Skip to content
Agentmart
Submit a tool

image-prompt

Write four distinct prompts for AI image models.

Skill by davidondrej

IndexedMIT★ 4,167
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/davidondrej-image-prompt

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Write four distinct prompts for AI image models. Use for image, thumbnail, banner, logo, or illustration prompts. Produces prompt text only.

Use cases: design, writing, agent-building, prompt-engineering, image-generation

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/image-prompt

What's inside

Skill name
image-prompt
Description
Write four distinct prompts for AI image models. Use for image, thumbnail, banner, logo, or illustration prompts. Produces prompt text only.
License
MIT
Instructions
46 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Image Prompt

Write 4 model-agnostic prompt variations that the user can paste into any image model.

## Output format

- Exactly 4 variations, each one paragraph in its own fenced code block. Nothing else inside.
- Always use a `text` fence so it renders as a full copyable block, not a tiny snippet. Pattern:
  1. Blank line after the label
  2. Opening fence on its own line: three backticks then `text`
  3. The prompt paragraph
  4. 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

Describe a finished piece in clear, expressive, descriptive prose. Each paragraph covers:

1. **Subject** — what it is, what it is doing, the defining details.
2. **Environment** — where it sits, background, lighting, time of day.
3. **Color palette** — name the dominant colors and the contrast.
4. **Vibe and feeling** — the mood and the emotion the image should evoke.
5. **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**

```text
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.)

From davidondrej/skills, MIT.

More from davidondrej/skills

Compare and alternatives

Featured in

Signals

Stars
4,167
Forks
612
Open issues
0
Last commit (this folder)
Oct 18, 2026 (9 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from davidondrej

Cite this page

image-prompt skill. Agent Mart. https://agentmart.dev/skills/davidondrej-image-prompt (last updated 2026-10-24)