Amazon Live Chat Integration

You can now integrate Amazon Connect with DRUID to allow your customers chat with the bot and get support from your Contact Center agents who are using Amazon Connect.

Prerequisites

You need to have the call center set up in Amazon Connect.
You need an Amazon Console account.

Configuring the integration

This section provides the steps necessary to place a customer live chat in Amazon Connect (that is, start an Amazon live chat session).

Step 1: Create AWS user dedicated for the Amazon Connect - DRUID Integration

Create a DRUID-dedicated user and assign an IAM policy which grands this user access to Amazon Connect service and to method StartChatContact. You can provide specific flow access in the Policy Editor as per examples below.

Example of IAM Policy which grants user access to all Amazon Connect resources:

Example of IAM Policy which grants user access to specific Amazon Connect resources:

You can use the AWS Policy Generator to generate the policy.

On the user details page, click the Security credentials tab and generate access key for this user by clicking the Create access key button. When the access key is created, the Access key ID displays in the table and you also get the secret key.

Write down or copy / paste in a document the Access key ID and secret key, you will need them later to configure your DRUID bot for Amazon Connect integration.

Note: If you lose or forget your secret key, you cannot retrieve it. Instead, create a new access key and make the old key inactive. You can have a maximum of two access keys (active or inactive) at a time.

Step 2: Get Amazon Connect Instance ID and Flow ID

In Amazon Connect, go to the details page of the flow you want to trigger from DRUID. On the flow menu, click Show additional flow information. From the ARN section, copy the Instance ID and the flow Id and paste them in a document, you will need these IDs to configure your DRUID bot for Amazon Connect integration.

Step 3: Configure DRUID for Amazon Connect

Go to bot details page and click the LiveChatSettings section header. The section expands.

Set the Amazon Live Chat fields as described in the table below

Field	Description
LiveChatProvider	Select Amazon.
AWSConnectServiceUrl	Enter your Amazon Connect endpoint in the following format: https://connect.[REGION].amazonaws.com For the complete list of Amazon regions, see Amazon Connect documentation. Example: https://connect.eu-central-1.amazonaws.com for region eu-central-1.
AWSParticipantServiceUrl	Enter your Amazon Connect Participant Service endpoint in the following format: https://participant.connect.[REGION].amazonaws.com. For the complete list of Amazon regions, see Amazon Connect documentation. Example: https://participant.connect.eu-central-1.amazonaws.com for region eu-central-1.
AWSApiId	The Access key ID created for your DRUID dedicated user (the one from Step 1.)
AWSApiKey	The Access key secret provided to you when creating the access key for your DRUID dedicated user (the one from Step 1).
AWSInstanceId	Amazon Connect Instance ID (the one you got at Step 2).
AWSFlowId	The flow ID that will be triggered in Amazon Connect (the one you got at Step 2).

After you set up the Amazon Live Chat fields, click the Save & Close button to save the settings.

Step 4: Configure the flow step for connecting to Amazon Agent

On the flow of choice, add a step dedicated to connecting clients to Amazon Connect. On the flow step set the following variables:

[[Chatuser]].Name = [[HelpdeskClient]].FirstName
[[ChatUser]].HelpdeskSubject = [[HelpdeskClient]].Subject

On the dedicated step, in the PreActions section, call the internal action ConnectToHelpdeskAgent.

Note: If you do not set the two variables [[ChatUser]].Name and [[ChatUser]].HelpdeskSubject, the chatbot will fail placing the customer in the Amazon Connect live chat. Please note that in figure above, the [[HelpdeskClient]] entity in the right hand side of the SetVariables section are used only as an example - you can set any values in [[ChatUser]].Name and [[ChatUser]].Subject, as per your specific flow configuration.

Send the conversation transcript to Amazon Live Chat Agents

Most contact centers want their helpdesk agents to get the transcript of the conversation the client had with the bot prior to being engaged in a live chat session.

To save the conversation transcript and make it available for Amazon Live Chat agents, on the flow that connects the client with the agent, make the following settings:

On the step where you call the internal action ConnectToHelpdeskAgent(Post Actions section), in the Pre Actions section, call the internal action GetConversationTranscript.

Important! When calling the GetConversationTranscript internal action on the same step where you call ConnectToHelpdeskAgent, make sure to call it in the PreActions section. Calling it in the PostActions section will not send the conversation transcript to agents.

Hint: You can call the GetConversationTranscript internal action on a previous step.

On the ConnectToHelpdeskAgent internal action add the following:

Copy

{
    ”SendConversationTranscript” : true
}

Hint: You can send additional client information Amazon Live Chat Agents together with the conversation transcript. For more information, see Send Contact Center Parameters.

Save the step.

Send Contact Center Parameters

The table below lists the Contact Center parameters bot authors can send to Amazon Live Chat.

Parameter	Data Type	Description
firstName	String	The first name of the client.
lastName	String	The last name of the client.
question	String	The question asked by the client.
queueId	Long	The ID of the queue to which the client will be assigned. Note: If you have multiple queues, this is the Id of the Amazon Live Chat queue to which the client will be assigned.
contactId	Long	The contact ID of the client.
organizationId	Long	The client organization ID.

To send Contact Center parameters Amazon Live Chat, on the ConnectToHelpdeskAgent internal action, provide the desired parameters inside the ContactCenterParameters object in JSON format.

Note: The payload must follow a key-value format where both the key and value are strings. When using values from DRUID conversation context (entities, local variables) as parameter values, ensure they are enclosed in double quotes ("") to maintain valid JSON formatting..

Copy

Example - Send Contact Center parameters

{
     "SendConversationTranscript":true,
     "ContactCenterParameters":
        {
                "firstName":"[[Customer]].FirstName",
                "lastName":"@lastName",
                "question":"[[Customer]].Question",
                "queueId":"[[Entity]].QueueId",
                "contactId":"@contactId",
                "organizationId":""
        }
}