AI Prompt Library

Clean case-style visual layout. Structured grid composition, subtle abstract data cues (lines, blocks, rhythm), neutral professional color palette, soft lighting with no dramatic contrast, designed to feel credible, calm, and serious, like a consulting-grade presentation cover.

{\n"prompt": "Convert the uploaded image into a soft handcrafted paper-cut layered illustration style inspired by premium papercraft diorama aesthetics. Preserve the exact same person, pose, long wavy hair, clothing details, garden environment, trees, leaves, flowers, pathways, and classical columns from the uploaded image, but reinterpret everything as a cozy handcrafted layered paper artwork. Use smooth rounded shapes, simplified cute proportions, minimal facial features with tiny dot eyes, soft blush cheeks, and charming storybook-style expressions. Create stacked paper layers with strong dimensional depth, visible separation between layers, soft realistic shadows, and clean laser-cut paper edges. Add a distinct thick white outer outline around the main character like a realistic sticker border or layered cut-paper backing, clearly separating the subject from the background. The white border must appear like a real physical paper layer, not a glow effect. Use matte tactile textures resembling thick cardstock, craft foam, and premium colored paper. Apply a soft pastel color palette using muted greens, warm creams, light browns, beige tones, dusty pinks, and gentle neutral shades. Lighting should be soft, diffused, even, and calming, enhancing the layered-paper dimensional look without harsh contrast. Keep the environment whimsical, wholesome, elegant, playful, and visually soothing like a luxury children’s storybook illustration or handcrafted decorative paper art scene. Ultra detailed layered papercraft diorama, realistic paper texture, cozy aesthetic, premium handmade illustration style.",\n\n"negative_prompt": "photorealistic skin, sharp realistic eyes, harsh shadows, neon colors, glossy plastic texture, metallic surfaces, blurry details, extra people, distorted hands, text, watermark, low quality, messy edges, dark horror mood, realistic facial anatomy, overexposed lighting",\n\n"style": "handcrafted papercut diorama illustration",\n"aspect_ratio": "4:5",\n"quality": "high",\n"lighting": "soft diffused lighting",\n"texture": "matte layered cardstock paper",\n"color_palette": "soft pastel muted tones",\n"outline": "thick white layered paper border"\n}

Ultra-wide-angle hyper-realistic top-down flat lay photography. The central focus is a custom physical conference table meticulously shaped like {argument name="Shape" default="[SHAPE/NUMBER/SYMBOL]"}. The table features {argument name="Material" default="[MATERIAL/TEXTURE/COLOR]"} and placed on a [FLOOR COLOR] floor, creating significant negative space around it for a clean, minimalist composition. A group of {argument name="Number of People" default="[NUMBER]"} real people are sitting around the edges of this [SHAPE]-shaped table. [POSITIONING: e.g., 3 on each side, 1 top/bottom] Actions: [DESCRIBE PEOPLE'S ACTIONS, e.g., working on laptops, drinking coffee]. Lighting & Quality: Professional studio lighting with crisp, distinct drop shadows that define the [SHAPE] silhouette on the floor. 8k resolution, sharp focus, cinematic commercial aesthetic.

{argument name="Character Name" default="Mia"} Medium shot of the 3D character gently reaching out to turn off a transparent floating interface panel, other panels fading softly, warm light slowly replacing cool tones, calm, focused expression, modern minimal environment, emotional realism, clean composition

Hyper-realistic liquid metallic chrome graphic with perfect reflective surfaces, fluid mercury-like distortions, high-gloss specular highlights, smooth gradients and dynamic liquid metal flow. Ultra-modern luxury product and tech branding aesthetic.

Create a 90s version of my website.

Given that the visual goal is completely determined, ask the AI to reverse-engineer a reusable prompt template. This prompt must support custom variables for subsequent long-term use.

Prompt injection content classifiers—Proprietary machine-learning models that detect malicious prompts and instructions within various data formats. Security thought reinforcement—Targeted security instructions that are added around the prompt content. These instructions remind the LLM (large language model) to perform the user-directed task and ignore adversarial instructions. Markdown sanitization and suspicious URL redaction—Identifying and redacting external image URLs and suspicious links using Google Safe Browsing to prevent URL-based attacks and data exfiltration. User confirmation framework—A contextual system that requires explicit user confirmation for potentially risky operations, such as deleting calendar events. End-user security mitigation notifications—Contextual information provided to users when security issues are detected and mitigated. These notifications encourage users to learn more via dedicated help center articles. Model resilience—The adversarial robustness of Gemini models, which protects them from explicit malicious manipulation.

