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

# Creativity from random source material

> Seed architectural concepts with random objects from a museum collection

Unexpected inputs make unexpected ideas. This workflow pulls random objects from the Smithsonian's open-access collection and abstracts each one into an architectural form.

![Random museum objects and the white card architectural models generated from each](https://mintcdn.com/runchat/ExzxwftWKPCOoCF3/images/examples/random-source-material.webp?fit=max\&auto=format\&n=ExzxwftWKPCOoCF3\&q=85\&s=c7261b37699d88e0203e6c6f98809962)

> [card] [Open the example workflow](https://runchat.com/editor?id=d723f533-8cfe-4e1c-bfea-99e41fe1928a)
>
> Make a copy of this workflow in your account to try it with your own images.

## How it works

1. **Pick random objects from the Smithsonian** is a code node that calls the Cooper Hewitt open-access API (CC0) and returns images.
2. **One architectural object per museum object (Nano Banana Pro)** runs once per object: *abstract one quality of it (its structure, rhythm, proportion, a fold, a pattern or a profile) into an inhabitable building form*.

## Tips

* Present results as white card models on a neutral backdrop. It keeps the focus on form, not rendering.
* Run the code node again for a new set of objects.


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