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

# Core concepts

> Docs, folders, workspaces, and how Cleve uses your context

A few ideas explain most of how Cleve works.

## Docs

A **doc** is a page in your workspace: a draft, a brief, meeting notes, research, a plan. The first line of a doc is its title. Everything else is the body, which can hold text, lists, tables, images, diagrams, and interactive blocks.

Docs save and sync as you type, so there is no save button. Every doc keeps a [version history](https://docs.cleve.ai/features/version-history.md) so you can compare versions and restore an earlier one.

> [!TIP]
>
> You may still see "notes" in older places, such as the API and MCP tool names. Notes and docs are the same thing.

## Folders

Folders group docs, and can be nested inside each other. Drag docs and folders in the sidebar to reorganise them, or ask Cleve to propose and create a folder structure for you.

Sharing a folder shares everything inside it, including docs and subfolders you add later.

## Workspaces

A **workspace** holds your docs, folders, chats, and settings. Everyone gets their own workspace (named after you, such as "Alex's Space"). Teams can create a shared workspace where members work from the same docs and context. See [Workspaces and teams](https://docs.cleve.ai/features/workspaces.md).

If you belong to more than one workspace, switch between them from the sidebar. Docs that other people share with you appear under **Shared with you**, even if they live in a workspace you aren't a member of.

## Cleve, your AI agent

**Cleve** is the AI agent built into your workspace. Open it from **Chat** in the sidebar or with `⌘J` / `Ctrl+J`.

Cleve works in one of two modes, toggled from the message box:

| Mode | What Cleve can do |
| - | - |
| **Edit mode** (default) | Read your docs and also create, edit, move, and organise docs and folders |
| **Chat mode** | Read your docs and reply in chat only, without changing anything |

Every change Cleve makes is saved in the doc's version history, so you can review what changed and restore the previous version. See [Chat with Cleve](https://docs.cleve.ai/features/ai-chat.md).

## Context

Context is what Cleve knows when it answers. It comes from:

* **The doc you have open** when you start a chat from it
* **Docs and folders you mention** by typing `@` in the message box
* **Your whole workspace**, which Cleve can search and read when it needs to
* **Files you attach**, such as PDFs, images, and text files
* **The web**, when web search is turned on
* **Apps you connect**, such as Gmail, Google Drive, Notion, and Slack
* **[Memory](https://docs.cleve.ai/features/memory.md)**, where Cleve keeps durable facts about you and your work (manage it in **Settings → Memory**)

The more relevant context you give Cleve, the better and more specific its answers are.

## Skills and Resources

A **skill** is a reusable instruction, such as "turn these notes into a LinkedIn post in my voice". Type `/` in the message box to run one. Create your own in **Settings → Skills**. See [Skills](https://docs.cleve.ai/features/skills.md).

**Resources** are skills and ready-made docs shared by Cleve and the community. Browse them from **Marketplace** in the sidebar and add them to your workspace. See [Resources marketplace](https://docs.cleve.ai/features/resources.md).

## Usage

Every reply from Cleve uses some of your plan's usage. How much depends on the model and on how long and complex the work is. See [How usage works](https://docs.cleve.ai/pricing/usage.md).


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