Image Prompt
Turn the request into 4 prompt variations. The user pastes them into whatever image model they are using, so the prompts must be model-agnostic.
Output format
- Exactly 4 variations. Each is ONE paragraph inside its own fenced code block. Nothing else inside the code block.
- Always use a
text fence so it renders as a full copyable block, not a tiny snippet. Pattern:
- Blank line after the label
- Opening fence on its own line: three backticks then
text
- The prompt paragraph
- Closing fence on its own line
- Never wrap a prompt in single backticks. Never use a bare unlabeled fence. Those render as small snippets and are hard to copy.
- Every prompt ends with
--ar 1:1. It is the Midjourney aspect-ratio flag and is harmless in other models. If the user names an aspect ratio, use that instead (e.g. --ar 16:9).
- A 3-5 word label above each block is fine. No commentary, no explanations, no questions after.
How to write each prompt
Write like a great expressive artist describing a finished piece. Concise, clear, highly descriptive. Every paragraph covers, in flowing prose:
- Subject — what it is, what it is doing, the defining details.
- Environment — where it sits, background, lighting, time of day.
- Color palette — name the dominant colors and the contrast.
- Vibe and feeling — the mood and the emotion the image should evoke.
- Style / medium — photograph, cinematic still, 3D render, oil painting, flat vector, etc.
Rules
- Descriptive prose only. No bullet lists inside a prompt, no "imagine a...", no "create an image of".
- Make the 4 variations genuinely different — different style, composition, mood, or palette. Not rewordings of one idea.
- No text inside the image unless the user asks; models render text badly.
- No negative prompts and no model-specific parameters other than
--ar.
- If the request is vague, still deliver 4 prompts. Pick sensible defaults, do not ask questions.
Example
Request: "logo concept for Vectal Labs, something about invention"
Filament in the void
A single glowing filament bulb suspended in a vast dark void, its wire coiled into a rising spiral, casting warm amber light onto faint blueprint lines etched into black glass beneath it. Deep charcoal and midnight blue with one accent of molten gold. Quiet and reverent, the feeling of a first idea arriving at 3 a.m. Minimal, high contrast, rendered like a luxury product photograph. --ar 1:1
(…then three more, each in its own code block, each a different direction.)
1---2name: image-prompt3description: Turn any idea, brief, or rough instruction into 4 ready-to-paste prompts for AI image models (Midjourney, Nano Banana, GPT Images, Seedream, Flux, any other). Use when the user asks for an image prompt, a thumbnail / banner / logo / illustration prompt, says "write a prompt for this image", or wants to turn a description into something an image model can render. Differentiator: writes the prompt text only — to actually generate the image use the deepapi skill.4---56# Image Prompt78Turn the request into 4 prompt variations. The user pastes them into whatever image model they are using, so the prompts must be model-agnostic.910## Output format1112- Exactly 4 variations. Each is ONE paragraph inside its own fenced code block. Nothing else inside the code block.13- Always use a `text` fence so it renders as a full copyable block, not a tiny snippet. Pattern:14 1. Blank line after the label15 2. Opening fence on its own line: three backticks then `text`16 3. The prompt paragraph17 4. Closing fence on its own line18- Never wrap a prompt in single backticks. Never use a bare unlabeled fence. Those render as small snippets and are hard to copy.19- Every prompt ends with ` --ar 1:1`. It is the Midjourney aspect-ratio flag and is harmless in other models. If the user names an aspect ratio, use that instead (e.g. `--ar 16:9`).20- A 3-5 word label above each block is fine. No commentary, no explanations, no questions after.2122## How to write each prompt2324Write like a great expressive artist describing a finished piece. Concise, clear, highly descriptive. Every paragraph covers, in flowing prose:25261. **Subject** — what it is, what it is doing, the defining details.272. **Environment** — where it sits, background, lighting, time of day.283. **Color palette** — name the dominant colors and the contrast.294. **Vibe and feeling** — the mood and the emotion the image should evoke.305. **Style / medium** — photograph, cinematic still, 3D render, oil painting, flat vector, etc.3132## Rules3334- Descriptive prose only. No bullet lists inside a prompt, no "imagine a...", no "create an image of".35- Make the 4 variations genuinely different — different style, composition, mood, or palette. Not rewordings of one idea.36- No text inside the image unless the user asks; models render text badly.37- No negative prompts and no model-specific parameters other than `--ar`.38- If the request is vague, still deliver 4 prompts. Pick sensible defaults, do not ask questions.3940## Example4142Request: "logo concept for Vectal Labs, something about invention"4344**Filament in the void**4546```text47A single glowing filament bulb suspended in a vast dark void, its wire coiled into a rising spiral, casting warm amber light onto faint blueprint lines etched into black glass beneath it. Deep charcoal and midnight blue with one accent of molten gold. Quiet and reverent, the feeling of a first idea arriving at 3 a.m. Minimal, high contrast, rendered like a luxury product photograph. --ar 1:148```4950(…then three more, each in its own code block, each a different direction.)
David Ondrej