Alexa integration steps

Before you start, you need to have done the following:
  • Created a skill model in the This topic will explain how to configure your skill model using a Converse JSON configuration file. You can then extend this configuration in Alexa.

    You will need the skill ID used when you created the skill model. In the alexa developer console, you can find the skill ID on the home page (click View skill ID).

  • The Converse bot should be complete. For example any interactions that use intentions should exist, and the intentions should contain some utterances. Intents are the Converse equivalent of intentions. Any intentions used in interactions are automatically included in the default Alexa configuration.

    You only need a few sample utterances in each intention as you can define more later in the alexa developer console. For more information, see Designing bots for Alexa.

Link the bot to the Alexa Skill

  1. In Designer, click Config and then go to the Integration tab.
  2. In the Skill ID field, paste in the skill ID you used in the alexa developer console to create the skill model.
  3. In Converse, at this stage (or later):
    • Select the interaction that will run first when a user launches the skill.
    • Enter the default help and quit messages that will be spoken by Alexa when the user speaks one of the utterances that triggers the Help or Quit intents. These are required by Alexa.

Copy the JSON configuration file for the skill model

A basic interaction model is set up for you in Converse. This is configures an Alexa intent for each Converse intention used by the bot. It also contains a general intention that will be used by Alexa when no specific intent can be found.

To copy the JSON configuration:

  1. In Designer, click Config and then go to the Integration tab.
  2. Click Copy Code in the Interaction Model JSON field.
  3. In the alexa developer console paste this into the JSON Editor. This sets up:
    • The Skill Invocation Name, which defaults to the name of the bot. This value is used by a user to begin an interaction with a skill. You can change this.
    • An intent for each intention used by the bot.
    • An EOCMessage intent. For more details, see About the EOCMessage intent and the EOCCatchAll slot type.
Note: After pasting in a new or modified JSON configuration file, you must save and build the interaction model.

Link the Alexa Skill to the bot

To integrate Alexa with the deployed bot:

  1. In the alexa developer console on the Endpoint page, select HTTPS as the service endpoint type.
  2. Paste in the endpoint URL. You can copy the URL from the bot: in Converse click Config and then go to the Integration tab.
  3. For deployment environments provided by EngageOne, select the wildcard certificate option (that is the option My development endpoint is a sub-domain of a domain that has a wildcard certificate from a certificate authority).

Configuration for analytics

If you use Dashbot for analytics then add the API key provided by Dashbot to the EngageOne Converse configuration:
  1. In Dashbot, get the Amazon Alexa API key from your Dashbot account.
  2. In EngageOne Converse Designer, click Config and then go to the Analytics tab.
  3. Under Dashbot, paste in the API key.

Account linking

For details of how to set up account linking, see the documentation for the alexa developer console. There is nothing to configure on the Converse side.