> For the complete documentation index, see [llms.txt](https://docs.avis.net/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.avis.net/avis-creative-studio/agent-and-integrations.md).

# Agent & Integrations

**What is Agent?**

Agent (Studio Agent, in beta) is the AI assistant built into Avis Creative Studio. It helps you brainstorm ideas, write scripts and prompts, and create images and videos on your Canvas. You can use it on the Canvas or in 3D Staging. The Agent runs on AI models provided by Avis, including models from Claude and ChatGPT.

<figure><img src="https://3712251980-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fd1iOtdhEqFbDy8JzwAnH%2Fuploads%2FSrC5Hl1fzJCr3TDxU1HE%2Fimage.png?alt=media&amp;token=c9c9a5be-7996-43cf-9f90-9846b32cb54d" alt="" width="359"><figcaption><p>Studio Agent panel</p></figcaption></figure>

**The Agent panel**

The Agent panel docks beside the Canvas, so you can chat while you work. At the top of the panel:

* **Menu (☰):** Opens your earlier conversations. Conversations are saved for each Workspace and synced with your account.
* **New chat (+):** Starts a new conversation.
* **Expand:** Opens the chat in a larger view.
* **Panel:** Shows or hides the Agent panel.

Your chat is private: other members of the project cannot see it. Chat data may be used to improve Studio Agent.

**Start a chat**

Type your request in the chat box, or start from a suggestion:

* **Preset:** Example projects you can start from, such as an ad or an app UI. Click the refresh button to see more.
* **Skill:** Quick starts for common tasks: **Start from an idea**, **Draft a script**, **Generate multi-view character** and **Polish prompt**.

Type **@** in the chat box to use an image from your Workspace.

<figure><img src="https://3712251980-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fd1iOtdhEqFbDy8JzwAnH%2Fuploads%2Fdr0FofcIxxpniJnY9bIV%2Fimage.png?alt=media&amp;token=c7f39621-cea5-476e-add9-32bce59b0ef4" alt="" width="356"><figcaption><p>Start from a Preset or a Skill</p></figcaption></figure>

**Chat box tools**

* **Add (+):** Adds media or files to the chat.
* **Model:** Chooses the AI model the Agent uses.
* **Skills:** Opens the Skills menu (see **Skills**).
* **Settings:** Sets the Agent mode and how the Agent uses credits (see **Agent mode and credits**).
* **Microphone:** Lets you speak your request instead of typing (see **Voice input**).

**Agent mode and credits**

Click **Settings** in the chat box:

* **Agent mode**
  * **Brainstorm:** The Agent helps you think, plan and write. It never generates media, so there is nothing to confirm.
  * **Create:** The Agent can also generate images, videos and audio on your Canvas.
* **Credits autonomy:** In **Create** mode, turn it on to let the Agent generate without asking you again for the rest of this session. Leave it off to confirm each paid generation.
* **Subagent model:** The model the Agent uses for smaller helper tasks. By default, it is the same as the chat's model.

<figure><img src="https://3712251980-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fd1iOtdhEqFbDy8JzwAnH%2Fuploads%2FRFqryseQQOJenK72BaLc%2Fimage.png?alt=media&amp;token=d143304d-35a6-4eac-8ded-17b2ef257f99" alt="" width="371"><figcaption><p>Agent mode and credits</p></figcaption></figure>

**Skills**

Skills give the Agent extra know-how for a task. Click **Skills** in the chat box:

* **Suggestion:** Skills suggested for you.
* **My skills:** Skills you have added. Each one shows whether it is on or off.

Search for a skill by name, or click **View all** to see and manage all your skills.

<figure><img src="https://3712251980-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fd1iOtdhEqFbDy8JzwAnH%2Fuploads%2F4wPMjIUI7SPTfyh4sxWc%2Fimage.png?alt=media&amp;token=5956257b-fe6d-4dac-8bed-516079049331" alt="" width="428"><figcaption><p>Skills menu</p></figcaption></figure>

**Chat tips**

* **No fixed turn limit.** The Agent keeps working until the task is done. You can cancel at any time.
* **Message queue.** You can type your next message while the Agent is still working. It is sent when the Agent finishes.
* **Copy.** You can copy the Agent's replies. Replies are shown in Markdown.

**Voice input**

Instead of typing, you can speak your request to the Agent.

1. Click the **microphone** button in the chat box and start speaking.
2. Click it again to stop. Your speech is turned into text in the chat box.
3. Check the text, then send it.

**Limits and privacy**

* Each voice message can be up to **60 seconds** long.
* Your voice is processed **locally on your computer**.


---

# 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.avis.net/avis-creative-studio/agent-and-integrations.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.
