1. Overview

Integrate Fullview AI into your Zendesk Ultimate workflows and enable Fullview AI to provide real-time, in-product assistance on your customer's screen - directly within your support flows.

To ensure seamless operation, you'll need to create a Conversation API Key configure specific triggers and data flows within your Zendesk workspace. These connections allow Fullview AI to respond intelligently at key points in the workflow—whether it's initiating a session, offering contextual help, collecting feedback, or escalating to a live agent.

Prerequisites

Fullview app installed from Zendesk Marketplace
Zendesk Ultimate Suites
Fullview AI enabled for your Fullview organization

2. Create a Conversations API Key

As a Zendesk admin you first have to generate an API key which allows Fullview to send messages in conversations. Navigate to:

Zendesk Admin Center → Conversations API → click Create API Key

❗️Securely save the App Id, Key Id and the Secret key for the following steps.

3. Webhook for escalations

Create a webhook to call Fullview's API when conversations are escalated to an agent. This webhook sends a note summarizing customer interactions. Creation steps:

Zendesk Admin Center → Webhooks → Create webhook
Choose Trigger and Automation as the type

Set the endpoint URL - depending on the region your Fullview org is in:

EU1

https://api.eu1.fullview.io/access/api/integrations/ai/unsuccessful-resolution

US1

https://api.us1.fullview.io/access/api/integrations/ai/unsuccessful-resolution

EU2

https://api.eu2.fullview.io/access/api/integrations/ai/unsuccessful-resolution

Method: POST
Request format: JSON
Authentication: None
Complete the webhook details and Save

4. Create a trigger

Zendesk Admin Center → Triggers → Create trigger
Name (eg): Ticket escalation via Fullview AI
Trigger category: Notifications (created the category if you don't have a similar one)
Condition 1 - under "Meet ALL of the following conditions"
1. Category: Ticket > Ticket
2. Operator: Is
3. Value: Created
Condition 2
1. Category: Ticket > Tags
2. Operator: Contains at least one of...
3. Value: fullview-ai-agent-escalation
Add an action action to call the webhook when conditions are met (scroll down)
1. Category: Notify by > Active webhook
2. Value: Escalation Fullview Notification
c. Down below you see the content you should add to the body of the POST

{
  "externalId": "{{ticket.requester.external_id}}",
  "email": "ticket.requester.email",
  "organisationId": "--replace-with-your-org--",
  "integrationData": {
    "platform": "zendesk",
    "caseId": "{{ticket.id}}"
  }
}

5. Zendesk Ultimate integration

The next step is to create 5 API integrations under your Zendesk Ultimate dashboard.

Zendesk Admin Center → AI → AI Agents → Manage AI agents - Advanced → API integrations (from the navigation) → Add integration

1. Start AI Session

This will be used to send Fullview AI the customer intent and start the assistance

Integration name: Start AI Session
Under the Request parameters, create the following 4 keys of type string:
- lastVisitorMessage
- email
- externalId
- conversationId

Under Environments

Method type: POST
URL - depending on the region your Fullview org is in:
- EU - https://api.eu1.fullview.io/access/api/integrations/ai/start
- US - https://api.us1.fullview.io/access/api/integrations/ai/start
- EU2 - https://api.eu2.fullview.io/access/api/integrations/ai/start

Request Body:

{
  "organisationId": "YOUR FULLVIEW ORG ID",
  "email": "{{email}}",
  "externalId": "{{externalId}}",
  "query": "{{lastVisitorMessage}}",
  "integrationData": {
    "caseId": "{{conversationId}}",
    "customerExternalId": "{{externalId}}",
    "subdomain": "YOUR ZENDESK'S ORG SUBDOMAIN",
    "platform": "zendesk"
  }
}

2. Terminate AI Session

Used when the customer wants to end the Fullview AI assistance

Integration name: Terminate AI Session
Under the Request parameters, create the following 2 keys of type string:
- email
- externalId
Under Environments
- Method type: POST
- URL - depending on the region your Fullview org is in:
  - EU - https://api.eu1.fullview.io/access/api/integrations/ai/abort
  - US - https://api.us1.fullview.io/access/api/integrations/ai/abort
  - EU2 - https://api.eu2.fullview.io/access/api/integrations/ai/abort
- Request Body:
```
{
  "organisationId": "YOUR FULLVIEW ORG ID",
  "email": "{{email}}",
  "externalId": "{{externalId}}",
  "integrationData": {
    "platform": "zendesk"
  }
}
```

3. Send Positive Feedback

Used for sending Fullview AI the positive customer feedback, when the session ended successfully

