Gera Tools

Midjourney /describe Reverse Engineer Guide

Use /describe output to extract reusable style tokens from reference images.

Guide for Midjourney's /describe command. Paste a /describe result and the tool extracts reusable style, lighting, medium and composition tokens, flags the noise to drop, and helps you rebuild a clean prompt from a reference image. It runs free in your browser on Gera Tools, with nothing uploaded.

Last updated Source: Gera Tools

What does /describe actually do?

/describe asks Midjourney to look at an uploaded image and write four text prompts that could plausibly have produced it. It is a reverse-engineering tool — it guesses descriptive language, not the exact original prompt.

Midjourney /describe reverse-engineer guide

Midjourney’s /describe looks at a reference image and writes four candidate prompts that could have produced it. The real value isn’t the whole prompt — it’s the reusable style tokens hidden inside. Paste a /describe result and this tool splits it into phrases, buckets them into style, lighting, medium and composition, and flags generic filler to drop.

How to mine /describe output

/describe reconstructs language, not the literal original prompt — so don’t copy it wholesale. Instead, harvest the transferable parts:

  • Medium — “oil painting”, “3D render”, “35mm film”, “vector art”.
  • Lighting — “golden hour”, “rim light”, “soft studio lighting”.
  • Mood / aesthetic — named styles, eras, and atmospheres.
  • Composition — “close-up”, “wide shot”, “centered”, “rule of thirds”.

Discard the subject nouns (they tie the prompt to that specific image) and the generic filler (“highly detailed”, “trending”, “8k”) that adds noise without steering the look.

Tips for building a reusable style

  • Keep a token library of medium + lighting + mood combos you like.
  • Swap the subject, keep the style to carry a look across different scenes.
  • Combine tokens from multiple /describe runs to invent a hybrid aesthetic.
  • Test with a --sref style reference too — it often beats text tokens for pure style transfer.

A step-by-step workflow for mining /describe

Here is the practical sequence that turns a reference image into reusable prompt components:

  1. Upload the reference in Midjourney. Use /describe and select the image. Midjourney returns four numbered prompts. You do not need to use all four — focus on the one that best captures the qualities you want to borrow.

  2. Paste the output into this tool. The tool splits each candidate prompt into individual phrase tokens and categorizes them.

  3. Accept the style, lighting, and medium tokens. These transfer cleanly to new subjects: a phrase like “warm analog grain” or “editorial studio lighting” works the same whether the subject is a person, product, or landscape.

  4. Discard the subject nouns. If the reference was a cat on a windowsill, the phrases “tabby cat” and “wooden windowsill” are tied to that specific image. Drop them.

  5. Discard generic filler. Phrases like “highly detailed”, “4k”, “trending on ArtStation”, and “beautiful” add almost nothing to v6 and later — the model is already trained to produce quality imagery. They dilute the effective signal of the useful tokens.

  6. Rebuild with your own subject. Take the cleaned style tokens and wrap them around whatever subject you actually want to generate.

/describe vs —sref (style reference)

/describe gives you text tokens you can manipulate and reuse across any prompt. A style reference (--sref) directly transfers the visual aesthetic of an image without needing to describe it in words. The two approaches complement each other: use /describe when you want to understand why an image looks the way it does, or when you want to blend elements of a style into a different creative direction. Use --sref when you want direct, efficient style transfer and do not need to verbalize the look.