Dmytro Struk

> @dmytrostruk, I think you set up the code coverage checking? Can you make a recommendation for what to do here? @stephentoub Yes, I can see some changes in OpenAI...

> I think the problem is that the library is already razor-thin above the code coverage line (main is sitting at 81% branch coverage for the OpenAI connector, and at...

> 3\. _Using moderation classification model_ – The developer should be able to use a [classification model](https://platform.openai.com/docs/guides/moderation) to determine if a prompt is not ok. If a prompt is not...

> 1. _Approving functions before they're run_ – if a function is "consequential", it should require the user to first approve the action before it happens. If the function is...

> 2\. _Semantic caching_ – after a prompt has been invoked, a developer should be able to cache the response by using the original question as the key (i.e., embedding)....

Example of PII detection: https://github.com/microsoft/semantic-kernel/pull/6171

Example of text summarization and translation evaluation: https://github.com/microsoft/semantic-kernel/pull/6262

Example of FrugalGPT: https://github.com/microsoft/semantic-kernel/pull/6815

@ericvdberge Thanks for reporting this issue. Semantic Kernel SDK contains some experimental features, which may be changed/updated in the future, so these errors just notify you about that. In order...

@markwallace-microsoft @matthewbolanos Regarding this issue: Kernel specific collections (e.g. `kernel.FunctionInvocationFilters` etc) allow to modify filter ordering in runtime. Users can insert/remove/sort filter at any position (beginning, middle, end) in collection,...