> For the complete documentation index, see [llms.txt](https://docs.dapta.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.dapta.ai/ai-text-agents/setting-up-your-text-agent/conversation-behavior.md).

# Conversation behavior

Set your Text Agent's inactivity timeout, reply delay, split replies and WhatsApp typing indicator. Check follow-up timing before publishing changes.

Use **Settings → Conversation behavior** to control when an inactive conversation closes and how your Text Agent delivers replies. These controls are separate from the instructions in your prompt.

<figure><img src="https://3835013762-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FCy5rSNtQmtqYCGzJlNEB%2Fuploads%2Fgit-blob-e1c3387b3d54e25f0d47227668c4d0eac364aca0%2Ftext-agents-settings-01-fields.png?alt=media" alt=""><figcaption><p>The agent settings: language, name, purpose, company context, model and temperature.</p></figcaption></figure>

## Set the inactivity timeout

In **Close after no reply for**, enter the **Hours** and **Minutes** your agent should wait without a customer reply before closing the conversation.

* The default is **24 hours**.
* The total must be between **10 minutes and 24 hours**.
* The minutes field accepts **0–59**. At **24 hours**, the minutes value is **00**.
* After a conversation closes, the customer's next message starts a new conversation.

For example, enter **1 Hour** and **30 Minutes** for a 90-minute timeout. The minimum is **0 Hours** and **10 Minutes**.

### Keep follow-ups within the timeout

Your timeout cannot be shorter than the longest configured follow-up delay. If a follow-up is set for 120 minutes, a 60-minute timeout prevents saving until you shorten or remove that follow-up, or increase the timeout.

The maximum follow-up delay is also capped by this timeout. Schedule reminders comfortably before closure; an accepted delay is not a guarantee of delivery after the conversation closes.

{% hint style="info" %}
The agent's inactivity timeout and WhatsApp's messaging window are different constraints. Increasing the timeout does not extend permission to send WhatsApp messages. Follow-ups still need to satisfy channel and conversation rules.
{% endhint %}

For message types, quiet-hour windows, timer resets and cancellation behavior, see [Follow-ups](/ai-text-agents/setting-up-your-text-agent/follow-ups.md).

***

## Choose how replies are delivered

| Control                                    | What it changes                                                                                         |
| ------------------------------------------ | ------------------------------------------------------------------------------------------------------- |
| **Wait a moment before replying**          | Adds a short pause before replying. Longer responses wait longer. It does not set a follow-up schedule. |
| **Reply in several short messages**        | Lets the agent deliver its answer as shorter messages rather than one long message.                     |
| **Show a typing indicator while replying** | Shows a typing indicator while the reply is prepared. Available only on WhatsApp.                       |

{% hint style="warning" %}
The WhatsApp typing indicator also marks the contact's unread messages as read, including earlier messages. Review this setting before enabling it if your team relies on unread messages to track work.
{% endhint %}

New Text Agents created through the creation wizard or Dapti enable split replies and the typing indicator by default. Existing agents keep their saved settings; the change does not automatically switch those controls on for them. Check the switches on the specific agent you are editing.

***

## Save, test and publish

1. Set the timeout and reply controls.
2. Click **Save Agent** to save the draft. Resolve any timeout or follow-up validation message before continuing.
3. Open **Playground**, reset the test conversation and check the saved draft's responses.
4. When the draft is ready, click **Publish** and confirm to make the changes live.
5. Use a controlled conversation on the connected WhatsApp line to verify channel-specific delivery and typing behavior. Playground alone does not verify WhatsApp read receipts or delivery.

Saving a draft does not publish it. Avoid testing real customer workflows or external actions without appropriate test contacts and credentials.

## Troubleshooting

| Symptom                               | What to check                                                                                               |
| ------------------------------------- | ----------------------------------------------------------------------------------------------------------- |
| The timeout cannot be saved           | The total must be 10–1440 minutes, and cannot be shorter than a configured follow-up delay.                 |
| Minutes return to 00 at 24 hours      | The total is capped at 24 hours; extra minutes would exceed the maximum.                                    |
| Customers still see the old behavior  | Confirm the draft was saved and published, and test a fresh conversation on the intended agent and channel. |
| No typing indicator in Playground     | The indicator is a WhatsApp feature; verify it on the connected line.                                       |
| Earlier WhatsApp messages became read | This is a consequence of the typing indicator, not a separate conversation assignment.                      |

## Related guides

{% content-ref url="/pages/6oDVCzyHD3gYg9lsWKDJ" %}
[Playground: Testing your Text AI Agents](/ai-text-agents/setting-up-your-text-agent/playground-testing-your-text-ai-agents.md)
{% endcontent-ref %}

{% content-ref url="/pages/UTZwUa601OoA7hZESXMb" %}
[Follow-ups](/ai-text-agents/setting-up-your-text-agent/follow-ups.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://docs.dapta.ai/ai-text-agents/setting-up-your-text-agent/conversation-behavior.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.