Announcement · Something I Built For You

7,000+ AI image prompts.
Searchable. Organized.
Yours to browse.

I got tired of saving prompts in random notes and scattered bookmarks. So I built a tool that organizes all of them in one place. Then I added something nobody else has.

Every week in this newsletter I share prompts. Photography terms that steer AI models. Social media workflows with copy-paste prompts. Camera angles, lighting setups, film stocks. Over time that library grows. But a prompt in an email gets read once and buried in your inbox. I wanted something better.

So I built PromptLens. It started as a personal tool to organize the prompts I was collecting and testing. Then I opened it up. Right now there are over 7,000 prompts in the gallery, organized by category, browsable, and searchable. Portraits, products, landscapes, food, architecture, fashion, fantasy, cinematic, street photography, anime, and more. You can go through them, find the ones that match what you are working on, and use them immediately.

But here is the part I am most excited about.

The feature nobody else has

See the prompt inside any image.

You know that feeling when you see an AI-generated image that is exactly what you have been trying to create and you think: what prompt made that? PromptLens answers that question. Drop any AI-generated image into it and it instantly reverse-engineers the prompt that would reproduce it.

Not a vague description. A detailed, model-specific prompt optimized for whichever tool you use. Upload an image and get the prompt formatted for Midjourney, ChatGPT, Gemini, Stable Diffusion, Flux, DALL-E, Grok, Ideogram, or Recraft. Each version uses the right syntax, the right parameters, and the right keywords for that specific model.

It also scores each prompt on quality and relevance. Quality measures how completely the prompt captures the visual elements of the image. Relevance predicts how closely the AI output would match the original if you used that prompt. So you know before you generate whether the prompt is going to work.

How to use it

Three ways PromptLens makes you better at prompting.

1. Browse the gallery for inspiration. Open the explore page and scroll through 7,000+ prompts organized by category. Find a style you like. Copy the prompt. Paste it into your AI tool. Modify it for your needs. This is the fastest way to go from "I do not know what to type" to a finished image.

2. Reverse-engineer images you love. Found an AI image on Pinterest, Instagram, or Twitter that is exactly the aesthetic you want? Save it. Upload it. Get the prompt. Now you have the recipe, not just the finished dish. This is how you learn what makes specific styles work. It is like having X-ray vision for prompts.

3. Translate prompts across models. You have a Midjourney prompt that works beautifully but you want to try it in ChatGPT Images 2.0 or Gemini. The syntax is different. The parameters are different. What works in one model does not always translate directly. Upload the Midjourney output to PromptLens and it gives you the optimized version for every other model. One image in, nine model-specific prompts out.

Why I built this

The gap between seeing and recreating.

Every issue of this newsletter teaches you photography vocabulary for AI prompts. Lenses, lighting, film stocks, composition. That knowledge is powerful. But even with all that vocabulary, there is still a gap between seeing an image you love and knowing how to describe it precisely enough to reproduce it.

It bridges that gap. It is the tool I wish I had when I started prompting. Instead of guessing what combination of words produced a specific look, you just ask. Upload. Read. Learn. Use.

The gallery is open to browse. The image-to-prompt tool gives you enough uses to analyze the images that matter most to you. The gallery and the core tool are available to everyone.

You can try it here: promptlens.polsia.app I would love to hear what you think.

Try it right now

One image. Five minutes. You will see.

Here is what I want you to do tonight. Think of an AI-generated image you have seen recently that you wished you could recreate. Maybe it was on Instagram, Pinterest, or Reddit. Save that image to your phone or desktop. Open the tool. Upload it. Read the prompt it generates. Then paste that prompt into whatever AI tool you use and hit generate.

The moment you see the output match the original is the moment you understand why prompt analysis matters as much as prompt writing. You have been learning the language in this newsletter. Now you have a translator.

Coming next

The one word that fixed all my portraits.

One photography term. Three prompts. The difference is immediate.

Have you tried PromptLens yet?

Hit reply and tell me what image you analyzed. I am reading every response. Your feedback is shaping what I build next.

Know someone who should read this? Forward this email to them.

See you next week.

Luxe Prompting

AI image generation for creators.