You can use the below to test your script. This is a utility that POSTs to your endpoint and simulates an incoming request. Use the mic button to talk and listen to your script.

Request

Intent:
Locale:
Slot Values:
{{slot.name}} :

Raw

{{request|json}}

Response

{{responseSpeech}}

Raw

{{response|json}}

Session

Raw

{{session|json}}

Your Violet Voice App/Conversation needs to be registered so that Alexa knows about it.

Registering

1. Go to Amazon Dev URL

You can find it here: https://developer.amazon.com/alexa/console/ask

If you don't have an account, register as a new developer.

Amazon Developer Accounts are different from Amazon Web Services accounts - the former are around capabilities where Amazon might need to pay you while the later focus on on capabilities where Amazon might need to charge you.

2. Create a new Skill

Skills are created by going to the 'Your Alexa Consoles' menu option, clicking on Skills, and then 'Create Skill'. After that you will need to enter a name for the skill, click on next, and select to create a custom skill.

3. Declare the Interaction Model

On the left side of the navigation, under 'Interaction Model', click on 'JSON Editor' (it is the last item). Copy the value below and 'Save Model'.

{{interactionModel|json}}

4. Skill Information

Beyond the interaction model there are a few key settings to be filled:

  1. On the left side of the navigation, under 'Interaction Model', click on 'Invocation'. Enter an invocation name click on 'Save Model' - this is the name that you will call your script when you launch Alexa
  2. On the left side of the navigation, click on 'Endpoint'.

    If you have deployed this service on Heroku, select 'HTTPS', enter url: and certificate type: 'My development endpoint is a sub-domain of a domain that has a wildcard certificate from a certificate authority'.

    Once you have filled the values out click on 'Save Model'
  3. Build your Model by clicking on 'Custom' on the left side of the navigation, and selecting 'Build Model' on the right side.

5. Testing

Once the model has been built you can now test the skill. You will need to click on the 'Test' tab and enable testing.

You only need to Launch the skill when you want others to be able to discover it without following the above steps.

Your Violet Voice App/Conversation needs to be registered so that Dialogflow knows about it.

Registering

1. Go to Dialogflow Console

You can find it here: https://console.dialogflow.com/

2. Create an Agent

Agents are created easily by clicking on the 'Create Agent' button on the left navigation, entering a name for the agent and clicking on next.

If the agent creation fails, please try verifying that you have a Google Cloud account setup (you can access it here: https://console.cloud.google.com/)

3. Upload the Agent Configuration

On the left side of the navigation, next to the project name, click on the gear symbol to get the agent configuration information. From there click on the 'Export and Import' tab and click on the 'Import From Zip' button. You can access the zip file here:

4. Test on Dialogflow

You should now be ready to test the agent on Dialogflow. To do that:

  1. Verify that the Violet endpoint has been configured correctly. To do this, go to 'Fulfilment' on the left side of the navigation, and make sure that the Webhook is enabled and that the Url is:
  2. Test the agent on the right side in Dialogflow.

5. Testing for Google Home / Actions on Google

Dialogflow Agents are used to have Voice Apps running in Google Home. These are done by creating Actions on the Google Assistant. To do this:

  1. Click on 'Integrations' on the left side of the navigation, and then click on 'Google Assistant'. Once there, click the 'Manage Assistant App'.
  2. Once the 'Actions on Google' console is open, you can click on 'Simulator' on the left navigation to test the Voice App there.
  3. Test on your Google Home or Google's Mobile app, by going in the 'Actions on Google' console and onto the 'Simulator' page, then make sure to click on the button the top-right that enables 'Testing on Device'.