GPT Image 2 prompts

Create a {argument name="aspect ratio" default="16:9"} energetic {argument name="subject" default="villain-duel"} cinematic storyboard sheet image.

Create a premium, modern {argument name="style" default="high-fashion"} {argument name="type" default="character concept art sheet"} in a {argument name="layout" default="16:9 widescreen layout"} on a pure white background, following this exact structured composition and visual aesthetic: [STYLE]: high-contrast digital anime illustration with clean,

Premium product shot of a {argument name="product name" default="iPhone 17 Pro Max"} floating diagonally above a glossy white circular pedestal in a minimal {argument name="background color" default="pale mint green and warm cream"} studio environment. Show exactly one smartphone, front-facing at a three-quarter angle, tilted slightly left, with a slim black bezel, metallic dark side rails, visible side buttons, bottom speaker holes, charging port, and a pill-shaped dynamic-island cutout at the top. The screen displays an elegant green abstract wallpaper with soft translucent oval shapes and centered white text reading “{argument name="screen text" default="iPhone 17 Pro Max"}”; include small lock-screen icons near the bottom. Use soft studio lighting, smooth gradients, a large curved mint backdrop, a creamy vertical light area on the right, realistic glass reflections, subtle highlights along the metal frame, and a diffused shadow directly beneath the floating phone on the glossy surface. Hyper-realistic textures, commercial advertising style, ultra-clean composition, wide landscape crop, no extra objects, no hands, no watermark.

Create an ultra-cute high-end 3D chibi-style portrait of two young people, {argument name="girl description" default="a fashionable young girl with warm fair skin, huge expressive brown eyes, a high voluminous brown updo with loose curled strands, and a cluster of small white daisy flowers tucked into her hair"} and {argument name="boy description" default="a fashionable young boy with warm fair skin, huge expressive brown eyes, tousled fluffy dark brown hair, and a small black earring"}, sitting on separate modern round stools and facing each other in a playful mirrored pose. Count exactly 2 characters: 1 girl on the left and 1 boy on the right. Count exactly 2 stools: round caramel-brown cushioned seats with slim gold metal legs and circular footrests. Both characters rest their chins on one hand, lean slightly forward, smile softly, and gaze into each other's eyes through stylish small black sunglasses. Dress them in coordinated burgundy and black outfits: the girl wears a burgundy pinstriped long-sleeve blouse tucked into wide-leg black trousers with chunky white sneakers and small dangling earrings; the boy wears a burgundy short-sleeve collared shirt, loose wide-leg black trousers, chunky white sneakers, and a small earring. Use exaggerated chibi proportions with oversized heads, tiny noses, soft rounded cheeks, delicate lips, glossy eyes visible behind tinted lenses, and realistic Pixar-like 3D rendering. Place them in {argument name="setting" default="a warm sunlit indoor garden cafe or studio conservatory"} with tall grid windows on the left, golden afternoon sunlight streaming in, soft reflections on a polished concrete floor, hanging green vines with glowing fairy lights overhead, potted plants, pink flowers, a wooden shelf with plants, and a small black pendant lamp in the background. The composition is vertical 4:5, eye-level, centered on the two figures with symmetrical spacing, shallow depth of field, creamy bokeh, warm amber color grading, cinematic soft lighting, detailed fabric folds, smooth toy-like skin, and no visible text, logos, or watermark.

Goal: Create a cinematic side-by-side comparison frame showing how different grading constraints affect the same nighttime romance scene in New York City.

Canvas: Very wide horizontal comparison image, approximately 4:1 aspect ratio, with a thin black outer border and a black divider between two equal panoramic panels.

Layout: Use exactly 2 side-by-side panels. The left panel is labeled at the top-left: “A  No skill: generic NYC romance grade” with a smaller subtitle beneath it: “normal saturation / brighter sky volume / modern night polish”. The right panel is labeled at the top-left: “B  With eterna-image2image: standard ETERNA” with a smaller subtitle beneath it: “muted night color / soft highlights / deeper readable shadows”. Keep the labels small, clean, white, and unobtrusive, like a film test contact sheet.

Scene content: Both panels show the same setup: exactly 2 human subjects, a woman on the left and a man on the right, standing face-to-face on a rainy New York street at night. The woman has {argument name="woman appearance" default="dark shoulder-length hair with bangs, light beige wool coat, soft natural makeup"}. The man has {argument name="man appearance" default="curly dark hair, glasses, navy suit jacket, white shirt"}. They stand in profile, emotionally restrained, as if in a quiet romantic conversation. The street is wet and reflective, with blurred traffic, storefronts, streetlights, building windows, and a yellow taxi in the distance. Include visible steam or mist rising from the street in the background.

