A minimalist, high-speed text-to-image generator built to eliminate prompt friction and keep your digital art history saved directly in the browser.
You can experience the live deployment of the app right here: 👉 Try Project Live Demo (Note: Replace this with your actual Vercel/GitHub Pages live link)
- Type any creative description in the input bar (e.g., "A cosmic cat wearing a space helmet").
- Optional: Click on one of the Enhance Style buttons (Realistic, Cyberpunk, Anime) to automatically optimize your keywords.
- Select your desired Aspect Ratio (1:1, 16:9, or 9:16).
- Hit Generate and watch your imagination come to life!
Unlike heavy full-stack setups, this application is engineered to be lightweight, responsive, and completely client-side.
When you trigger a creation, the script captures your raw text input, marries it with the metadata of your selected style tags, and dynamically computes the absolute resolutions based on your aspect ratio choice (
To solve the friction of losing generated art upon browser refreshes, I designed a lightweight state sync using browser LocalStorage. Every time an image fully loads (img.onload), its direct secure URL and original prompt parameters are prepended into a structured JSON array, capped at a maximum of 6 elements to prevent local storage bloat, and re-rendered into a responsive CSS Grid. Clicking any item in the history sidebar instantly pulls the snapshot back onto the main focus view.
Want to run or tweak this project on your machine? It requires no complicated backend environments.
- Clone the repository:
git clone [https://github.com/Murodbektuychiyev/Frictionless.git](https://github.com/Murodbektuychiyev/Frictionless.git)
cd Frictionless