For the actions execution, we provide an SDK. Using this SDK, you can transform simple actions,or functions, into an intelligent conversations that you can have with the virtual assistant.
Here is an example for a Python function that is capable of getting the pull request on github for a specified repository.
And the result as you can see is an intelligent conversation where the virtual assistant is able to collect all of the relevant context in order to execute this action.
To power this experience and for the AI to know which parameters are required for each action and gather them during a conversation, Kuibya uses pedantic models.
So you can define a pedantic model as a function signature: It gets the model as an input., and based on the model, the AI is able to ask all of the relevant questions based on the parameters.
These parameters can be optional strings, any of any of the specified types in the programming language.
Organization can wrap these actions and create their own, thus templating the experience to their organizational users. Have a home-grown system? Any system that has an API or an SDK can be supported using these model. Creating actions takes minutes and Kubiya providers a template.
Read more about action stores on our docs.
Ready to play around with our DevOps assistant? Signup here.