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

# Use Custom Logic

> Choose between AI logic and code for transformations and branching

# Custom Logic

You can implement logic in two ways.

## Option 1 - AI Agent Node

Describe the logic in plain English. Great for flexible or fuzzy rules like classification or summarization.

* Pros - fastest to build, adapts to varied inputs.
* Cons - not deterministic. The Agent may choose different actions on similar data.

## Option 2 - Code Editor

Write logic in free text or Python to ensure consistent behavior.

* Pros - deterministic and repeatable across inputs.
* Cons - Breakable.