3d squishy kawaii letters from {argument name="letters" default="A to Z"}, {argument name="theme" default="candyland"} theme, glossy pastel colors, soft rounded shapes like marshmallows and gummies, each letter made of different sweets such as lollipops, jelly beans, cotton candy, and chocolate, arranged in alphabetical order on a {argument name="background" default="plain blank white background"}, soft studio lighting, high-resolution, ultra-cute and whimsical style

the prompt help nano banana strictly maintain the exact layout, wall positions, door swing, shelving placement, toilet and sink position, and dimensions as shown. No structural or proportional changes

Apply an ASCII-art rendering effect to the reference image, where the ASCII character density and placement is driven by the underlying tonal contrast and edges of the photo, like a true edge-aware ASCII-art conversion. Dense character clusters follow the natural contours, silhouette lines, and high-detail areas (such as hair strands, fabric folds, and background structure), while flatter, low-contrast areas (such as smooth skin) stay comparatively clean with minimal or no texture. The effect should look like the ASCII characters are tracing the actual shapes and edges present in the photo, not a uniform overlay pasted on top of everything equally. Use a classic {argument name="color scheme" default="green-on-black or grayscale"} terminal character set ({argument name="character set" default="@ # % & * + . :"}), {argument name="density" default="medium"} overall density, with the original photo still legible underneath.

2D editorial illustration inspired by Egyptian North Coast summer, unveiled {argument name="ethnicity" default="Middle Eastern woman"}, wide-leg breezy trousers in {argument name="trouser color" default="sandy beige"} with hand-drawn wave prints, embroidered blouse with {argument name="blouse motif" default="palm frond motifs"}, warm sun-drenched palette, minimal background suggesting sea and dune, elegant relaxed pose, 3:4

Recreating a copy of Airbnb using a self-made Image-to-Text model to generate a prompt for Nano Banana Pro.

You are an expert-level UI/UX presentation designer. Generate high-fidelity, futuristic 16:9 presentation slides. Based on visual balance aesthetics, automatically choose the most perfect composition among a cover, grid layout, or data visualization. For the global visual language, the style should seamlessly blend Apple Keynote minimalism, modern SaaS product design, and glassmorphism. The overall mood should feel premium, immersive, clean, and breathable. Use cinematic volumetric lighting, soft ray-traced reflections, and ambient occlusion. For the color scheme, choose either deep void black or pure ceramic white as the base, accented with flowing aurora gradients such as neon purple, electric blue, soft coral orange, and cyan for the background and UI highlights. For the content modules on each slide, intelligently integrate the following elements: 1. Use a bento grid system as the layout engine, organizing content inside modular rounded-rectangle containers. The containers must be frosted glass with blur, featuring delicate white edges and soft shadows, and strictly preserve generous internal whitespace to avoid crowding. 2. Insert gift-like 3D objects, rendering unique, high-end abstract 3D artifacts as visual anchors. They should look like tangible, expensive gifts or collectibles, with materials such as polished metal, iridescent acrylic, transparent glass, or soft silicone, and shapes like floating capsules, spheres, shields, Möbius strips, or fluid waves. 3. For typography and data, use clean sans-serif fonts with high contrast. If there are charts, use glowing 3D donut charts, capsule-shaped progress bars, or floating numbers; the charts should look like glowing neon toys. Composition logic reference: If generating a cover, place a huge, complex 3D glass object in the center with bold large text over it and aurora waves stretching in the background. If generating a content slide, use the bento grid layout with 3D icons on small cards and text on large cards. If generating a data slide, use a split-screen design: typography on the left, and a large, glowing 3D data visualization chart floating on the right. Rendering quality requirements: Unreal Engine 5 rendering, 8K resolution, ultra-detailed textures, strong UI design sense, UX interface, Dribbble trending style, award-winning design quality.

"semi-transparent 3D wireframe grid overlay in {argument name="grid color 1" default="blue"} horizontal lines and {argument name="grid color 2" default="red"} vertical lines to show volume, like a digital modeling sketch." + "photo faded to 50% opacity for emphasis on the grid" + "show only the grid, white bg"

Make a beautiful 9:16 wallpaper. Leave natural negative space near the top and bottom. Give the wallpaper a subtle top to bottom gradient. No blur effects or cropping. Keep the scene simple. Put your subject and style at the end of the prompt.

A highly realistic, photorealistic close-up macro photography of a used, physically weathered {argument name="bank name" default="JP Morgan Chase"} bank check resting on a scratched wooden desk. The check is completely filled out with a blue ballpoint pen. The handwriting is distinctly human, messy, slightly slanted, and uniquely stylized with inconsistent ink pressure and minor skips, completely avoiding any uniform or computerized script-font appearance. The date written is '{argument name="date" default="May 1, 2026"}'. The signature on the bottom right clearly reads '{argument name="signature" default="iqrasaifiii"}' in a fast, authentic, slightly erratic cursive scrawl. The paper of the check shows significant physical wear and tear: noticeable horizontal and vertical folding creases, bent and frayed corners, faint dirt smudges, and subtle, faded coffee spots. A cheap black ballpoint pen lies slightly out of focus in the background. Shot on DSLR, 50mm lens, natural daylight casting soft, realistic shadows, 8k resolution.

