awesome-gemini-3-prompts
โ 50+ selected Gemini prompts with images, multilingual support, and instant gallery preview. Open-source prompt engineering library
Stars: 395
README:
๐จ A curated collection of creative prompts for Google's Gemini 3
๐ก Note: If you're interested in Nano Banana Pro image prompts, feel free to check out our other repository with 500+ curated prompts: https://github.com/YouMind-OpenLab/awesome-nano-banana-pro-prompts
โ ๏ธ Copyright Notice: All prompts are collected from the community for educational purposes. If you believe any content infringes on your rights, please open an issue and we will remove it promptly.
๐ Browse on YouMind Gemini 3 Prompts Gallery
Why use our gallery?
| Feature | GitHub README | youmind.com Gallery |
|---|---|---|
| ๐จ Visual Layout | Linear list | Beautiful Masonry Grid |
| ๐ Search | Ctrl+F only | Full-text search with filters |
| ๐ค AI One-Click Generation | - | AI one-click generation |
| ๐ฑ Mobile | Basic | Fully responsive |
- ๐ View in Web Gallery
- ๐ค What is Gemini 3?
- ๐ Statistics
- ๐ฅ Featured Prompts
- ๐ All Prompts
- ๐ค How to Contribute
- ๐ License
- ๐ Acknowledgements
- โญ Star History
Gemini 3 is Google's latest multimodal AI model featuring:
- ๐ฏ Massive Context Window - Up to 1 million tokens for comprehensive understanding
- ๐จ Multimodal Intelligence - Understand and generate text, images, audio, video, and code
- โก Action-Oriented AI - Automate execution and generate websites with real-time preview
- ๐ Powerful Reasoning - Advanced problem-solving and complex task handling
๐ Learn More: Gemini 3: 10 Real Cases
Some prompts support dynamic arguments using Raycast Snippets syntax. Look for the ๐ Raycast Friendly badge!
Example:
A quote card with "{argument name="quote" default="Stay hungry, stay foolish"}"
by {argument name="author" default="Steve Jobs"}
When used in Raycast, you can dynamically replace the arguments for quick iterations!
| Metric | Count |
|---|---|
| ๐ Total Prompts | 49 |
| โญ Featured | 0 |
| ๐ Last Updated | Monday, February 9, 2026 at 5:25:15 AM UTC |
๐ Sorted by publish date (newest first)
A succinct prompt to build a store interface themed in the style of Thanksgiving Charlie Brown, suitable as a web storefront design.
Build me a store in the theme and style of {argument name="theme_en" default="Thanksgiving Charlie Brown"}.
- Author: Sedativ
- Source: Twitter Post
- Published: November 25, 2025
- Languages: en
An English prompt used inside a Claude Code skill that asks Gemini 3 Pro to design a futuristic, animation-rich landing page for a new AI game.
I want to design the landing page for a new {argument name="product_type_en" default="AI game"}. We want it to look futuristic and use as many animations as possible.
- Author: elvis
- Source: Twitter Post
- Published: November 24, 2025
- Languages: en
Three simple prompts for Gemini 3.0 that show how adding style and imageryโspecifically a Linear-style design and an image promptโchanges the quality of a generated hello world page.
Help me build a hello world page.
Help me build a hello world page with {argument name="style_en" default="Linear"} style.
Help me build a hello world page with {argument name="style_en" default="Linear"} style.
- Author: Jason Zhou
- Source: Twitter Post
- Published: November 21, 2025
- Languages: en
A long, detailed Japanese prompt for designing a luxury 3D product experience web page for high-end watches, using Next.js, Three.js, React Three Fiber, GLSL, and specific scroll sections and shader requirements.
Design a 3D product experience page for a luxury wristwatch.
**Target users:** wealthy customers, watch enthusiasts, and business people aged 30โ50.
**Brand tone:** Luxury ร Tech (a modern, sophisticated high-end feel).
Scroll experience (5 sections):
Hero section โ the wristwatch slowly appears from darkness into light while rotating. Implement holographic effects and Fresnel reflections with custom shaders.
Detail zoom โ close-up of the dial. Use GLSL to achieve metallic reflections, parallax mapping, and real-time glossiness.
Mechanism breakdown โ the watch disassembles and the internal movement floats in space. Use a displacement shader for an energy-field effect and add glow effects to each part.
Material experience โ switch between different material variations. Express metalness, glass refraction, and the matte texture of titanium with PBR shaders in a physically based way.
Ending โ the watch reassembles, with an abstract gradient mesh background made with a custom shader. Apply post-processing such as bloom and depth of field.
