Schemas

This section explains how you tell anyformat what information matters and how it should be structured. You don’t need to think in code or formats. You describe data the way a human would, anyformat handles the rest.

---

​

What is a schema?

A schema defines the shape of the data you want to extract from a document. Think of it as a contract:

• What fields should exist
• What type of data each field contains
• What “correct” looks like

Schemas do not extract data by themselves. They describe the goal that workflows and AI work toward.

---

​

When you use schemas

You use a schema when:

• You want consistent output across many documents
• You care about structure, not just raw text
• You plan to validate or improve results over time

Most workflows are built around a single schema.

---

​

What schemas are not

Schemas are not:

• Rules engines
• Code
• OCR configurations

They don’t tell anyformat how to read documents — only what the result should look like.

---

​

What’s next?