> For the complete documentation index, see [llms.txt](https://docs.katara.ai/about/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.katara.ai/about/start-here/readme.md).

# Katara documentation

Katara helps your organization turn source documents into knowledge that people and AI clients can use with explicit access controls. Use the application to upload content, maintain Brains, organize searchable collections, configure Q\&A agents and connect AI clients through MCP.

## Choose your starting point

| I want to…                                         | Start here                                                                                                    |
| -------------------------------------------------- | ------------------------------------------------------------------------------------------------------------- |
| Set up my workspace and understand the application | [Quickstart](/about/start-here/quickstart.md)                                                                 |
| Build a shared or personal knowledge space         | [Brains](/about/brains/overview.md)                                                                           |
| Add files to an existing Brain                     | [Brain sources](/about/brains/sources.md)                                                                     |
| Read or improve the Wiki                           | [Knowledge and pages](/about/brains/knowledge.md)                                                             |
| Record a fact for future retrieval                 | [Facts and memory](/about/brains/facts.md)                                                                    |
| Organize existing documents for answering          | [Collections](/about/documents-and-collections/collections.md)                                                |
| Configure and test answers                         | [Q\&A agents](/about/agents-and-answering/qna.md) and [Prompt Lab](/about/agents-and-answering/prompt-lab.md) |
| Connect an external AI client                      | [MCP setup](/about/connect-ai-clients/connect.md)                                                             |
| Invite teammates or manage access                  | [Members and groups](/about/organization-administration/members.md)                                           |

## Understand the building blocks

**Documents** are the accepted source content. **Collections** select and synchronize documents for source-based answering. A **Brain** combines its directly uploaded originals with a Wiki and memory, with its own ownership and audience. A **Q\&A agent** binds answering to one Collection or Brain. An **MCP client** accesses the operations its authenticated identity is allowed to use.

These are related, but not interchangeable. Sharing one original document does not share its Brain. Upload acceptance does not mean indexing is complete. Saving export settings does not publish the Wiki to GitHub.

Read [Concepts and navigation](/about/start-here/concepts.md) for a fuller map.

## Availability and permissions

The guides follow the current application source, including newer Brain workflows. Your deployment, subscription, organization membership and resource-specific permissions determine which controls work for you. Brain knowledge, Brain-backed answering and GitHub export require deployment configuration; they are not assumed to be enabled everywhere.

See [Feature availability](/about/reference/availability.md) for the distinction between dashboard workflows, optional capabilities, backend-only features and retired integrations. These guides are not a claim that every feature has been verified on your live deployment.