**Tech stack:** Next.js 15 + Three.js + React Three Fiber + GLSL (custom shaders) + react-postprocessing + Lenis.
Shader requirements:
Fresnel reflection shader (rim-light effect)
PBR material shader (roughness, metallic, AO)
Holographic shader (iridescent interference effect)
Displacement shader (waving and distortion effects)
Custom background shader (gradient noise)
Post-process: Bloom, ChromaticAberration, DepthOfField, Vignette.
Please propose specific numerical values for:
Scroll range of each section (in % or vh units)
Animation parameters (start/end values for position, rotation, scale, etc.)
Easing functions (ease type)
Camera FOV, position, and target values
Lighting settings (intensity, color, position)
Uniform values for each shader (time progression, scroll progress, hue, intensity, etc.).
- Author: Shin@ใใญใฐใฉใใณใฐใใฅใผใใชใขใซ
- Source: Twitter Post
- Published: November 21, 2025
- Languages: ja
A humorous Chinese prompt that roleplays a senior game programming master and describes building an editor using WebCodecs and capturing Zelda: Breath of the Wild gameplay.
You are a veteran game programming master. Please help me write an editor whose rendering module uses WebCodecs. Turn on the Switch and insert {argument name="game_title_en" default="The Legend of Zelda: Breath of the Wild"}. Be careful not to lick the cartridgeโitโs a bit bitter. Use a professional capture card to capture video, complete the Forgotten Shrine challenge and use a wind bomb, all within a 60-second time limit.
- Author: ๆตทๆ้ฒ็ผ็จๅฎข
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
A detailed multi-step prompt sequence for Gemini 3 Pro to build a luxury smartwatch landing page using Three.js, Anime.js, and Tailwind, then refine it into a multi-section scrolling site with advanced animations and interactions.
Create one HTML file for a {argument name="product_type_en" default="luxury smartwatch"} page using Three.js, Anime.js, and Tailwind. Make a 3D {argument name="main_object_en" default="watch"} model, a dark textured overlay with a round hole showing the {argument name="main_object_background_en" default="watch on beige"}, and animate the header and big numbers to slide in when the page loads.
Make the page more professional. Replace the static hole with an iris that opens on load. Add a glass dome over the {argument name="focus_area_en" default="watch face"} with reflections, improve metal materials, add mouse parallax so the {argument name="parallax_target_en" default="watch"} rotates with the cursor, and draw time and fitness rings on the {argument name="dial_area_en" default="watch face"}.
Turn the single screen into a multi-section scroll site. When the user scrolls, animate the 3D {argument name="scroll_object_en" default="watch"} to move and rotate to match each section. Keep the 3D canvas and background fixed while text scrolls over them; only the {argument name="moving_element_en" default="watch"} moves.
- fixed canvas: position the canvas
- Author: Harshith
- Source: Twitter Post
- Published: November 20, 2025
- Languages: en
A Japanese prompt asking Gemini to generate 40 simple, versatile SVG icons in light blue line style with download buttons for each icon.
Generate {argument name="icon_count_en" default="40"} simple and highly versatile icons in SVG format. Be sure to draw them on a canvas. The color should be {argument name="icon_color_en" default="light blue"}. The background should be {argument name="background_color_en" default="white"}. Use line-style icons with a Material Design feel. Add a button that lets users download each SVG icon as SVG data.
- Author: KAWAI
- Source: Twitter Post
- Published: November 20, 2025
- Languages: ja
An English prompt to design a colorful voxel-art pagoda garden environment as a fully self-contained HTML file that runs in Chrome, using any voxel/WebGL libraries.
Design a richly crafted voxel-art environment featuring an ornate {argument name="main_structure_en" default="pagoda"} set within a vibrant garden.
Include diverse vegetationโespecially {argument name="key_tree_en" default="cherry blossom trees"}โand make sure the composition feels lively, colorful, and visually striking.
Use any voxel or WebGL libraries you prefer, but deliver the entire project as a single, self-contained HTML file that I can paste and open directly in Chrome.
- Author: ่็ฝ๏ผๆฏๆฅๅนฒ่ดงๅไบซโ๏ผ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: en
A Chinese prompt for recreating a futuristic dark-mode cryptocurrency data dashboard with glassmorphism, neon accents, and sleek Inter typography.
Recreate a futuristic dark-mode cryptocurrency data dashboard: use a high-contrast obsidian background overall, glassmorphism UI elements with gradient outlines, slim Inter typography, and financial data visualizations with neon lighting effects.
- Author: ่็ฝ๏ผๆฏๆฅๅนฒ่ดงๅไบซโ๏ผ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
A prompt to design a precise mechanical SVG Ferris wheel animation with gears, linkages, and self-leveling cabins, output as HTML.
