> For the complete documentation index, see [llms.txt](https://atlas.anywhere.now/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://atlas.anywhere.now/channels-integrations/dialogue-studio-integrations/integrate-dialogue-ai-assist-with-dialogue-studio.md).

# Integrate Dialogue AI Assist with Dialogue Studio

Connect an environment-specific AI Assist integration to Dialogue Studio by sending the event and transcript data the integration needs.

Use this page to connect Dialogue AI Assist to Dialogue Studio for voice scenarios where the flow must forward event and transcript data to the assistant integration.

{% hint style="warning" %}
Use the Dialogue AI Assist flow package, transcription prerequisite, and credential model approved for the customer environment before changing a production flow.
{% endhint %}

## Before you begin

* Dialogue AI Assist is approved and configured for the UCC.
* Dialogue Studio is available for the UCC.
* The required voice transcription capability is available, if the environment depends on transcription for Dialogue AI Assist.
* The Dialogue AI Assist webhook URL and secret values are supplied through the approved delivery or support process.
* You know who owns endpoint, credential, and operational support for the integration.
* You know which dialogue events and transcript data the flow should forward.
* You have a fallback path for integration failures or unusable responses.
* You can validate the integration in a controlled test flow before production exposure.

{% hint style="warning" %}
Event-driven integrations can fail because of authentication, endpoint availability, or payload-shape mismatches. Always validate the connection ownership and the fallback path before you rely on this integration in production.
{% endhint %}

## Import or update the flow pattern

Some environments provide an example flow for Dialogue AI Assist nodes. Use the example only when it is available and approved for the customer environment.

| Scenario                                      | Setup approach                                                                                                                             |
| --------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------ |
| Existing flow already uses an InDialogue node | Import the example into the existing flow, then update the existing flow so its connections match the approved Dialogue AI Assist pattern. |
| New setup                                     | Import the example as a new flow and select the target UCC in the InDialogue and EventBus nodes.                                           |

When product or support supplies the standard example flow, use the supplied location. The location can be:

```
Examples > flows > anywherenow-dialogue-ai-assist-nodes > indialogue-flow
```

{% hint style="info" %}
If the example path, node names, or node availability differ in the Dialogue Studio environment, use the flow package supplied for that environment.
{% endhint %}

## Configure one owned connection configuration

Keep the connection setup centralized so the flow reuses one owned set of integration values.

This setup pattern uses a shared Dialogue AI Config when that node label is present in the Dialogue Studio environment.

1. Store the required endpoint or connection target once.
2. Store the required credential or authentication value once.
3. Reuse that configuration from each related integration step in the same flow.
4. Confirm who owns renewal or rotation of the authentication value.

In the Dialogue AI Assist Events node, create or select the shared configuration and enter the webhook URL and secret value supplied for the environment. In each Dialogue AI Assist Transcript node, select the same shared configuration.

{% hint style="warning" %}
Treat webhook secrets and client secrets as managed secrets. Do not expose secret values in screenshots, flow notes, issue tickets, or unmanaged documentation.
{% endhint %}

## Forward dialogue events deliberately

Use the event-focused integration path when Dialogue AI Assist should react to selected dialogue events.

Before production exposure, verify:

* which event types matter to the integration
* how each event is scoped to the intended UCC before forwarding
* which event object or session-state values the next step must pass forward
* whether the event branch should also trigger transcript handling
* what the flow should do when the event-forwarding call fails

{% hint style="warning" %}
Selecting the target UCC in the EventBus connection does not filter generic bus payloads to that UCC. Events can include other UCCs on the underlying Core service, not only UCCs connected to this Dialogue Studio deployment, even with a dedicated plugin port. Filter on event-specific UCC or dialogue metadata where present before forwarding. Not every event has a UCC field; use validated correlation where available and handle unresolved events explicitly instead of forwarding them as if they belonged to the selected UCC. See [EventBus event scope](/contact-center-configuration/building-your-callflow/via-dialogue-studio/core-nodes.md#eventbus-event-scope).
{% endhint %}

## Send transcript data in a consistent shape

Use the transcript-focused integration path when Dialogue AI Assist depends on transcript content from the conversation.

Some flow patterns use two Dialogue AI Assist Transcript nodes that use the same shared Dialogue AI Config as the event node. Follow the node labels and required transcript timing shown in the customer environment.

Keep the transcript pattern simple:

* map the incoming transcript format into the supported payload shape
* include only the stable transcript fields the integration requires, such as session ID, transcript ID, participant URI, language, final flag, and transcript text
* send the normalized payload through the owned connection configuration

## Keep debugging and status checks operational

Before production exposure, validate:

* how the integration signals success or failure
* how the flow should recover from a failed integration call
* which payload details are useful for troubleshooting and which are only low-level debugging noise
* whether repeated failures should stop the branch or route to a fallback path

## Verification

1. Confirm the connection values point to the intended integration environment.
2. Run one controlled event-driven test and confirm the intended event data is forwarded.
3. Run one transcript-driven test and confirm the transcript payload reaches the integration in the expected shape.
4. Force one authentication, endpoint, or payload failure and confirm the fallback route works as designed.
5. Confirm the integration owner has a documented process for rotating credentials and investigating repeated failures.
6. Where another UCC is available on the same Core service, confirm its generic bus events are not forwarded to this UCC's assistant integration. Check the handling of events whose UCC cannot be resolved.

{% hint style="info" %}
Keep this page focused on event type, event object, shared connection configuration, and the small stable transcript field set. Leave raw debugging payloads, assistant API configuration, tool configuration, and JSON examples out of this setup guidance.
{% endhint %}

## Related topics

Go to **Configure Dialogue AI Assist** for the broader WebAgent, configuration, and validation sequence.

{% content-ref url="/pages/ujDYWWgUDposUaNAQvXs" %}
[Configure Dialogue AI Assist](/channels-integrations/dialogue-ai-assist/configure-dialogue-ai-assist.md)
{% endcontent-ref %}

Go to **Integrate external services with Dialogue Studio** for the broader credential, request, response, and fallback pattern used by external integrations.

{% content-ref url="/pages/HOLQTTPFTbI5AMesxcMD" %}
[Integrate external services with Dialogue Studio](/channels-integrations/dialogue-studio-integrations/integrate-external-services-with-dialogue-studio.md)
{% endcontent-ref %}

Go to **AI nodes** for the broader admin guidance on AI-driven flow behavior, routing, and fallback design.

{% content-ref url="/spaces/gIhRKIrtrwqPt8cNlAy6/pages/GlCSD1yum3Por3TyDayO" %}
[Use AI-driven nodes in Dialogue Studio](/contact-center-configuration/building-your-callflow/via-dialogue-studio/ai-nodes.md)
{% endcontent-ref %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://atlas.anywhere.now/channels-integrations/dialogue-studio-integrations/integrate-dialogue-ai-assist-with-dialogue-studio.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.