Integration name: Send Positive Feedback
Under the Request parameters, create the following 3 keys of type string:
- email
- externalId
- conversationId
Under Environments
- Method type: POST
- URL - depending on the region your Fullview org is in:
  - EU - https://api.eu1.fullview.io/ai-agent/integrations-sessions-feedback/positive-feedback
  - US - https://api.us1.fullview.io/ai-agent/integrations-sessions-feedback/positive-feedback
  - EU2 - https://api.eu2.fullview.io/ai-agent/integrations-sessions-feedback/positive-feedback
- Request Body:
```
{
  "organisationId": "YOUR FULLVIEW ORG ID",
  "email": "{{email}}",
  "externalId": "{{externalId}}",
  "caseId": "{{conversationId}}"
}
```

4. Send Negative Feedback

Used for sending Fullview AI the negative customer feedback, when the session ended unsuccessfully and the customer rated the interaction with a thumbs down

Integration name: Send Negative Feedback
Under the Request parameters, create the following 3 keys of type string:
- email
- externalId
- conversationId
Under Environments
- Method type: POST
- URL - depending on the region your Fullview org is in:
  - EU - https://api.eu1.fullview.io/ai-agent/integrations-sessions-feedback/negative-feedback
  - US - https://api.us1.fullview.io/ai-agent/integrations-sessions-feedback/negative-feedback
  - EU2 - https://api.eu2.fullview.io/ai-agent/integrations-sessions-feedback/negative-feedback
- Request Body - same as at the previous connector:
```
{
  "organisationId": "YOUR FULLVIEW ORG ID",
  "email": "{{email}}",
  "externalId": "{{externalId}}",
  "caseId": "{{conversationId}}"
}
```

5. Send Negative Feedback Message

Used for sending Fullview AI the additional feedback from the customer, when the session ended and the customer gave additional feedback besides the thumbs down

Integration name: Send Negative Feedback
Under the Request parameters, create the following 4 keys of type string:
- email
- message
- externalId
- conversationId
Under Environments
- Method type: POST
- URL - depending on the region your Fullview org is in:
  - EU - https://api.eu1.fullview.io/ai-agent/integrations-sessions-feedback/negative-feedback-details
  - US - https://api.us1.fullview.io/ai-agent/integrations-sessions-feedback/negative-feedback-details
  - EU2 - https://api.eu2.fullview.io/ai-agent/integrations-sessions-feedback/negative-feedback-details
- Request Body - same as at the previous connector:
```
{
  "organisationId": "YOUR FULLVIEW ORG ID",
  "email": "{{email}}",
  "externalId": "{{externalId}}",
  "caseId": "{{conversationId}}",
  "message": "{{message}}"
}
```

Here is an example of how the configuration on an API integration should look like:

6. Create Actions

Now let's create the following 3 actions. To do that navigate to:

Zendesk Admin Center → AI → AI Agents → Manage AI agents - Advanced → Content → Actions → Create action

1. Get Conversation Data

This action will be used to get the conversationId when a support request is submitted.

Name: Get Conversation Data
Target: Sunshine Conversations
Task: Get conversations
Field to retrieve: id
Save as parameter: conversationId

2. Get User Data

This action will be used to get the email and the externalId from the user that started the conversation.

Name: Get Conversation Data
Target: Sunshine Conversations
Task: Get user
Field to retrieve: email
Save as parameter: email
Field to retrieve: externalId
Save as parameter: externalId

3. Escalate via Fullview AI

This action will be used to add a tag on the ticket that will be created on escalation. The tag should be called fullview-ai-agent-escalation

Name: Escalate via Fullview AI
Target: Sunshine Conversations
Task: Add conversation tags
Tags: fullview-ai-agent-escalation

7. Integrate Fullview AI in your workflows

Now that you have all the building components in place, you can now leverage the API integrations and Actions you created above, to integrate Fullview AI in your existing Zendesk Ultimate workflows.

To do so navigate to:

Zendesk Admin Center → AI Agents → Manage AI agents - Advanced → Content

7.1 Example

Here is an example of a template we created, which is used to start a Fullview AI session.

The example flow above captures the last visitor message and runs two actions: Get Conversation Data and Get User Data - to provide context. A dedicated use case links to the flow template and initiates the AI Session, enabling a contextual and responsive support experience.

When a Fullview AI session ends, we send a message in the conversation that triggers the feedback flow. The use case for feedback looks like this:

Here's an example of the entire feedback flow

Fullview AI <> Zendesk workflow