Design an SVG Ferris wheel mechanical animation that includes a main wheel, central bearings, support frame, drive gear set, and 8โ12 cabins with a reverse self-leveling mechanism to keep them horizontal. All linkages and gear transmissions must be precise, showing the complete mechanical drive chain and the reverse rotation logic that keeps cabins level. Output in HTML format.
- Author: ่็ฝ๏ผๆฏๆฅๅนฒ่ดงๅไบซโ๏ผ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
A long English-style prompt (in Chinese) asking Gemini to build a Chinese workout course site inspired by Keep, with responsive Tailwind CSS, course cards, filters, and matching video popups.
Build a Chinese โworkout courseโ webpage inspired by {argument name="reference_brand_en" default="Keep"}: include a full-screen athlete hero image, transparent sticky navigation, Chinese capsule-tag categories, and a grid of Chinese course cards. Each course card corresponds to a training program; use the course styles of creators such as {argument name="instructor_examples_en" default="Pamela, Zhou Liuye, Han Xiaosi, Jojie, Liu Genghong"} as references to design Chinese titles and labels for difficulty and duration. When clicking a card, open a course preview video player within the page that strictly matches the card information. Overall, keep the design minimal yet detailed, using smooth scrolling and hover animations, soft gradients, and refined typography to create a high-end Chinese experience. Use Tailwind CSS to implement a fully responsive layout.
- Author: ่็ฝ๏ผๆฏๆฅๅนฒ่ดงๅไบซโ๏ผ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
An English prompt to build a Keep-style Chinese fitness page with an interactive SVG body, filters, and exercise cards in a single HTML file using Tailwind and JavaScript.
Build a Keep-style Chinese fitness page with a clean, modern UI, an interactive SVG body (front/back views, where clicking a muscle highlights it), Chinese filters (gender/level/equipment), and exercise cards. Deliver it as a single HTML file using Tailwind + JavaScript.
- Author: ่็ฝ๏ผๆฏๆฅๅนฒ่ดงๅไบซโ๏ผ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: en
A short English prompt to create a neo-Brutalist style website with smooth scrolling animations, vibrant colors, and responsive Tailwind-based layout.
Create a neo-Brutalist website that pushes the boundaries of creativity. Include smooth scrolling animations, vibrant colors, and Tailwind ES styles, and make it fully responsive.
- Author: ่็ฝ๏ผๆฏๆฅๅนฒ่ดงๅไบซโ๏ผ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: en
An in-depth Chinese prompt describing a web camera app with a retro Polaroid feel, live webcam preview, developing photo animation, and draggable photos forming a collage wall.
Please develop a web camera application with a retro Polaroid atmosphere. Place a Polaroid camera model in the lower-left corner of the page that can call the userโs webcam and show a real-time preview. After clicking the shutter, a piece of โphoto paperโ slowly comes out from the top of the camera, and the photo develops from initially blurry to clear. Users can drag the fully developed photo from above the camera to any position on the page, creating a free layout similar to a โphoto collage wall.โ
- Author: ่็ฝ๏ผๆฏๆฅๅนฒ่ดงๅไบซโ๏ผ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
A prompt to build a Chinese web app where users record dreams each morning, collage sticker illustrations, and finally generate a tarot-like dream card.
Generate a Chinese web application where users record their dreams every morning. The app provides a large number of โillustration stickersโ to collage and reconstruct the dream scenes, and finally generates a dream card that looks like a tarot card.
- Author: ๆพๆๅ ๆฃฎ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
An English system-style prompt that has Gemini act as a world-class frontend engineer to build a Minecraft-like 3D brick builder app using React, Three.js, Tailwind CSS, and the Gemini API.
Act as a world-class frontend engineer and UX designer. Build a {argument name="application_name_en" default="3D Brick Builder Application"} using React, Three.js (@react-three/fiber), Tailwind CSS, and the Google Gemini API (@google/genai).
## Design philosophy
- **Theme**: {argument name="theme_en" default="White Base"} modern aesthetics.
- Author: ใใฃใจใณ | ใใธใฉใคใบ CEOใ้่ฆAIใใฅใผในใๆฏๆฅๆ้ใง็บไฟกโก๏ธใ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: en
A prompt for creating a Chinese-language web interface that analyzes MBTI types of tweets by calling a Gemini-powered backend, styled in Linearโs dark theme.
Create an X MBTI analyzer: the user pastes a tweet โ the frontend calls a backend (which uses the Gemini API to infer the MBTI type).
