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

# Types of automation

> Choose a focused automation, a sequence of saved steps, or a workflow that uses AI judgment during execution.

Ideatura offers three ways to organise automated work. Choose according to whether your task needs a defined operation, a fixed sequence, or AI judgment while it runs.

## Compare the options

| Type | How it works | Architectural example |
| - | - | - |
| **Normal automation** | Runs a defined procedure using your selected inputs. | Dimension grids with a chosen style, sides and offsets. |
| **Chained automations** | Runs saved automations in a fixed sequence. | Create room sheets, plans, RCPs and elevations, then place the views on their corresponding sheets. |
| **AI Flow** | Combines AI judgment during execution with model operations. | Review a drawing package against a brief, interpret presentation issues and return an action list with evidence. |

## Normal automation

Use a normal automation when you know the procedure and need to choose its inputs. For grid dimensions, parameters can specify the views, dimension style, placement sides and offsets.

Start with [Find an automation](https://docs.ideatura.ai/automation/automation-basics/shared-automation-library-guide.md).

## Chained automations

Use a chain when several saved procedures need to run in order. Each step keeps its own purpose, while later steps use the outputs created earlier. Configure the inputs and check how those outputs are matched between steps.

See [Chained automations](https://docs.ideatura.ai/automation/chains.md).

## AI Flow

An AI Flow is one saved automation that can include parameters, internal AI prompts and model operations. Use it when the procedure needs to interpret information and decide what to do while it runs.

Branches, checks and retries must be designed into the Flow; they do not happen automatically. A chain can also include an AI Flow as one of its steps.

See [AI Flows](https://docs.ideatura.ai/automation/ai-flows.md).

## Choose the simplest fit

Start with a normal automation for a defined task. Use a chain to connect reusable steps, or an AI Flow when decisions depend on information found during execution. Test a small case and inspect the result before extending the scope.


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