Left panel look: A generic modern NYC romance grade, with brighter lifted shadows, slightly exaggerated cool moonlit highlights on faces and coats, higher saturation, glossy wet-road reflections, visible bokeh lights, and a more polished digital night look. The woman and man are placed closer to the foreground, cropped at mid torso, with the woman near the left third and the man near the right third.

Right panel look: A more restrained Fuji ETERNA-inspired cinematic grade, using muted colors, soft highlights, deeper but still readable shadows, lower saturation, natural practical streetlight warmth, less artificial blue spill, smoother tonal rolloff, and a more grounded filmic night atmosphere. The composition is slightly wider than the left panel, with more negative space and more visible sidewalk, columns, taxi, steam, and wet street environment. The woman and man remain facing each other in profile, centered in the panel.

Visual style: Photorealistic cinematic still, anamorphic-feeling wide frame, natural rain-slick pavement, shallow depth of field, realistic skin tones, soft practical lighting, subtle film grain, restrained contrast, no fantasy elements.

Constraints: Use exactly 2 panels, exactly 2 people per panel, exactly the 2 panel labels and 2 subtitles listed above, no extra captions, no logos, no watermark, no UI controls, no additional characters in the foreground.

Create a cinematic vertical close-up portrait of {argument name="character description" default="a young East Asian woman styled as an ethereal ancient Chinese beauty"} at night, framed from the chest upward and slightly off-center to the right. She has pale porcelain skin, delicate features, soft red lips, downcast eyes, and a calm melancholic expression, with one hand raised lightly to her chin showing long pale manicured nails. Her {argument name="hair color" default="jet-black"} hair is long, voluminous, and loosely pinned up in a traditional-inspired style, with wispy strands falling across her face. Add one small red plum-blossom forehead ornament centered above her brows. She wears {argument name="robe color and material" default="a shimmering silver-white silk hanfu robe"} with subtle floral jacquard embroidery, layered lapels, and a soft white inner garment. The background is very dark and blurred, with exactly one visible branch area of soft-focus white blossoms on the left side, suggesting moonlit plum flowers. Use cool blue moonlight, low-key lighting, glossy highlights on the silk, soft shadow modeling on the face, shallow depth of field, realistic skin texture, high-fashion fantasy photography, elegant Chinese historical drama aesthetic, vertical 9:16 composition, no text, no watermark, no modern accessories.

Goal: Create a colorful 8-panel storyboard sheet for an animated sequence featuring {argument name="character name" default="Souichirock"}, a chibi punk-rock boy performing an intense invisible air-guitar solo and ending with a dramatic guitar-smash pose.

Canvas: Wide horizontal storyboard image, clean white background, divided into exactly 8 equal rectangular panels arranged in a 2-row by 4-column grid. Use thin light-gray panel borders. Each panel has a small circled black number in the upper-left corner, labeled exactly 1 through 8.

Visual style: Hand-drawn digital illustration with bold black ink outlines, exaggerated manga/chibi proportions, dynamic comic action poses, expressive angry rock-performance face, and high-energy paint-splash effects. Use vivid splashes and sweeping arcs of cyan, magenta, purple, orange, green, blue, and black around the character to visualize the sound and motion of the invisible guitar. Keep the background mostly white and uncluttered.

Character details: The same chibi male character appears in all 8 panels. He has short spiky {argument name="hair color" default="black"} hair, pale skin, thick black eyebrows, an open shouting mouth with red interior, and an intense scowling expression. Outfit: short-sleeve cream shirt covered in colorful paint-like marks, black pants, black shoes, black studded wristbands, and a small punk-rock attitude. He does not hold a visible guitar; his hands mime playing and smashing an air guitar.