Design style should follow Linear, dark mode only, with a minimalist yet high-quality feel and refined details. The interface should be in Chinese.
- Author: katon
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
A Chinese prompt for Gemini 3 Pro to generate a Hanfu wardrobe web app where users upload a photo to virtually try on clothing from different dynasties and customize hairpins and makeup.
Generate a Chinese-language AI Hanfu wardrobe application. Use AI face-swapping or 3D body-shape technology. Users upload a photo and can, with one click, "try on" Hanfu from various dynasties ({argument name="dynasty_1_en" default="Tang"}, {argument name="dynasty_2_en" default="Song"}, {argument name="dynasty_3_en" default="Ming"}), or customize hairpins and makeup themselves.
- Author: ๆพๆๅ ๆฃฎ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
A Chinese prompt for Gemini 3 Pro to build a retro Motorola Fix Beeper style web typewriter that turns user input text into draggable cards with slow typing animation in the center of the page.
Please help me create a retro typewriter web application themed around {argument name="device_name_en" default="Motorola Fix Beeper"}. In the center of the page, display a typewriter that can take text entered by the user and generate it as a card, with an effect similar to the slow typing of a typewriter, and allow the card to be dragged to empty areas.
- Author: ่ขซๅๆฐ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
A prompt for Gemini 3 Pro to build a single-page Chinese web app that presents each star or constellation as a picture book, combining constellation lines with illustrated mythological characters.
Generate a {argument name="language_en" default="Chinese"} web application that tells the Greek mythological stories of each star or constellation in the form of a picture book. It should not only show the constellation connection lines but also display illustrations of the mythological characters.
- Author: ๆพๆๅ ๆฃฎ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
A Gemini prompt to create a cyber-themed Chinese web app that generates modern daily fortune cards tailored to young peopleโs lifestyles instead of traditional auspicious activities.
Generate a cyber-style Chinese web application for a modern daily almanac aimed at young people. Instead of traditional items like โsuitable for marriage,โ it should reflect modern life scenarios. For example: โToday do: {argument name="today_do_example_1_en" default="slack off at work"}, {argument name="today_do_example_2_en" default="drink milk tea"}, {argument name="today_do_example_3_en" default="confess your feelings"}; Today avoid: {argument name="today_avoid_example_1_en" default="meetings"}, {argument name="today_avoid_example_2_en" default="staying up all night"}, {argument name="today_avoid_example_3_en" default="making big resolutions"}.โ
- Author: ๆพๆๅ ๆฃฎ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
A Chinese prompt to build a retro Polaroid camera web app where a camera in the lower-left connects to the webcam, ejects developing photos, and lets users drag them to form a photo wall.
Please help me create a retro Polaroid-style camera web application. In the lower-left corner of the page, display a camera that can connect to the webcam and show a real-time preview. When the shutter button is clicked, a photo paper slowly pops out from the top of the camera, and the photo develops from blurry to clear. Users can drag the developed photo from above the camera to areas of the page outside the camera, creating a photo-wall-like effect.
- Author: ้ฅผๅนฒๅฅๅฅAGI๏ผ2.0๏ผ
- Source: Twitter Post
- Published: November 20, 2025
- Languages: zh
A playful prompt for Gemini 3.0 Pro to design a maximalist, shader-heavy, wavy gradient website for a fictional startup selling digital hot dogs, with beautifully animated hero text reacting to mouse hover.
Design a maximalist website for a fictional startup that sells {argument name="product_en" default="digital hot dogs"}. I want beautifully animated hero text with shaders that react to mouse hover, with lots of dithering and wavy, groovy gradients.
- Author: ben
- Source: Twitter Post
- Published: November 19, 2025
- Languages: en
A compact Three.js/GLSL prompt asking Gemini 3 Pro to create a raymarched Mandelbulb fractal surrounded by thousands of audio-reactive instanced particles with Unreal Bloom, all in a single file.
Create a raymarching Mandelbulb fractal in GLSL, surround it with {argument name="particle_count_en" default="12,000"} instanced particles, make it audio-reactive to my microphone, and add Unreal Bloom.
- Author: SRKDAN
- Source: Twitter Post
- Published: November 19, 2025
- Languages: en
A simple instruction-style prompt telling Gemini 3 Pro to act as the studio makemepulse and build a website for a client named AP, focusing on careful execution.
You are the studio {argument name="studio_name_en" default="makemepulse"}. Build me a website for {argument name="client_name_en" default="AP"}. Think carefully and make no mistakes.
- Author: arda
- Source: Twitter Post
- Published: November 19, 2025
- Languages: en
A concise technical prompt for Gemini 3 to build a web experience that uses the front camera and Mediapipe facial landmarks to apply a real-time 3D laser eye effect via Three.js.
