# AI Assistant

In BoltAI, an AI Assistant includes a set of configuration that help you with specific tasks: Software Developer, Technical Writer, Marketing Expert... 

I'm working to migrate toward the official Assistants API from OpenAI.

### How to create an AI Assistant?

Creating a new AI Assistant is simple. Go to **Settings > AI Assistant**, click on the "+" button.

Fill the form:

1. Give your Assistant a friendly name. For example: "Life Coach"
2. Set the System Instruction for the Assistant.
3. (Optional) Set a custom ID for the Assistant. Note that this will change the **Trigger Phrase** for this Assistant. For the example below, you will need to type `//ask_life_coach`.
4. (Optional) Set other advanced configurations such as AI Service & Model (only applicable for inline usage), Temperature or Output Language
5. Click `Save`

<figure><img src="/files/9I7wzRULjf1S6zcwRyky" alt=""><figcaption></figcaption></figure>

### How to use an AI Assistant?

You can select the assistant when starting a new chat (1). Clicking on the info button to view the detail information, and optionally set it as the default assistant (2)

<figure><img src="/files/dQCHjIPC8098746AkiOM" alt=""><figcaption><p>Choose an AI Assistant</p></figcaption></figure>

You can change the AI Assistant mid-conversation with the Statusbar. You can toggle the Statusbar in the "View" menu item, or with the keyboard shortcut `Command+/`

<figure><img src="/files/gIdbJRvyl0kywlZwYC7V" alt="" width="375"><figcaption></figcaption></figure>

For inline usage, follow this guide

{% content-ref url="/pages/B4FQ658yCMwrFc7jIm3i" %}
[Inline Assistant](/docs/ai-inline/inline-assistant.md)
{% endcontent-ref %}


---

# Agent Instructions: Querying This Documentation

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

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.boltai.com/docs/chat-ui/ai-assistant.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.