> This page is for version Generative (default).
> For other versions, use one of these documentation indexes:
> - Generative (default): https://developers.ada.cx/generative/llms.txt
> - Scripted: https://developers.ada.cx/scripted/llms.txt

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://developers.ada.cx/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://developers.ada.cx/_mcp/server.

# Zendesk SDK Deployment

You can use Zendesk Messaging SDKs to allow customers to interact with
your AI Agent on their web or mobile devices. There are several other ways to
do this, including using [Ada's SDKs](/generative/docs/channels/chat/widget-embedding#app-experience),
but this method is convenient if you already have an integration with
Zendesk configured.

## Understand customer experience with Zendesk Messaging SDKs\[#understand-customer-experience-with-zendesk-messaging-sdks]

The customer experience through Zendesk Messaging SDKs is very similar to
Ada's default web experience in a lot of ways, with some exceptions.
Here are some basics:

* Customers can use the same conversation window to exchange messages
  with both your AI Agent and with human agents, with seamless handoffs to
  Zendesk Messaging in between.

* Customers and agents can send files to each other. For information on
  how Zendesk validates files, see [Working with attachments](https://support.zendesk.com/hc/en-us/articles/4408821905434#topic_i32_zhh_3nb)
  at Zendesk's documentation.

* After the customer has been handed off to an agent, the chat window is
  reserved for their conversation with the agent. Only the agent can
  disengage the conversation by marking the ticket as "solved." After
  this, depending on your organization's [trigger settings](#configure-settings-in-zendesk-admin-center),
  Zendesk automatically marks the ticket as "closed."

  Until the ticket is marked as "closed," any messages the customer sends
  go to the agent, not the AI Agent. The customer has to wait until the ticket
  is closed to be able to send messages to your AI Agent again.

* You can use Zendesk's multi-instance routing feature. For more information,
  see [Connect Zendesk social channels to multiple AI Agents using Sunshine Conversations](/generative/docs/channels/social/zendesk-social-integration).

### Understand how Ada features work with Zendesk Messaging SDKs\[#understand-how-ada-features-work-with-zendesk-messaging-sdks]

* The language your AI Agent sends messages in depends on the customer's
  browser settings. You can override the default language by setting the
  locale manually. For more information, see the [Set locale](https://developer.zendesk.com/api-reference/widget-messaging/web/core/#set-locale)
  section of Zendesk's Core messaging Web Widget API documentation.

* Some [detailed reports](/generative/docs/monitoring/performance/reports/performance-reports)
  may not contain information about conversations that took place on
  Zendesk channels, depending on the data they require.

* You can manage whether your AI Agent remembers customers between
  conversations in Agent Workspace. You can choose to retain
  conversation history, or to treat all customers as anonymous, with no
  saved previous conversation history. For more information, see the
  instructions in Zendesk's
  [troubleshooting](https://support.zendesk.com/hc/en-us/articles/4531075858458)
  documentation.

  This setting only applies to Zendesk's Web Widget, and not mobile; on
  mobile, Zendesk saves customer history by default.

* You can control which website domains customers can see your Zendesk
  Web Widget on. For more information, see [Creating an allowlist for the Web Widget](https://support.zendesk.com/hc/en-us/articles/4500748175258-Installing-the-Web-Widget#topic_edv_xd5_2tb)
  at Zendesk's documentation.

## Set up Zendesk Messaging SDKs with your AI Agent\[#set-up-zendesk-messaging-sdks-with-your-ai-agent]

Setting up your AI Agent to work with Zendesk Messaging SDKs requires some
work in Sunshine Conversations, in Zendesk, and finally in
your own website. If you have any trouble with any portion of these
procedures, don't hesitate to contact your Ada team for assistance.

### Before you begin\[#before-you-begin]

Before proceeding, do a quick check to see that you have the following:

* Access to Zendesk Messaging (on Agent Workspace)

* A subscription to Zendesk Suite Professional, Enterprise, Enterprise
  Plus, **or** a standalone Sunshine Conversations license add-on

* Access to the Zendesk Conversations API through Zendesk Admin Center,
  so you can create API keys

* A completed Sunshine Conversations integration with your AI Agent. For
  more information, see [Configure and use Sunshine Conversations](/generative/docs/channels/social/sunshine-conversations-setup).

* Your AI Agent's name configured in your Ada dashboard, so the AI Agent's name
  appears on messages in your widget and in Zendesk tickets. For more
  information, see [Customize your AI Agent's general settings](/generative/docs/channels/chat/appearance-settings#general).

Additionally, contact your Ada team to make sure your AI Agent is ready for
the next steps. There are some features you may need Ada staff to enable
on your behalf that you can't proceed without.

