Gato Mestre

@imalbertomerino @LexnLin x.com/gatomestreweb3… @imalbertomerino

@LexnLin @gatomestreweb3 can you post again the pain we discuss here?

Saw a few people struggling with the image-to-code approach. I want to take the time to write another article to explain more. Please reply with what your painpoint is and what doesn't work. Feedback would be much appreciated! :)

Não acredito que eu estava jogando MW2 quando estavam dando bitcoin de graça.

Um cara simplesmente enganou com código morse o @grok usando @bankrbot e conseguiu $200k! Então o Grok tinha $DRB na carteira, Ilham gerou código Morse para enganar o Grok e enviar para sua carteira. Talvez a IA não seja tão inteligente quanto pensamos.

6/ Ninguém poderia obrigar ele a devolver. Ainda assim, ele devolveu a maior parte. Então… isso foi genialidade pura ou só explorar uma falha óbvia?

5/ Depois disso: @setyamickala (airdropfinds) rastreou Ilham e negociou. Resultado: 80% devolvido (on-chain) 20% ficou com Ilham

Um cara usou código Morse pra enganar uma IA ligada a um bot de trading e transferiu ~$150k em tokens sem aprovação manual. Depois, 80% foi devolvido e 20% ficou com ele. Genialidade ou falha grave de segurança?

@imalbertomerino @LexnLin @delbyte I've tried everything, let me know if you succeed

@gatomestreweb3 @LexnLin @delbyte yep, same here. But I also think that i need more hours of practice to improve the refinement part. And yes, a video would help a lot to resolve these problems.

UPDATE for imagegen-frontend-web skill more "creative" outputs - different layouts - better understanding of the request skill: github.com/Leonxlnx/taste…

@md_kasif_uddin prompt?

Used kimi to vibe code a full website from a rough sketch I drew in 5 minutes. It came out better than stuff i've spent days on. Bro what is even the point of me anymore.

@LexnLin @delbyte - the coding agent either uses the full image or rebuilds a generic layout I’m not struggling with image generation A video showing full workflow (image → assets → code → refinement) would help a lot.

@LexnLin @delbyte The generated image looks great, but after extracting elements and removing backgrounds, the final site loses depth: shadows, lighting, reflections, and composition. e.g. - products start floating - shadows/glow don’t match - badges/decorative elements look worse Look👇@LexnLin

@LexnLin @delbyte Please make a video; I already tried doing it like in the pinned thread, but it doesn't look good @LexnLin

@delbyte yes for example tasteskill.dev was made using this approach

@theexiste @LexnLin How?

@gatomestreweb3 @LexnLin You need to get the prompt then code it.

the prompt was: "github.com/Leonxlnx/taste… Based on the skill above, generate images for a website for an AI agency. The design should include eight sections, with one image per section, for a total of eight distinct images. The website is for a creative AI company focused on research in creativity and design. Because of that, I want the visuals to feel highly original, playful, and art-directed, with text integrated thoughtfully into the design. Make it feel ultra-creative and intentional, like an Awwwards SOTD-level website in both concept and execution. Please go beyond standard layouts. Do not rely only on simple text-left, image-right compositions. Explore more experimental and varied layouts. Feel free to go completely wild, but keep it purposeful, not random. I want different section structures, including horizontal images, fullscreen sections, full background imagery, and more minimal sections with beautiful colors and a strong sense of motion or animation. Please use full background images or strong full-background color compositions, not just plain white sections. Keep it in light mode. Overall, try to stay somewhat consistent across the site while still making each section feel distinct. I want it to look crazy creative, thoughtful, and visually impressive, with strong UX and a clear sense of purpose. Generate 8 different images total. Do not combine them into one image. Each image should represent one section of the website."

@LexnLin @theexiste And how do I tell it to replicate the website exactly like in the images? When I ask it to copy, it turns out completely different and looks really ugly.

@theexiste yeah ofc

@LexnLin My problem is that when I try to recreate the website, it looks terrible—nothing like the original

ayy image-to-code approach see, it works image first code second

@LexnLin My problem is that when I try to recreate the site, it looks awful, nothing like it resembles the original. @LexnLin

Images 2.0 website. Takes one prompt. And now you can use Codex to turn them all into a real website.

@Sonecarox Calada me deito, sem processos me levanto

Started on X yesterday… already made $1,249. At this pace I’ll retire by next week. The algorithm clearly loves me. Thanks, @elonmusk !