Prompt anatomy: subject, style, light, framing

"A cat" gets you a random cat. The formula below gets you your cat.

The formula: subject → style → light → framing

a ginger cat asleep on a stack of old books by a window (SUBJECT+ACTION),
cosy watercolour illustration with soft edges (STYLE),
warm evening light through the window, long shadows (LIGHT),
close-up, shallow depth of field, blurred background (FRAMING)

Layer 1: subject and action

Specifics do the work: not "a cat" but "a ginger cat asleep on a stack of books". One anchor detail — the stack of books, a mug on the sill — turns a picture into a story instead of a stock photo.

Layer 2: style

Name the medium and the mood: "watercolour illustration", "shot on film", "flat vector graphics", "Pixar-style 3D render". Style is the strongest layer you have — one phrase changes everything.

Layer 3: light

Light is mood: "warm evening light through a window", "neon glow", "soft diffused studio light", "dramatic backlight". Beginners forget light, and their pictures come out flat.

Layer 4: framing

Shoot it the way a photographer would: "close-up", "top-down view", "wide shot with lots of room", "shallow depth of field". This is how you steer the viewer's eye.

The rule: describe what you WANT to see

Generators are bad at negatives — "no people" can hand you people. Instead of "not gloomy", say "bright and airy". Positive phrasing was a habit in the prompting course; here it's closer to a law of physics.