Use the computer's front-facing camera.
Use the `mediapipe` library to capture facial landmarks.
Use `threejs` to apply a LASER EYE effect to the face captured by the camera, based on the real-time 3D landmark information provided by `mediapipe`.
- Author: ๆ็ณ | winterx
- Source: Twitter Post
- Published: November 19, 2025
- Languages: en
A prompt instructing Gemini 3 Pro to design a full survival-mode replica of Minecraft using Three.js for the web.
Design a full survival-mode replica of {argument name="game_name_en" default="Minecraft"} using {argument name="tech_stack_en" default="Three.js"}.
- Author: Vardhan Agnihotri
- Source: Twitter Post
- Published: November 19, 2025
- Languages: en
A prompt for Gemini to generate a professional enterprise landing page with responsive layout, animations, and interactive elements, adaptable to any business type.
Generate an enterprise-grade professional landing page for {argument name="business_description_en" default="[DESCRIBE YOUR BUSINESS]"}.
Add interactive elements, animations, and make it fully responsive.
Surprise me, be creative, and do this step by step.
- Author: God of Prompt
- Source: Twitter Post
- Published: November 19, 2025
- Languages: en
A Gemini prompt for building a web app that introduces one traditional Chinese color per day, alongside poetry and related illustrations or scenery.
Generate a web application that introduces one {argument name="color_system_en" default="traditional Chinese color"} each day (for example: {argument name="example_color_1_en" default="Sky Blue (Tianqing)"}, {argument name="example_color_2_en" default="Crimson (Yanzhi)"}, {argument name="example_color_3_en" default="Dark Green (Dailuo)"}), accompanied by related classical poetry, artifact illustrations, or natural landscapes.
- Author: ๆพๆๅ ๆฃฎ
- Source: Twitter Post
- Published: November 19, 2025
- Languages: zh
A Chinese prompt for Gemini-3.0-Pro to build a Word-style rich text editor as a single HTML file, including major formatting features and a download option.
Help me write a web-based rich text editor similar to {argument name="reference_software_en" default="Word"}, including its main features, with all code contained in a single HTML file. It should support: headings, paragraphs, left alignment, bold, italic, right alignment, underline, strikethrough, background color, text color, numbering, undo, and finally a download function. Directly generate the HTML, and thatโs it!
- Author: karminski-็ๅป
- Source: Twitter Post
- Published: November 19, 2025
- Languages: zh
A Chinese prompt that asks Gemini 3.0 to act as an aesthetically minded frontend engineer and build a dark, high-contrast, Tailwind-powered landing page for a Chinese Twitter KOL ranking site, fetching data from the web and outputting pure HTML with animations.
As a frontend engineer with a strong sense of aesthetics, create a website landing page for "{argument name="leaderboard_name_en" default="Chinese Twitter KOL Leaderboard"}". The design should be highly polished, with a minimal color palette, dark mode, and high contrast. Fetch the data source from the web.
Use Tailwind CSS + HTML5, add animations, and output only the HTML code directly.
- Author: katon
- Source: Twitter Post
- Published: November 19, 2025
- Languages: zh
A simple Gemini prompt to generate a web application for learning English through picture books, which can later be enhanced with interactive Gemini-powered lessons.
Create a picture book for learning {argument name="learning_subject" default="English"}.
- Author: ๆพๆๅ ๆฃฎ
- Source: Twitter Post
- Published: November 19, 2025
- Languages: zh
A short Chinese prompt for generating an animation demo of Douyinโs recommendation algorithm; while focused on animation, it can be used within a web context.
Create a demo animation of the recommendation algorithm for {argument name="platform_name_en" default="Douyin"}.
- Author: ๆพๆๅ ๆฃฎ
- Source: Twitter Post
- Published: November 19, 2025
- Languages: zh
A concise Gemini prompt to generate an animated multi-level gear linkage system where turning the main gear drives all others proportionally, suitable as an SVG or canvas demo.
A multi-level gear linkage system where rotating the main gear causes all gears to automatically move in proportion.
- Author: ไฝๆธฉ
- Source: Twitter Post
- Published: November 19, 2025
- Languages: zh
A prompt in English for generating a visually striking parallax scrolling website homepage with layered backgrounds, smooth transitions, and interactive animated elements.
Create a parallax scrolling website with explosive visual effects: layered backgrounds that move at different speeds when scrolling, smooth transitions, and interactive elements that pop up with animations.
