It’s easy to make a chatbot fail. I am not talking about nonsense input, or input not related to the chatbot domain — literally all state-of-the-art chatbots will fail in these cases. No, I am talking about input making perfectly sense in the domain of the chatbot — simple and short instructions which should trigger one of the main use cases for the chatbot.
Today I was asked to build a proof-of-concept for using Tricentis Tosca to test an Alexa Skill — challenge accepted!
Tosca will be used for Test Case Design, Test Execution and Test Management. Botium provides connectivity to the chatbot technology, is used for conversation scripting and execution and delivers meaningful error logs on test case failure...
The number of Alexa skills exploded to more than 30.000 alone in U.S. in March 2018. It's time to introduce some test automation to Alexa skills developers :-)
With Botium you could combine Cucumber feature files and your chatbot to automate Acceptence Testing.
This article shows how to setup test automation for a Google Assistant App built with Dialogflow. The test object for this article is the typical “tell me a joke” chatbot included as prebuilt agent in Dialogflow.
With Botium we are enabling chatbot developers and testers to ensure consistency and correctness of IBM Watson Chatbots without a single line of code.
This is the third part of an article series about chatbot testing. This part will show up why testing a chatbot is incredibly hard, even when using the right tool.