> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.instabase.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.instabase.com/_mcp/server.

# Project and app basics

> Create a project, classify documents, and add text extraction fields in the app editor.

Create a project, upload sample documents, classify document types, and add your first text extraction fields.

## Creating a project

This video shows how to create a project and upload documents:

1. Start from the homepage **Create** tile under projects, or from a workspace via **Create** > **App**.

2. Choose a blank project or start from an existing project schema.

3. Rename the project from the title if needed.

4. Upload sample documents that represent production file types.

## Adding classification

See how AI Hub organizes uploaded documents into classes based on content, structure, and layout. After classification, each class can have its own extraction schema so different document types are processed with the fields they need.

## Split classification

Learn how to split multi-page files into separate documents and classify them:

* Add classes (for example, invoice and passport) with optional descriptions; unmatched pages go to an **other** class.

* Enable splitting multi-page files by document boundaries rather than fixed page counts.

* Optionally use a custom splitting function when built-in splitting isn't enough.

## Adding text fields

Text fields, the most commonly used field type, extract specific values that appear directly on the document. Use them for straightforward values. Prefer other field types or custom logic when the answer requires calculation or reasoning that isn't printed on the page.

## Related documentation

* [Working with projects](/automate/projects)

* [Extracting data from documents](/automate/document-schema)