CropUp Ai
07
Prompt Fixer and Optimizer
Paste a bad prompt into this Prompt Fixer and Optimizer to get a layer-by-layer diagnosis, a fully rewritten version, and a before/after score comparison.
6-layer audit
Before/after score
Fix it again loop
01
Paste your prompt
Enter the information above. The more specific you are the better your results.
02
Claude analyzes and generates
Our AI processes your input using the 6-layer prompt framework and brand vocabulary system.
03
Copy and use instantly
Copy your results directly into Midjourney, ChatGPT, or any other AI tool and generate.
Last updated: July 2026
Curious why your prompts keep coming out flat? Read Why Your AI Images Look Generic.
Recent fixes
Diagnosing your prompt...
Your results will appear here
Fill in the details above and click generate to see your fixed prompt.
Layer audit
Prompt score
0/10
→
0/10
OriginalFixed
Your original prompt
Fixed prompt from the Prompt Fixer and Optimizer
Now optimize this fixed prompt for your platform. Open the Platform Optimizer.
What changed and why
Fix it again
The 10 most common prompt mistakes
Bookmark this page - these mistakes account for 90% of bad AI image results.
| Mistake | Bad example | Fixed example |
|---|---|---|
| Too vague | "a woman in nice clothes" | "a woman in a tailored ivory linen blazer, fitted at the waist, standing with relaxed confidence" |
| No lighting defined | "product on table" | "product on white marble surface, soft diffused window light from the left, warm undertones" |
| No camera or lens | "close-up of face" | "close-up beauty shot with 85mm lens at f/2.0, shallow depth of field, single catchlight" |
| Generic background | "nice background" | "seamless warm beige background, subtle center gradient, no visible edges or shadows" |
| No mood or atmosphere | "coffee shop scene" | "cozy coffee shop interior, warm amber tungsten light, soft bokeh background, steam rising" |
| Vague color description | "blue dress" | "deep cerulean silk midi dress with cool purple undertones catching soft window light" |
| Missing composition | "model standing" | "model positioned on left third of frame, generous negative space to the right, head room above" |
| No material detail | "leather bag" | "cognac leather crossbody with visible natural grain texture, subtle gloss on hardware" |
| No depth or layers | "outdoor scene" | "foreground blurred foliage framing edges, middle ground subject in sharp focus, soft city blur background" |
| Using AI language | "make it realistic" | "shot on Canon EOS R5 50mm f/1.8, natural skin texture, visible pores, asymmetrical features" |