Guides
Learn how AI image prompts actually work
Promptify writes prompts for you, but the results get much better once you understand what a prompt is made of and which words a model is actually sensitive to. These guides are written from the same reading of images that powers the tools — no listicles, no copied prompt packs.
What Is an AI Image Prompt? A Complete Guide
The anatomy of a prompt — subject, composition, lighting, camera language, style, colour and detail — and the mistakes that quietly ruin results.
Read the guide →How to Write Better AI Image Prompts
A practical ordering method: start with the subject, build outward to environment and light, then control style and colour without contradicting yourself.
Read the guide →Image-to-Prompt Generation: How It Works
What a model can genuinely read from a picture, what it can only infer, where the limits are, and how to get a sharper description out of a reference.
Read the guide →How to Create AI Prompts for Design Inspiration
Prompting for layout work — websites, posters, editorial spreads, branding, social graphics and product interfaces — where typography and grid matter more than mood.
Read the guide →AI Image Prompt Examples
Worked examples with the creative objective, the prompt itself, and a breakdown of which parts are doing the heavy lifting.
Read the guide →Website Theme Grab: A Full Worked Example
A complete theme grab written out — layout order, colour roles with hex values, the type scale and the finished build-ready prompt.
Read the guide →Poster and Infographic Grab: A Full Worked Example
Two full poster briefs — a festival sheet and a data explainer — with the grid, hierarchy, palette and print treatment spelled out.
Read the guide →Midjourney Image to Prompt Generator
Turn a reference image into a Midjourney-ready prompt, with a parameter reference for --ar, --v, --style raw and --stylize.
Read the guide →When you want to try the ideas out, open the prompt tools or browse the annotated examples.