A workflow for generating and refining AI visuals with clarity and control
Each prompt is built from four core components:
1. Subject
What’s in the image
Example: "A still life with ceramic bowls and fresh herbs"
2. Style / Medium
How it should look or feel
Example: "In the style of a Japanese woodblock print"
Example: "Hyperreal render, soft natural lighting"
3. Scene / Context
The setting, perspective, and composition
Example: "On a marble countertop, top-down angle, minimal clutter"
4. Mood / Tone
The emotional or atmospheric layer
Example: "Warm, organic, calming, early morning light"
[Subject], in [style/medium], [scene/context], with a [mood/tone] atmosphere
Example:
"A minimalist reading nook with linen cushions, illustrated in soft pencil style, surrounded by plants and filtered morning light, quiet and peaceful"