Alexa Horoscope Workshop

Workshop Goals

Complete our Horoscope sample skill by

• Creating a HoroscopeForSign Intent which includes a sign slot
• Add a response handler to the code which handles the new intent

Setup instructions (for macOS)

• Install nvm (if you need to switch between node versions)
• Install Node 8.10.0 by running nvm install 8.10.0
• Create Alexa Developer Console account
• Create a new skill on the Alexa Developer Console
  • Select Developer Console on the top menu bar
  • Select Alexa Skills Kit
  • Click Create Skill
  • Input your selected skill name
  • Select the Custom skill model
  • Select Start From Scratch
  • Navigate back to the Your Skills section, and take note of skill ID for your new skill (you will need this later)
• Download the sample project from Git
• Navigate to lambda/custom and run npm install
• Open lambda/custom/lib/constants.js and set the YOUR_APP_ID variable to the skill ID from above
• Open models/en-US.json and set the YOUR_INVOCATION_NAME to your desired invocation name (The default is the skill name)

How to run the skill

In production, Alexa skill code typically runs on a lambda function. For development this is not very practical as you cannot easily change and debug your code on lambda.

To run code in development we use a library called bespoken-tools. It allows us to run our code locally by routing Alexa requests to our local machines.

There are two approaches that you can use to run the project for this workshop:

• Using Webstorm
  • Copy the runConfiguration folder from the sample project into .idea folder
  • Select the new configuration from the top menu bar, and click Run or Debug
• Using the Command line
  • Ensure you are running Node 8.10.0 by running nvm use 8.10.0
  • Install bespoken-tools globally npm install -g bespoken-tools
  • Navigate to lambda/custom
  • Run this command on the command line AWS_REGION="us-east-1"
  • Start the skill using bespoken tools by running the following command bst proxy lambda index.js --verbose

If everything is running correctly, bespoken-tools should output something like the following:

Your public URL for accessing your local service:
https://sick-huxley-JuRHxZ.bespoken.link

This URL is where the Alexa Skills kit will need to route requests to.

Setup on Alexa Developer Console

• Navigate to your skill on the Alexa Developer Console
• In the build tab, do the following:
  • Navigate to the Endpoints section
    • Select HTTPS
    • Input the bespoken-tools URL displayed in the last step
    • Select My development endpoint is a sub-domain of a domain that has a wildcard certificate from a trusted certificate authority
    • Select Save Endpoints at the top of the page
  • Navigate to the JSON Editor section
    • Upload model/en-US.json
    • Click Save Model and then click Build Model
    • Once the build is completed and successful, move on to the next step

Test your skill

Navigate to Test tab on the Alexa Developer Console

• Enable testing using the toggle on the top of the page
• Test if you can communicate with the skill running locally. For example, enter the following into the input field: open YOUR_INVOCATION_NAME

Workshop

Add HoroscopeForSign Intent with a Sign slot in your models

1. Define a Sign custom slot in models/en-US.json
2. Define HoroscopeForSign intent in models/en-US.json that utilizes the Sign slot
3. Add your invocation name at the top of the models file under invocationName
4. Upload and build your modified models file to the Alexa Developer console

Add HoroscopeForSign Response Handler

1. Define a HoroscopeForSign ResponseHandler in lambda/custom/handlers/requests.js.
   • Within your handler, use the following to access the inputted slot: handlerInput.requestEnvelope.request.intent.slots.YOUR_SLOT_NAME
   • Make sure to validate the slot value, responding to the user with an error if necessary
   • Use horoscopeService.fetchBySign() to fetch the horoscope content for the inputted sign.
   • Add any required string responses in lambda/custom/strings/en-us.js
   • Remember that you can use const requestAttributes = handlerInput.attributesManager.getRequestAttributes(); const response = requestAttributes.t("YOUR_STRING_KEY", replaceObj) to fetch strings and make string replacements
2. Run your skill (following the instructions above) and make sure it responds correctly using the Test tab in the Alexa Developer Console

Add Visuals to the HoroscopeForSign Response

1. In the Alexa Developer Console, Navigate to the Interfaces section. Enable Display Interface
2. In the strings file lambda/custom/strings/en-us.js, add display data to the HoroscopeForSign strings (https://github.com/Hearst-DD/ask-toolkit/blob/master/README.md#objects)
3. Run your skill and test that the visuals show up as desired using the Test tab in the Alexa Developer Console.

Helpful links

• Custom skill documentation
• Slot type documentation
• Bespoken tools tutorial
• ask-tools library developed by Hearst ET Private repository. Contact netteam@hearst.com for access.
• https://www.npmjs.com/~hearst-et