A screenshot of a winamp style software called {argument name="software name" default="Nano Banana Pro"}

"Platform": " {argument name="platform" default="WeChat"}", "Left_Person": " {argument name="left person" default="Enter name or upload reference image"}", "Right_Person": "{argument name="right person" default="Enter name or upload reference image"}", "Message_Text": "{argument name="message text" default="Enter text"}"

Extreme macro photograph of the {argument name="monitor area" default="bottom-left corner of a computer monitor"}, shot at a slight angle from the right. The sheet tab bar of a spreadsheet fills the frame — three tabs legible and sharp, reading "{argument name="tab 1" default="Quotes"}", "{argument name="tab 2" default="Copy of Invoices"}", and "FINAL_dont_delete", with "FINAL_dont_delete" the active white tab. Visible pixel structure in the screen surface at this magnification, a fine layer of dust, one small fingerprint smudge on the glass. Very shallow depth of field so the tabs at the left edge blur out. Soft window light from the side raking across the screen surface. Natural colour, macro photography, no device bezel visible beyond the lower edge.

{ "scene": "AI engineer working at futuristic desk with multiple holographic displays", "camera": { "angle": "{argument name="camera angle" default="over-the-shoulder, slightly elevated"}", "focal_length": "{argument name="focal length" default="50mm portrait lens"}", "depth_of_field": "{argument name="depth of field" default="shallow, subject in sharp focus, background slightly blurred"}" }, "lighting": { "type": "{argument name="lighting type" default="dramatic side lighting with blue accent lights"}", "time_of_day": "{argument name="time of day" default="late night"}", "mood": "{argument name="mood" default="focused, tech-noir atmosphere"}", "shadows": "deep but not harsh" }, "color_grade": "{argument name="color grade" default="cool tones, teal and orange color scheme"}" }

Subject: image1 - use the uploaded photo as the exact face reference (100% facial accuracy). Do not change facial features, proportions, skin texture, or expression. Keep my natural dark facial tone, no extra brightness or whitening. (Use face reference exactly.) A dreamlike, movie-quality augmented-reality street scene. I am {argument name="pose" default="leaning casually against a streetlight on an Indian city street, wearing an open yellow hoodie, black jeans, and white sneakers, holding a phone and looking at its screen"}. Around me, multiple floating Spotify/Apple-Music style player cards orbit in full 3D space. Some cards appear in front, partially overlapping my body, while others float behind and beside me at varying depths. The UI cards have rounded corners, slight perspective tilt, translucent frosted-glass look, soft glowing edges, playback bars, and album art featuring songs by {argument name="artist" default="[Artist Name]"}. Lighting is {argument name="lighting" default="cool-to-neutral with preserved shadows, subtle cinematic bloom, and natural interaction between Ul glow and street lighting"}. Foreground cards are sharp; distant cards softly blurred for depth. Medium-high contrast, realistic motion accents. Photorealistic subject, central composition, seamless blend of real environment and AR visuals.

Vertical social media story design (9:16 aspect ratio), beauty vlogger Christmas makeup tutorial interface. Background is a soft, high-end pink gradient, creating a sweet and dreamy atmosphere. Core Visual: Character Consistency (Highest Weight): The center of the image features an Asian female beauty vlogger. Facial features, features, and hairstyle must strictly and perfectly replicate the person in the uploaded reference image. Makeup and Accessories: Extremely delicate “Christmas Reindeer Sweet Cool Makeup.” Eye makeup is red-brown blending, with white fawn spots on the face, wearing a fluffy reindeer antler headband. Red Pen Markings and Step Correspondence (Core Logic): On the face of the finished character on the right, use a striking red ink pen to draw 5 indicator lines, pointing to 5 key parts of the makeup, and label them sequentially in Chinese: 1. Eyeshadow; 2. Gold Glitter; 3. Eyelashes; 4. Reindeer Blush; 5. Full Red Lips. Layout: Top: Chinese title “Christmas Makeup Tutorial,” paired with a lipstick icon. Subtitle “Reindeer Girl Makeup Sweet Cool Style.” Middle (Comparison Area): Split-screen design. The left side is the character “Bare Face,” the right side is “Finished Makeup.” Connected by a dashed line and a scissor icon in the middle. Bottom (Strongly Associated Step Preview Area): Set a horizontally scrolling rounded card bar, displaying 5 step images that strictly correspond to the facial markings: Card 1: Close-up of applying eyeshadow to the eye area; Card 2: Close-up of dotting gold glitter with a finger; Card 3: Close-up of curling or brushing eyelashes; Card 4: Close-up of sweeping blush and drawing fawn spots on the cheek; Card 5: Close-up of lips applying red lipstick. Each card's top-left corner must clearly label numbers 1-5, corresponding to the red pen markings on the face. Bottom (Interaction Area): “Swipe up for full tutorial” arrow, “Click to view product” button, like and save icons. Style: High-definition commercial photography texture combined with UI design, transparent light and shadow, vivid colors, delicate and realistic skin texture, fashion magazine layout style. --ar 9:16

Make the character in the image hold their passport while sitting in front of a computer, showing the ID clearly to the camera to prove they are a real person. The ID information should be clearly visible, with the name {argument name="name_en" default="[NAME]"}, the affiliation {argument name="affiliation_en" default="[YOUR DATA]"}, and the birthday {argument name="birthday_en" default="[YOUR DATE]"}. All other details can be random and fictional.