application.
3. Run the chat bot
4. Ask him what repository you can query?

Expected behavior
A github repository "microsoft/semantic-kernel".

Screenshots

Platform

• Language: [C#]
• Source: [e.g. NuGet package version 0.1.0, pip package version 0.1.0, main branch of repository]
• AI model: [gpt-4o-mini] gpt-5 gives a similar result
• IDE: [Visual Studio]
• OS: [Windows]

Other notes
Same behavior (the KernelArguments are not matched with keys in the prompt) with this code