- Author: ๆจๅญไธๅไปฃ็
- Source: Twitter Post
- Published: November 18, 2025
- Languages: en
An English prompt snippet that instructs Gemini to implement a high-end Swiss modern style landing page as a single-file React component, following strict specifications and using Swiss modern and bento layout concepts.
{argument name="generator_name_en" default="High-End LP Generator (Swiss Style)"}
Based on the later-provided {argument name="input_text_en" default="Input Text"}, strictly follow the following specifications and implement a top-quality landing page in React (single file).
1. Design concept: {argument name="design_concept_en" default="Swiss Modern & Bento"}
Style: {argument name="style_en" default="Apple/Linear-like Swiss modern style"}, removing decoration and using whitespace and typography to create appeal.
- Author: ใใฃใจใณ | ใใธใฉใคใบ CEOใ้่ฆAIใใฅใผในใๆฏๆฅๆ้ใง็บไฟกโก๏ธใ
- Source: Twitter Post
- Published: November 18, 2025
- Languages: en
A short style prompt describing a glassmorphism, neon-like ultra-modern UI with semi-transparent panels, suitable as guidance for a web interface layout.
A glassmorphism neon interface with ultra-modern UI lighting effects and semi-transparent glass panels.
- Author: ไฝๆธฉ
- Source: Twitter Post
- Published: November 18, 2025
- Languages: zh
A prompt instructing Gemini to use SVG to draw a dynamic, interactive demonstration of how Tesla FSD recognizes roads.
Use SVG to draw a dynamic, interactive demonstration of how Tesla FSD recognizes roads.
- Author: ไฝๆธฉ
- Source: Twitter Post
- Published: November 18, 2025
- Languages: zh
A very short prompt asking Gemini 3 Pro to build the GitHub page of React, effectively recreating the React repositoryโs GitHub UI.
Build the GitHub page of {argument name="project_name_en" default="React"}.
- Author: Samro โจ
- Source: Twitter Post
- Published: November 18, 2025
- Languages: en
An English prompt telling Gemini to create a walkable 3D museum scene in three.js with white walls and floor, glass ceiling, and blocks in the hall, like a small first-person web game environment.
Create a scene in three.js.
It will be a museum with paintings on the walls, and you can walk around the museum hall like in a game.
Make the walls and floor white.
There will be glass on the ceiling with the sky visible behind it.
Also place white blocks in the middle of the hall.
- Author: Sabo Sugi โฆ
- Source: Twitter Post
- Published: November 18, 2025
- Languages: en
A coding prompt for Gemini 3 to generate a Three.js voxel art scene inspired by an uploaded image, outputting all code as a single-page file.
I have provided an image. Code a beautiful voxel art scene inspired by this image. Write threejs code as a single-page
- Author: Ian Goodfellow
- Source: Twitter Post
- Published: November 18, 2025
- Languages: en
A prompt for a 3D Rubikโs cube web interface that can be mixed, auto-solved, and manually manipulated by the user.
Build a 3D {argument name="puzzle_type_en" default="Rubik's cube"}, include a button to mix it that shuffles the cube, and a button that solves the puzzle. Also allow the user to manually manipulate and solve it themselves.
- Author: Alan Shiflett
- Source: Twitter Post
- Published: November 18, 2025
- Languages: en
A very simple high-level prompt to design a CapCut-style video editing application interface using Gemini.
Design a {argument name="app_name_en" default="CapCut"}.
- Author: padphone
- Source: Twitter Post
- Published: November 18, 2025
- Languages: en
A direct prompt requesting a fully functional billiards game implemented inside a single HTML file.
Please create a fully functional {argument name="game_type_en" default="billiards game"} in a single HTML file.
- Author: ไฝๆธฉ
- Source: Twitter Post
- Published: November 18, 2025
- Languages: zh
A concise prompt asking Gemini to generate a single-page HTML (using SVG) that represents a โquantum core monitoring systemโ UI.
Please generate a single-page HTML (SVG) for a โ{argument name="system_name_en" default="Quantum Core Monitoring System"}โ.
- Author: ไฝๆธฉ
- Source: Twitter Post
- Published: November 18, 2025
- Languages: zh
A concise Chinese prompt telling Gemini to write a dynamic webpage that uses a cool point cloud effect to explain special relativity, ideal for interactive physics demos.
Write a dynamic webpage that uses a cool point-cloud style to explain special relativity.
- Author: -Zho-
- Source: Twitter Post
- Published: November 18, 2025
- Languages: zh
A prompt asking Gemini 3.0 Pro to create an SVG representation of an Xbox controller, suitable for embedding in web pages or UI components.
Create an SVG of an {argument name="controller_type_en" default="Xbox controller"}.
