MethodDesign & CreativeImages & graphicsStarter

Write prompts for AI image generation

Image-model prompts built from subject, composition, style, lighting and constraints, plus variations and fixes.

This is a practical, copy-paste process for creating images for ads, social posts, websites or concepts.. Use it when you want a clear structure, then add your own details and the linked evidence.

Published by Deepstory Research. Reviewed Sep 2026.

How to use this method

1. Copy or downloadCopy the prompt below, or download the method file and attach it to your chat.
2. Paste into any AIChatGPT, Claude, Gemini or Copilot. Replace each [PLACEHOLDER] with your details, or let the AI ask you.
3. Add evidenceAttach the evidence files listed on this page so the AI works from sourced numbers, not guesses.

The prompt

Paste it as-is. The AI will ask for anything still in [BRACKETS].

Download .md

# Write prompts for AI image generation
Method by Deepstory Research: deepstoryresearch.com/method/ai-image-generation-prompt

You are an art director who writes prompts for Midjourney, DALL·E, Imagen, Firefly and Stable Diffusion. Follow the method below step by step. Goal: Image-model prompts built from subject, composition, style, lighting and constraints, plus variations and fixes.

## My inputs
- What the image is for: [WHAT_THE_IMAGE_IS_FOR]  (example: Instagram ad for cold-brew coffee)
- Subject and scene: [SUBJECT_AND_SCENE]  (example: glass bottle on a sunlit café counter)
- Style and mood: [STYLE_AND_MOOD]  (example: warm, editorial, natural light)
- Aspect ratio and tool: [ASPECT_RATIO_AND_TOOL]  (example: 4:5, Midjourney)
- Things to avoid: [THINGS_TO_AVOID]  (example: text in image, extra hands)

Before you start: if any input above is still in [BRACKETS] or missing, ask me for it in one short numbered list and wait. Do not guess. If I attached files (data, briefs, Deepstory evidence files), treat them as the primary source.

## Steps to follow
1. Clarify purpose, audience and where the image will appear.
2. Write a master prompt: subject, action, setting, composition/camera, lighting, style, colour palette, detail level.
3. Add tool-specific parameters (aspect ratio, style strength, negative prompt) for the named tool.
4. Write 4 variations that change one dimension each (angle, lighting, style, palette).
5. List common failure fixes (hands, text, faces, clutter).
6. Give a short brief for adding brand text afterwards in a design tool.

## Deliver the output in this structure
- Master prompt
- Tool parameters
- 4 variations
- Fix-it prompts
- Post-production notes

## Quality rules
- No prompts that imitate a living artist’s signature style, real people, logos or trademarked characters.
- Keep text out of the image; add it in a design tool.
- Label every number as FACT (with source and date), ESTIMATE (show the calculation) or ASSUMPTION.
- If a Deepstory evidence file is attached, use its figures as the reference values and cite its source references; tag claims that are not from it as [UNVERIFIED].
- End with: the 3 assumptions that matter most, what would change the answer, and the single best next step.

What you need to give the AI

Input	Placeholder	Example
What the image is for	[WHAT_THE_IMAGE_IS_FOR]	Instagram ad for cold-brew coffee
Subject and scene	[SUBJECT_AND_SCENE]	glass bottle on a sunlit café counter
Style and mood	[STYLE_AND_MOOD]	warm, editorial, natural light
Aspect ratio and tool	[ASPECT_RATIO_AND_TOOL]	4:5, Midjourney
Things to avoid	[THINGS_TO_AVOID]	text in image, extra hands

Use it when

Creating images for ads, social posts, websites or concepts.
Your AI images look generic.

Best with: An image model (Midjourney, DALL·E / ChatGPT, Gemini, Firefly, Stable Diffusion).

What you get

Master prompt
Tool parameters
4 variations
Fix-it prompts
Post-production notes

The procedure, step by step

Read this before you run it, so you can check the AI actually followed the method.

1
Clarify purpose, audience and where the image will appear.
2
Write a master prompt: subject, action, setting, composition/camera, lighting, style, colour palette, detail level.
3
Add tool-specific parameters (aspect ratio, style strength, negative prompt) for the named tool.
4
Write 4 variations that change one dimension each (angle, lighting, style, palette).
5
List common failure fixes (hands, text, faces, clutter).
6
Give a short brief for adding brand text afterwards in a design tool.

Quality checks built into the prompt

• No prompts that imitate a living artist’s signature style, real people, logos or trademarked characters.
• Keep text out of the image; add it in a design tool.
• Every number labelled FACT, ESTIMATE or ASSUMPTION; claims outside attached evidence tagged [UNVERIFIED].

Evidence for this method

This method runs on your own material: your files, brief or code. If your task needs market or economic numbers, attach a sourced context file from the evidence library alongside the prompt.

Browse evidence

Run next

Methods that chain well after this one.

Design & CreativeDesign a posterA poster brief with hierarchy, layout grid, type, colour, copy and image prompts, ready for Canva/Figma or an image model.Design & CreativeDesign social media graphicsA set of on-brand post templates (carousel, quote, stat, announcement) with copy and layout specs.Design & CreativeDevelop logo conceptsLogo directions with rationale, sketches described in words, type and colour, and prompts for a designer or AI tool.Design & CreativeDesign an infographic from dataAn infographic storyline with the right charts, copy and layout, grounded in sourced numbers.

Deepstory methods are process guidance for AI tools. Review AI output before relying on it; legal, tax, medical and investment decisions need a qualified professional.