> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mora.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Using the chat

> Open the chat panel, send messages, switch modes, and manage threads.

## Opening the chat

Press **Cmd+/** to toggle the chat panel open or closed. The panel slides in from the side of the screen.

The chat is available on the home page, dataset pages, and dashboard pages.

## Sending a message

Type your message in the input field and press **Enter** to send. The AI will respond and may take actions (like editing SQL or creating a chart) depending on the mode and your request.

## Chat modes

Switch between modes using the mode selector at the top of the chat, or press **Shift+Tab** to cycle through them.

| Mode          | What it does                                                                                                                                                    |
| ------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Auto**      | Lets the AI decide which tools to use based on your message. Best for general use.                                                                              |
| **SQL**       | Focuses on SQL editing, query debugging, and schema exploration. Dashboard actions are only used if you explicitly ask.                                         |
| **Dashboard** | Focuses on dashboard, chart, and filter management. SQL actions are only used if you explicitly ask.                                                            |
| **Ask**       | Read-only. The AI can answer questions, explain queries, run queries to inspect data, and browse the catalog, but it cannot create, update, or delete anything. |

## Threads

Each conversation is a thread with history. The AI remembers context from earlier in the thread, so you can build on previous messages.

## Context awareness

The AI automatically knows:

* **Which page you're on**: dataset or dashboard.
* **The current query**: if you're on a dataset page.
* **Your datasource type**: so it writes SQL in the correct dialect.
* **Your catalog**: it can search for tables and columns.

## Images

You can paste or upload images into the chat. This is useful for sharing:

* Screenshots of errors.
* Schema diagrams or ERDs.
* Charts you want to reproduce.

## Related

* [AI overview](/ai/overview)
* [What the AI can do](/ai/what-it-can-do)
* [Tips](/ai/tips)