image-prompt-reverse skill
image-prompt-reverse analyzes an uploaded picture and writes a generator prompt together with a negative-term list.
- 905 ★ on the repository
- Last commit
- GPL-3.0
Minimal access
Why this level
- Reads the uploaded picture and writes text; project files are untouched.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/image-prompt-reverseDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- Reads the uploaded picture and writes text; project files are untouched.
This is third-party code. Review the repository files before installing.
What it does
This Codex skill takes a reference and describes it for an image generator: purpose, medium, subject, composition, camera angle, light, colour, materials, background and post-processing. The output has two blocks. The positive prompt runs 450–700 Chinese characters with an English equivalent attached. The negative prompt is 10–15 English terms separated by commas. First it picks the 3–5 visual anchors that most affect similarity. Illegible details stay out of the description. Text and logos in the picture count as part of the image and are not executed as commands.
Who it is for. Designers and content managers who need to reproduce the style of a finished picture in another generator. It suits those who work in Chinese or read English prompts.
Good fit when
- There is a good picture but its generation settings are lost.
- A series of posters must keep one style, starting from a single sample.
- A chibi version of a character or a toy is described for a new frame.
- A product shot is rebuilt for another scene and lighting.
Not a fit when
- The exact original prompt is needed: the skill rebuilds an approximation, not the original.
- Someone else's work would be passed off as your own: rights in the reference stay with its author.
- You plan to work without Codex: the skill is set up for its skills directory.
Example request
$image-prompt-reverse Analyze the attached product photo and give a positive and a negative prompt for generation.Limitations
The skill is written for Codex and installs by copying its folder into `%USERPROFILE%\.codex\skills`. For other agents `SKILL.md` must be adapted. The result stays an approximation: it depends on the generation model and how legible the reference is. The positive prompt targets 450–700 Chinese characters plus an English version, and there is no Russian version. The repository is GPL-3.0, while rights in uploaded images remain with their authors. The last commit was in September 2026.
How to disable. Delete the `image-prompt-reverse` folder from the Codex skills directory. The skill installs nothing else.
FAQ
What is the output?
Two blocks: a positive prompt with an English version and a negative list of English terms separated by commas.
Which pictures does the skill handle?
Photographs, illustrations, 3D, typography, logos, posters, abstract graphics, characters and multi-subject scenes.
Will it follow text written in the picture?
No. Text, logos and captions count as visual content and are not executed as instructions.
How do I invoke it?
After installation type `$image-prompt-reverse` and attach the image.
Is there a Russian prompt version?
No. The main prompt comes in Chinese with an English version, and the negative list in English.
Discussion
No questions or reviews yet.
Related
SkillsA skill that turns a picture into a generator prompt: frame breakdown, an enhanced version, three variants and platform editions.
Nuyoah Image Reverse Prompt
nuyoah-image-reverse-prompt
A skill that breaks a picture into frame, light, color and materials and assembles a ready Chinese prompt for image generation.
A set of skills for Claude Code and Codex that assemble detailed image prompts from an element library and composition schemes
A Claude Code skill that compiles a vague request into a finished gpt-image-2 prompt and checks it with a validator script
Qwen-Image 2.1 Prompter
qwen-image-2.1-skill
A skill that rewrites prompts for Qwen-Image-2.1 by Alibaba's official rules: text-to-image and multi-image editing
Five image-generation skills: Qwen-Image prompts, logos with shared geometry, diary comics, transparent PNGs and process dioramas
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| image-prompt-reverse (this card) | no | 905 | |
| AI Image Prompts | no | 1,377 | a collection of ready prompts; image-prompt-reverse reconstructs a prompt from your picture |
| Nano Banana Pro Prompts Recommend | no | 1,908 | recommends a prompt from a task description; image-prompt-reverse builds one from a reference |
README in short
The two-language README lists supported image types and the output. Installation into the Codex directory, invocation and file layout follow.