### Configure your Zendesk widget\[#configure-your-zendesk-widget]

You can customize your web or mobile widgets in Zendesk Admin Center:

* For information on how to customize how your widget looks, see
  [Configuring the widget launcher](https://support.zendesk.com/hc/en-us/articles/4500747797914-Configuring-the-appearance-of-the-Web-Widget#id_jnl_xtm_gnb)
  at Zendesk's documentation.

* For more detailed information on additional settings you can change,
  see [Zendesk Web Widget and SDKs](https://developer.zendesk.com/api-reference/widget-messaging/introduction/)
  at Zendesk Developers' documentation.

### Install your Zendesk widget on your website\[#install-your-zendesk-widget-on-your-website]

After configuring your Zendesk widget, you can install it on your
website so customers can interact with it.

* To install your widget on your website, find the installation script
  in your Admin Center and add it to your website's code.

* Optionally, you can choose to enable your widget on your Zendesk Help
  Center.

For installation instructions, see [Installing the Web Widget](https://support.zendesk.com/hc/en-us/articles/4500748175258-Installing-the-Web-Widget)
at Zendesk's documentation.

### Configure settings in Zendesk Admin Center\[#configure-settings-in-zendesk-admin-center]

In Zendesk Admin Center, you can configure two settings for your AI Agent's
behavior:

* You can set up triggers to automatically close tickets a certain
  amount of time after the agent marks them as "solved." For more
  information, see [Understanding and managing handback events](https://support.zendesk.com/hc/en-us/articles/4408824482586#topic_ux4_cqq_5pb)
  at Zendesk's documentation.

* To collect customer satisfaction data, you can set up Zendesk's CSAT
  survey. For more information, see [About CSAT ratings in messaging](https://support.zendesk.com/hc/en-us/articles/4408832620570-About-CSAT-ratings-in-messaging)
  at Zendesk's documentation.

Lastly, to allow your AI Agent to start interacting with customers, you have
to connect your AI Agent with Zendesk. For more information, see [Connect your AI Agent with Zendesk Admin Center](/generative/docs/handoffs/zendesk/zendesk-messaging#connect-your-ai-agent-with-zendesk-admin-center).

### Set up handoffs using HTTP Request blocks\[#set-up-handoffs-using-http-request-blocks]

Instead of using Ada's usual handoff blocks, Zendesk Messaging SDKs use
API requests to hand customers off to human agents.

> **Note**
>
> When configuring handoffs using the HTTP Request block, make sure you
> toggle on the **Count as Handoff** setting on the block to collect data
> for your reports.

For complete documentation on Sunshine Conversations' API, see [Metadata sent to Zendesk](https://docs.smooch.io/guide/switchboard/#metadata-sent-to-zendesk)
at Sunshine Conversations' developer documentation. To get started,
however, you can use these values using the variables with your AI Agent.

<table>
  <thead>
    <tr>
      <th>
        Name
      </th>

      <th>
        Value
      </th>

      <th>
        Notes
      </th>
    </tr>
  </thead>

  <tbody>
    <tr>
      <td>
        URL
      </td>

      <td>
        `https://api.smooch.io/v2/apps/{appId}/
        conversations/{conversationId}/passControl`
      </td>

      <td />
    </tr>

    <tr>
      <td>
        appId
      </td>

      <td>
        `<sunco-app-id>`
      </td>

      <td>
        * Sent in URL
        * Found in **Zendesk Admin Center** >
          **APIs** > **Conversations API**
      </td>
    </tr>

    <tr>
      <td>
        conversationId
      </td>

      <td>
        `@sunshine_conversation_id`
      </td>

      <td>
        Sent in URL
      </td>
    </tr>

    <tr>
      <td>
        Authorization
      </td>

      <td>
        `Basic @sunshine_basic_token`
      </td>

      <td>
        Sent in header
      </td>
    </tr>

    <tr>
      <td>
        switchboardIntegration
      </td>

      <td>
        `“next”`
      </td>

      <td>
        Sent in body
      </td>
    </tr>

    <tr>
      <td>
        metadata:first_message_id
      </td>

      <td>
        `@sunshine_first_message_id`
      </td>

      <td>
        * Sent in body
        * Sends first message of the most recent interaction, overriding
          default behavior of sending just the last 10 messages in the
          conversation
      </td>
    </tr>
  </tbody>
</table>

Here's an example of a completed handoff using an HTTP Request block:

![](/_fern-files/_img/64099a2bbb92d868ac5560daad78b95a67a6ea977bec5cd8243fd33d7990411b.webp)

In the above example, the request block also passes tags into Zendesk.
This is how they look in the human agent's conversation with the
customer:

![](/_fern-files/_img/f081de884c068b43c0dbcb7dcaa107743acda7318b127d48c76333257cb8635c.webp)

---

Have any questions? Contact your Ada team, or email us at [](<mailto:help@ada.support%20%E2%80%9CHelp Docs inquiry%E2%80%9D>).