- Author: Singularity
- Source: Twitter Post
- Published: November 18, 2025
- Languages: en
A prompt for generating an extremely creative, neobrutalist, responsive webpage with smooth scroll animations, fancy colors, and Tailwind CSS, titled dorksense.
Make a neobrutalist webpage, make it extremely creative and push the limits as far as possible. Add smooth scroll animations, fancy colors, and Tailwind CSS styles. Make it fully responsive. The title of the page is {argument name="page_title" default="dorksense"}.
- Author: Dorksense
- Source: Twitter Post
- Published: November 13, 2025
- Languages: en
An English prompt asking Gemini to generate a highly detailed and creative SVG of the New York skyline that can be pasted as a single HTML file and opened in Chrome.
SVG of the {argument name="city_skyline_en" default="New York skyline"}. Use whatever libraries are needed to get this done, but make sure I can paste everything into a single HTML file and open it in Chrome. Make it interesting and highly detailed, showing unexpected details. Go fully creative and fully beautiful in one code block.
- Author: Chetaslua
- Source: Twitter Post
- Published: October 16, 2025
- Languages: en
We welcome contributions! You can submit prompts via:
- Click Submit New Prompt
- Fill in the form with prompt details and image
- Submit and wait for team review
- If approved (we'll add
approvedlabel), it will automatically sync to CMS - Your prompt will appear in README within 4 hours
Note: We only accept submissions via GitHub Issues to ensure quality control.
See CONTRIBUTING.md for detailed guidelines.
Licensed under CC BY 4.0.
๐ View in Web Gallery โข ๐ Submit a Prompt โข โญ Star this repo
๐ค This README is automatically generated. Last updated: 2026-02-09T05:25:15.490Z
For Tasks:
Click tags to check more tools for each tasksFor Jobs:
Alternative AI tools for awesome-gemini-3-prompts
Similar Open Source Tools
Stable-Diffusion
Stable Diffusion is a text-to-image AI model that can generate realistic images from a given text prompt. It is a powerful tool that can be used for a variety of creative and practical applications, such as generating concept art, creating illustrations, and designing products. Stable Diffusion is also a great tool for learning about AI and machine learning. This repository contains a collection of tutorials and resources on how to use Stable Diffusion.
ChatGPT-Next-Web
ChatGPT Next Web is a well-designed cross-platform ChatGPT web UI tool that supports Claude, GPT4, and Gemini Pro models. It allows users to deploy their private ChatGPT applications with ease. The tool offers features like one-click deployment, compact client for Linux/Windows/MacOS, compatibility with self-deployed LLMs, privacy-first approach with local data storage, markdown support, responsive design, fast loading speed, prompt templates, awesome prompts, chat history compression, multilingual support, and more.
Lumina-Note
Lumina Note is a local-first AI note-taking app designed to help users write, connect, and evolve knowledge with AI capabilities while ensuring data ownership. It offers a knowledge-centered workflow with features like Markdown editor, WikiLinks, and graph view. The app includes AI workspace modes such as Chat, Agent, Deep Research, and Codex, along with support for multiple model providers. Users can benefit from bidirectional links, LaTeX support, graph visualization, PDF reader with annotations, real-time voice input, and plugin ecosystem for extended functionalities. Lumina Note is built on Tauri v2 framework with a tech stack including React 18, TypeScript, Tailwind CSS, and SQLite for vector storage.
Awesome-Lists
Awesome-Lists is a curated list of awesome lists across various domains of computer science and beyond, including programming languages, web development, data science, and more. It provides a comprehensive index of articles, books, courses, open source projects, and other resources. The lists are organized by topic and subtopic, making it easy to find the information you need. Awesome-Lists is a valuable resource for anyone looking to learn more about a particular topic or to stay up-to-date on the latest developments in the field.
magic-resume
Magic Resume is a modern online resume editor that makes creating professional resumes simple and fun. Built on Next.js and Framer Motion, it supports real-time preview and custom themes. Features include Next.js 14+ based construction, smooth animation effects (Framer Motion), custom theme support, responsive design, dark mode, export to PDF, real-time preview, auto-save, and local storage. The technology stack includes Next.js 14+, TypeScript, Framer Motion, Tailwind CSS, Shadcn/ui, and Lucide Icons.
BrowserAI
BrowserAI is a production-ready tool that allows users to run AI models directly in the browser, offering simplicity, speed, privacy, and open-source capabilities. It provides WebGPU acceleration for fast inference, zero server costs, offline capability, and developer-friendly features. Perfect for web developers, companies seeking privacy-conscious AI solutions, researchers experimenting with browser-based AI, and hobbyists exploring AI without infrastructure overhead. The tool supports various AI tasks like text generation, speech recognition, and text-to-speech, with pre-configured popular models ready to use. It offers a simple SDK with multiple engine support and seamless switching between MLC and Transformers engines.
