Lesson 51 of 1570
The Craft of Image Prompting
Great image prompters aren't typing harder — they're using a mental framework. Subject, setting, style, composition, lighting, mood. Here's the system.
Lesson map
What this lesson covers
Learning path
The main moves in order
- 1The six-slot recipe
- 2image prompting
- 3composition
- 4lighting
Concept cluster
Terms to connect while reading
Section 1
The six-slot recipe
Nearly every great image prompt fills six slots, implicitly or explicitly. When you're stuck, check which slot is empty.
- 1SUBJECT — who or what (a woman, a dragon, a skyline).
- 2SETTING — where (in a cafe, on Mars, in a misty forest).
- 3STYLE / MEDIUM — watercolor, photo, pixel art, 3D render.
- 4COMPOSITION — close-up, wide shot, bird's-eye, symmetrical, rule-of-thirds.
- 5LIGHTING — golden hour, harsh noon, neon, candle-lit, rim-lit.
- 6MOOD / COLOR — moody, cheerful, monochrome, pastel, saturated.
Six slots merged into one natural-language prompt.
SUBJECT: a marine biologist
SETTING: on the deck of a research ship
STYLE: cinematic photograph, Fuji film
COMPOSITION: medium shot, rule of thirds
LIGHTING: overcast natural daylight
MOOD: focused, weathered, hopeful
FINAL PROMPT: Cinematic photograph of a weathered marine biologist standing on the deck of a research ship, medium shot composed on rule of thirds. Overcast daylight, shot on Fuji film. Focused, hopeful expression.Composition words worth knowing
- Close-up, medium shot, wide shot, extreme wide shot.
- Bird's-eye view, low-angle, dutch angle, over-the-shoulder.
- Rule of thirds, symmetrical composition, leading lines, centered.
- Shallow depth of field, bokeh, everything in focus, macro.
Lighting vocabulary
- Golden hour, blue hour, overcast, harsh noon.
- Rembrandt lighting, rim light, backlit, silhouette.
- Neon-lit, candle-lit, moonlit, practicals (in-scene lights).
- Studio softbox, ring light, chiaroscuro.
Negative prompts (Stable Diffusion / Flux)
Some tools accept a 'negative prompt' — things you explicitly don't want. Use this sparingly; too much degrades quality.
Midjourney and Flux accept negative prompts via --no or via a dedicated field.
positive: A serene mountain lake at dawn, mirror reflection, oil painting
negative: boat, person, watermark, logo, text, blurry, extra fingersThe iterate-then-upscale workflow
- 1Generate cheaply at lower resolution / draft quality (Flux Schnell, Midjourney Relaxed).
- 2Pick the best 1-2 compositions.
- 3Upscale or re-run at highest quality (Flux Pro, Midjourney Turbo).
- 4Touch up in Photoshop / Firefly / inpainting if needed.
Compare the options
| Amateur prompt | Pro prompt |
|---|---|
| A cool sports car. | A 1971 candy-apple-red Dodge Charger on a desert highway at golden hour, low-angle three-quarter front shot, 35mm film, motion blur on wheels. |
| A sad person. | A young woman looking out a rain-streaked cafe window, cinematic medium shot, overcast natural light, muted blue-grey palette, shallow depth of field. |
Key terms in this lesson
End-of-lesson quiz
Check what stuck
15 questions · Score saves to your progress.
Tutor
Curious about “The Craft of Image Prompting”?
Ask anything about this lesson. I’ll answer using just what you’re reading — short, friendly, grounded.
Progress saved locally in this browser. Sign in to sync across devices.
Related lessons
Keep going
Builders · 26 min
DALL-E vs. Midjourney vs. Flux
Five image models, five personalities. Here's when each one is the right pick — in 2026, with current strengths, costs, and quirks.
Builders · 32 min
Video AI — Sora, Veo, Runway, Kling
Text-to-video became practical in 2025 and cinematic in 2026. Here's the state of the art and how to choose.
Builders · 40 min
Builder Capstone: Ship a Short Creative Piece
Your first end-to-end AI-assisted creative project. Plan it, make it, and reflect on what surprised you. Small scope, real output.