Panel-by-panel action sequence: Include exactly 8 numbered panels. Panel 1: wide-legged stance, left hand low as if gripping the neck of an invisible guitar, right hand clawing near the body, shouting, with bright paint bursts behind him and blue-magenta arcs sweeping to the right. Panel 2: deeper lunge, legs spread wider, both hands actively strumming and fretting an invisible guitar, a large blue and magenta sound-wave swirl blasting from his right side. Panel 3: centered rock stance, knees bent, left hand extended like a guitar neck, right hand near the torso, shouting toward the viewer, with dense multicolor splatter behind him. Panel 4: leaning far backward to the right, one hand raised in a claw shape and the other near his waist, legs stretched diagonally, with an orange-blue-magenta spiral sweeping around him. Panel 5: crouched low and leaning forward, one knee near the ground, hands in aggressive air-guitar position, surrounded by looping purple, pink, green, and blue paint trails. Panel 6: low lunging pose, body angled forward, one hand reaching down and the other clawed outward, with explosive colorful paint splashes radiating behind him. Panel 7: airborne jump pose, knees bent, one fist raised overhead, mouth wide open in a scream, with vertical motion lines and a rainbow paint arc curving underneath and behind him. Panel 8: final guitar-crash climax, character bent forward with both hands driving downward as if smashing the invisible guitar into the ground, black impact lines and a huge multicolor splatter explosion spreading outward.

Constraints: Use exactly 8 panels and exactly one character per panel. Preserve the small circled panel numbers 1, 2, 3, 4, 5, 6, 7, and 8. No captions other than the panel numbers. Do not add a real guitar, stage, audience, watermark, logo, or extra characters.

Goal: Create a black-and-white manga page about a fashionable young man meeting a tiny anthropomorphic hamster friend at a busy indoor shopping plaza.

Canvas: Vertical manga page, roughly 2:3 aspect ratio, clean white page with thin black outer border and thick black gutters. Highly detailed monochrome pen-and-ink line art with screentone-style shading, realistic architectural perspective, elegant shoujo/seinen manga character rendering.

Layout: Use exactly 6 panels. Panel 1 is a wide establishing shot across the full top width. Panel 2 is another wide panel below it. Panel 3 row has two panels: a narrow close-up of the man on the left and a wider hamster panel on the right. Panel 4 row has two panels: a narrow vertical hamster-hand gag panel on the left and a wider final panel on the right.

Setting: A modern indoor mall or station concourse with glossy tiled floors, tall glass storefronts, crowds of pedestrians, overhead direction signs, a circular indoor fountain, and a clothing store in the final panel. Include English signage such as “Central Plaza” on a directional sign and “SELECTED MEN'S WEAR” above the clothing shop, plus a small “NEW ARRIVAL” sign near the shop entrance.

Characters: The main human character is {argument name="male friend description" default="a tall handsome young man with tousled wavy dark hair, sharp eyes, a calm aloof expression, wearing a black long coat over a black turtleneck with a long scarf and pendant necklace"}. The small companion is {argument name="hamster friend description" default="a round fluffy hamster wearing a tiny beret and bow tie, holding or standing near a smartphone, expressive and cute"}.

Panel details: Panel 1: Wide establishing shot of the shopping plaza fountain from a distance. Many pedestrians walk around. The tiny hamster sits on the rim of the circular fountain, looking down at a smartphone, with small sound text near it. Panel 2: Medium-wide shot showing the handsome man in the left foreground, facing slightly right, noticing something; the fountain with the tiny hamster appears on the right side in the background. Panel 3 left: Tight close-up of the man’s face, serious and slightly puzzled. Panel 3 right: Close shot of the hamster on the fountain rim, waving one paw energetically while standing on or beside the smartphone. Panel 4 left: Comedic close-up of a human hand pinching and lifting the hamster gently by the tiny beret; the hamster dangles with startled paws. Panel 4 right: Final medium-wide shot in front of the menswear shop. The man stands in profile, holding the hamster on his open palm while looking toward the store window with suits and racks of clothes.

Visible text and effects: Include exactly 8 text elements: 1) the overhead sign “Central Plaza”; 2) small Japanese sound effect {argument name="phone tapping sound" default="てちてち"} near the hamster in panel 1; 3) speech bubble from the man in panel 2 reading {argument name="man first line" default="ん？"}; 4) silent speech bubble in panel 3 left reading “……” ; 5) hamster speech bubble in panel 3 right reading {argument name="hamster greeting" default="あ、こっちだよー"}; 6) large Japanese sound effect in panel 4 left reading “わしっ”; 7) small hamster speech bubble in panel 4 left reading “え？”; 8) final speech bubbles in panel 4 right: the man says “……ハムスターに似合う服を見に行くから服屋から回るぞ” and the hamster replies “えぇ…………”.

