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

# Elephas Key Concepts: Brains, Documents, Chats and Citations

> A short tour of the ideas behind Elephas: workspaces, Brains, documents, Ask, chats, citations and AI options. Start here to see how the app fits together.

Elephas is built around a few simple ideas. You keep your documents in **Brains**, you **Ask** questions across the Brains you choose, and every answer comes with **citations** that show the exact passage from your files.

## The building blocks

### Workspace

A workspace holds all your Brains and chats. Everyone starts with a **Personal workspace** on their own computer. If you join a team, you also get a **Team workspace** with shared Brains. Switch between them with the **Workspace** switcher in the sidebar.

[Teams overview](/docs/teams/teams-overview)

### Brain

A Brain is a collection of documents about one subject: a client, a case, a deal, a research project or a course. Brains keep unrelated material apart, so an answer about one client never draws on another client's files.

Your first Brain is created for you during setup and called **My Documents**. Create more with **New Brain** (⌘B on Mac, Ctrl+B on Windows). You can pin up to 5 Brains to the top of the sidebar.

[Create and manage Brains](/docs/brains-and-documents/create-and-manage-brains)

### Documents

Documents are the files and web pages you add to a Brain: PDFs, Word, PowerPoint, Excel, Markdown and plain text, CSV, RTF, HTML, EPUB, and images. Elephas reads text in scanned pages and images on your computer.

When you add a document, Elephas extracts its text and indexes it. Indexing happens on your computer. Always. If you replace a document with a newer version, earlier answers keep citing the version they used.

[Add files and folders](/docs/brains-and-documents/add-files-and-folders) · [Supported file types](/docs/brains-and-documents/supported-file-types)

### Ask

Ask is where you ask questions. Choose which Brains to ask, or narrow it to specific documents, then type your question. Elephas finds the passages in your documents that match, and writes an answer grounded in them.

Ask only uses the Brains you select. It never browses the web, and it never pulls from other Brains. If your documents do not support an answer, Elephas tells you so instead of guessing. That is the difference from a generic chatbot: answers from your documents, not the internet.

[Ask questions about your documents](/docs/ask/ask-questions-about-your-documents) · [Ask specific documents](/docs/ask/ask-specific-documents)

### Citations

Every answer has small numbered citations. Click one to see the exact passage from your document in the Sources panel.

[Citations and sources](/docs/ask/citations-and-sources)

### Chats

Each conversation in Ask is a chat. Chats are saved in your workspace so you can come back to them, rename them or delete them. Start a new one with **New Chat** (⌘N on Mac, Ctrl+N on Windows).

[Manage chat history](/docs/ask/manage-chat-history)

## Where answers come from

### AI options

Elephas uses an AI model to write each answer from your passages. You choose where that model runs in **Settings › AI**:

* **Elephas AI** is included with your plan and works with no setup.
* **Own API key** sends requests directly to the AI provider you already use, with your key.
* **Local AI** runs a model on your computer, so your question and passages are not sent anywhere.

[Choose how Elephas answers](/docs/ai-options/choose-how-elephas-answers)

### Personal information protection

When you use a cloud option, Elephas detects names, contact details and ID numbers in your question and passages, and replaces them with placeholders on your computer before anything is sent. The real values are put back when the answer appears. It is on by default and hides many common personal details. You can also add names to a Brain that should always be hidden.

[Personal information protection](/docs/privacy-and-security/personal-information-protection) · [What data is sent to AI](/docs/privacy-and-security/what-data-is-sent-to-ai) · [Protected names](/docs/brains-and-documents/protected-names)

## Across devices and people

### Personal Sync

Personal Sync keeps your library in step across your own computers, end-to-end encrypted. It is off until you turn it on, and is available on the Professional plan.

[Personal Sync overview](/docs/sync-and-devices/personal-sync-overview)

### Team workspace

A Team workspace lets colleagues share Team Brains, so everyone asks the same files and gets the same answers. The team owner manages members, roles and which Brains each person can see.

[Teams overview](/docs/teams/teams-overview)

## Getting around

| Go to | Mac | Windows |
| - | - | - |
| Home | ⌘1 | Ctrl+1 |
| Ask | ⌘2 | Ctrl+2 |
| Brains | ⌘3 | Ctrl+3 |

**Home** shows what you last asked, anything that needs your attention, and where you left off.

## Related articles

<CardGroup cols={2}>
  <Card title="Quick start" href="/docs/getting-started/quick-start" />

  <Card title="Glossary" href="/docs/resources/glossary" />

  <Card title="Tips for better answers" href="/docs/ask/tips-for-better-answers" />

  <Card title="Privacy overview" href="/docs/privacy-and-security/privacy-overview" />
</CardGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.