Amazon Alexa: Deploy an Endpoint

This page describes the important steps for creating and configuring an Alexa Endpoint.

Flows created with Cognigy.AI can be deployed as an Alexa skill and used together with any Amazon hardware from Echo Dot to Echo Show.

This tutorial covers the necessary steps for creating and configuring an Alexa Endpoint and deploying it to your Alexa Skill. If you want information about assigning a flow to the endpoint and setting up Data Management see Endpoints.

Three Step Preparation Process

1. Create an Alexa skill in the Amazon Developer Portal
2. Add necessary intents to your Alexa skill in the Amazon Developer Portal
3. Create an Alexa endpoint and deploy it to your Alexa skill in Cognigy.AI

Create an Alexa Skill

🔗 In the Amazon Developer Portal

Requirements:
✔ An Amazon Developer Account

1.1 Amazon Developer Portal Overview

To create a skill, navigate to Amazon Developer Portal . click on the Alexa tab in the navigation bar.

Figure 1.1: Amazon Developer Portal landing page

1.2 Create the Skill

To create a new skill, click on the Create Skill button on the right. Here you can give the skill any name you want. You can choose whichever language you want, the steps to build the skill are the same for all languages.

Figure 1.2.1: Alexa Skills Overview

On the next page, you will be asked to choose a model. Let's choose the "Custom model" for now since it fits most needs. In this tutorial, you can see how to build Alexa models from scratch.

Figure 1.2.2: Choosing an Alexa Skill Model

Now we have successfully created an Alexa Skill. We can now proceed and add some necessary intents.

Adding neccessary intents to your Alexa Skill

🔗 In the Amazon Developer Portal

Requirements
✔ An Amazon Developer Account
✔ An Alexa Skill

2.1 Intents

If you are not using our sample package, but instead using a pre-configured Alexa Skill, then you have to add the following built-in intents to your Alexa Skill:

• AMAZON.YesIntent
• AMAZON.NoIntent

This ensures that you can use Yes/No type Question nodes in your Flow.

Configuring and Deploying an Alexa Endpoint

🔗 In Cognigy.AI

Requirements
✔ Amazon Developer Account
✔ Alexa Skill

2.1 Creating an Alexa Endpoint

In order to connect a Flow to your Alexa Skill, you have to create an Endpoint of type Alexa as described in Endpoints. You then need to connect the corresponding Endpoint to your Flow.

Figure 2.1: Creating an Alexa Endpoint in Cognigy.AI

2.2 Auto-deploying the Endpoint to your Alexa Skill

Cognigy.AI offers an easy way to automatically deploy your Endpoint to an Alexa Skill of your choice by just the click of a button. If you open the Deployment panel, you will be prompted to Login With Amazon if you haven't done so already. Clicking this button will send you to a secure Amazon login page where you can log in to the Amazon Developer Account that has your Alexa Skills.

Figure 2.2.1: Authenticating with Amazon

When you have successfully logged in to your Amazon Developer Account, a drop-down with all your Alexa Skills should appear. When you select the skill you wish to use, then you can click on the Deploy button. Clicking this button will configure your skill to use this endpoint when invoked.

Figure 2.2.2: Deploying the Endpoint

2.3 Manually deploying the Endpoint

You can also deploy the Endpoint to your skill manually. To do this, go to the top of the Alexa Endpoint and copy the EndpointURL. You need to add this URL as the Endpoint for your Alexa Skill.

Figure 2.3.1: Getting the EndpointURL

To add this EndpointURL in your Alexa Skill, go to your Alexa Skill and navigate to the Endpoints page. Here you have to choose HTTPS as the Service Endpoint Type. When you have selected this, then you can enter the EndpointURL you copied before in one of the URI fields for one of the region. If you have no preference as to which region you need, simply use the default region. The last thing to do is to make sure to select that the development endpoint is a sub-domain.

Figure 2.3.2: Manually deploying the Endpoint in your Alexa Skill