You set up a multi-turn conversation with the skill and make a few assertions on the expected response by using the Test SDK tool.
You have learned how to chain multiple requests to simulate multi-turn conversations with your skill in one user session.
- Test Lambda function you deployed in Lab 4
- S3 Test-Script bucket that was created in Lab 4
- Number guessing skill that was deployed to your Amazon developer account in Lab 1
We recommend you to use an advanced text editor or IDE. You could also use an Online YAML editor if you want.
6.1.1. Open ./scripts/lab06.yml.
The Launch node is still the entry point for the test client but it is not the only node anymore. A second top-level node called GuessFive has been added. The Launch node is referencing to it via a YAML anchor.
Side note: Anchors in YAML are defined as a value to parent nodes and start with an ampersand (e.g. &GuessFive). Referencing to it goes with a star followed by the anchor name (e.g. *GuessFive). An anchor definition always comes before its references. Learn more about YAML anchors
6.1.2. In line 8 there's an intent name missing. Please set it correctly according to what makes sense in this case. Looking at the interaction model of the skill might help you get the correct name.
6.1.3. Upload the lab06.yml file to S3 and again run the Test Lambda function. Look into the log output.
6.1.4. Change the script in a way it is guessing a 6. Upload the file an run the Test Lambda.
6.1.5. Change the script in a way it is guessing a 5 and then a 6. Upload the file an run the Test Lambda.
6.1.6. Change the script in a way it is guessing a 5 and then a 6 and then exits (stop) the session. Upload the file an run the Test Lambda. The intent for stopping a session is named AMAZON.StopIntent