Skip to content

Describe (visual)

🤖 AI Drafted (Not reviewed)

Visually describe each selected image or PDF page: what is depicted, layout, figures, handwriting versus print, physical condition, and notable visual details. Saves one description artifact per page child.

Folder /Describe
Steps 2
Tags preset, describe, visual, vision, per-page

Steps, in run order

1. Files

Tool: Files — Pass through input files from workflow context

2. Describe visually

Tool: Describe — Generate image descriptions

Settings this step uses:

Option Value
detail_level detailed
prompt Visually describe this page or image. Include what is depicted, the layout and reading order, figures or illustrations, handwriting versus print, visible condition or damage, and any notable marks, stamps, seals, labels, or symbols. Do not transcribe the whole page; summarize visible text only when it helps identify the object.
save_to_db yes
vision_mode auto

What this step asks the model:

Describe this image.

Provide a detailed visual description covering what is depicted, layout, figures, handwriting versus print, condition, composition, and context.