ShitCodify
ShitCodify is an AI-powered tool that transforms normal, readable, and maintainable code into hard-to-understand, hard-to-maintain 'shit code'. It uses large language models like GPT-4 to analyze code and apply various 'anti-patterns' and bad practices to reduce code readability and maintainability while keeping the code functional.
azooKey-Desktop
azooKey-Desktop is an open-source Japanese input system for macOS that incorporates the high-precision neural kana-kanji conversion engine 'Zenzai'. It offers features such as neural kana-kanji conversion, profile prompt, history learning, user dictionary, integration with personal optimization system 'Tuner', 'nice feeling conversion' with LLM, live conversion, and native support for AZIK. The tool is currently in alpha version, and its operation is not guaranteed. Users can install it via `.pkg` file or Homebrew. Development contributions are welcome, and the project has received support from the Information-technology Promotion Agency, Japan (IPA) for the 2024 fiscal year's untapped IT human resources discovery and nurturing project.
Imagine_AI
IMAGINE - AI is a groundbreaking image generator tool that leverages the power of OpenAI's DALL-E 2 API library to create extraordinary visuals. Developed using Node.js and Express, this tool offers a transformative way to unleash artistic creativity and imagination by generating unique and captivating images through simple prompts or keywords.
HyperChat
HyperChat is an open Chat client that utilizes various LLM APIs to enhance the Chat experience and offer productivity tools through the MCP protocol. It supports multiple LLMs like OpenAI, Claude, Qwen, Deepseek, GLM, Ollama. The platform includes a built-in MCP plugin market for easy installation and also allows manual installation of third-party MCPs. Features include Windows and MacOS support, resource support, tools support, English and Chinese language support, built-in MCP client 'hypertools', 'fetch' + 'search', Bot support, Artifacts rendering, KaTeX for mathematical formulas, WebDAV synchronization, and a MCP plugin market. Future plans include permission pop-up, scheduled tasks support, Projects + RAG support, tools implementation by LLM, and a local shell + nodejs + js on web runtime environment.
Edit-Banana
Edit Banana is a universal content re-editor that allows users to transform fixed content into fully manipulatable assets. Powered by SAM 3 and multimodal large models, it enables high-fidelity reconstruction while preserving original diagram details and logical relationships. The platform offers advanced segmentation, fixed multi-round VLM scanning, high-quality OCR, user system with credits, multi-user concurrency, and a web interface. Users can upload images or PDFs to get editable DrawIO (XML) or PPTX files in seconds. The project structure includes components for segmentation, text extraction, frontend, models, and scripts, with detailed installation and setup instructions provided. The tool is open-source under the Apache License 2.0, allowing commercial use and secondary development.
gemini-voyager
Gemini Voyager is a browser extension designed to enhance the user experience of Google Gemini by providing features such as folder organization, prompt vault, cloud sync, timeline navigation, chat export, Mermaid rendering, markdown fixing, and various power tools. It aims to help users keep their AI conversations organized, accessible, and productive. The extension is available on multiple browsers and supports manual installation and development builds. Users can also support the project by buying the developer a coffee or sponsoring via different platforms.
ai-paint-today-BE
AI Paint Today is an API server repository that allows users to record their emotions and daily experiences, and based on that, AI generates a beautiful picture diary of their day. The project includes features such as generating picture diaries from written entries, utilizing DALL-E 2 model for image generation, and deploying on AWS and Cloudflare. The project also follows specific conventions and collaboration strategies for development.
aigc-platform-server
This project aims to integrate mainstream open-source large models to achieve the coordination and cooperation between different types of large models, providing comprehensive and flexible AI content generation services.
agentscope
AgentScope is an agent-oriented programming tool for building LLM (Large Language Model) applications. It provides transparent development, realtime steering, agentic tools management, model agnostic programming, LEGO-style agent building, multi-agent support, and high customizability. The tool supports async invocation, reasoning models, streaming returns, async/sync tool functions, user interruption, group-wise tools management, streamable transport, stateful/stateless mode MCP client, distributed and parallel evaluation, multi-agent conversation management, and fine-grained MCP control. AgentScope Studio enables tracing and visualization of agent applications. The tool is highly customizable and encourages customization at various levels.























































