How to Structure an AI Image Prompt

The Modular Prompt Formula

When text inputs are unstructured, neural image models struggle with concept bleeding (e.g. coloring the background blue instead of the character's coat). Structuring your prompt from macro-elements to micro-details resolves ambiguity.

The 7 Core Components Explained

1. Subject: The primary focus (e.g. an elderly watchmaker with spectacles).
2. Action: What the subject is doing (e.g. inspecting intricate brass gears under magnifying glass).
3. Environment: The surrounding world (e.g. in a cluttered vintage European workshop with mahogany shelves).
4. Composition & Camera: Frame orientation and perspective (e.g. close-up portrait, 85mm lens, f/1.8 shallow depth of field).
5. Lighting: Light direction and quality (e.g. warm tungsten lamplight from the side, dust motes dancing in light beam).
6. Style: Aesthetic genre (e.g. cinematic film still, rich chiaroscuro contrast).
7. Negative Constraints: What to exclude (e.g. no modern electronics, no blurry hands, no artifacts).