Style constraints: Do not use color. Keep the page polished and cinematic, with delicate hatching, realistic crowd silhouettes, crisp panel borders, detailed clothing folds, and expressive manga faces. The hamster should remain tiny compared with the man, creating a humorous contrast. Avoid extra panels, extra dialogue, watermarks, or modern UI overlays.

Goal: Create a dark, highly detailed Japanese ukiyo-e style poster featuring a skeletal salaryman as a grim motivational figure, combining Edo-period woodblock print texture with modern corporate burnout satire.

Canvas: Vertical poster, roughly 4:5 aspect ratio, aged parchment background with heavy black border, weathered paper stains, ink splatter, scratches, distressed print grain, muted tan, black, charcoal, bone white, and deep crimson palette.

Layout: Split the image visually into two zones. On the left, place a tall rectangular parchment calligraphy panel occupying about one quarter of the width, bordered with thin black ornamental lines. On the right and center, place the main character large and dramatic, standing or perched on a rocky mound, framed by a huge black circular ink-brush halo behind the skull. Add waves, smoke-like clouds, and a small Mount Fuji landscape in the lower left background. Include exactly 3 main text areas: 1 large vertical Japanese quote panel on the left, 1 small English translation block near the bottom left, and 1 small handwritten label on the briefcase at bottom right.

Main subject: A full-body human skeleton wearing a tattered dark navy-black business suit or overcoat, ornate but decayed, with frayed sleeves, ripped fabric, exposed ribs and bones, red torn necktie ribbons blowing to the right, decorative gold trim, and small hanging talismans. The skull has hollow black eye sockets and a slightly open grinning jaw. The figure holds a vintage black and brass megaphone near its mouth in the left hand; the megaphone has a small vertical paper label reading 「社畜課」. The right skeletal hand hangs down holding a battered black leather briefcase with metal corners. A vertical name tag on the character’s chest reads 「終電厳守」. Add a small black cat with glowing yellow eyes partially hidden behind the briefcase near the lower right edge.

Text content: On the left calligraphy panel, write large expressive vertical Japanese brush calligraphy: 「{argument name="Japanese quote" default="道を選ばずして、何が人生だ。"}」. Beneath it in smaller vertical Japanese text: 「武者小路実篤　生きて行く私」, plus two red seal stamps. In the lower left corner, add a small serif English translation in off-white: “To live is to choose. To choose is to risk. To risk is to become.” followed by “Saneatsu Mushanokōji” and “Ikiteyuku Watashi”. On the briefcase label, write 「本日も苦労様でした」 in rough handwritten Japanese.

Style: Intricate linework, dramatic manga-meets-ukiyo-e illustration, antique woodblock print, gritty etching detail, rough ink outlines, high contrast shadows, asymmetrical composition, ornate decayed textures, solemn but darkly humorous mood. Make it look like an old Japanese illustrated proverb poster rather than a clean digital painting.

Constraints: Use exactly 1 skeleton character, 1 megaphone, 1 briefcase, 1 black cat, 1 Mount Fuji, 1 large ink-brush circle, 2 red seal stamps, and 3 main text areas. Keep all Japanese text vertical where specified, avoid modern logos or watermarks, and do not add extra characters.

Create a highly detailed vertical fantasy storybook illustration in a soft anime watercolor style: a magical desert café built into the side of an old crashed propeller airplane at twilight, with a glowing Milky Way sky overhead and warm lantern light spilling from the café counter. The setting is called {argument name="cafe name" default="STARDUST CAFÉ"}; show the name painted on the airplane fuselage and on a chalkboard sign. The airplane nose and large propeller dominate the left side, while the right wing extends overhead as a canopy covered with ivy, hanging bulbs, star ornaments, glass lanterns, and a pale scarf with gold stars draped from the wingtip. On the underside of the wing, write the small hand-lettered phrase “Come for Coffee, Stay for Wonder.” The café counter is packed with tiny details: brass coffee kettles, glass jars, teacups, baked cookies, cloches, bottles, shelves of ingredients, a vintage clock, coffee drippers, plants, and magical trinkets. Include exactly four main characters: 1) {argument name="girl character" default="Alice"}, a blonde young girl in a blue dress with a white apron, black shoes, white stockings, and a bow, standing at the left and looking curiously toward the café; 2) {argument name="rabbit character" default="the White Rabbit"}, an anxious white rabbit in a blue waistcoat and jacket, holding a pocket watch and looking hurried; 3) {argument name="cat character" default="orange cat waiter"}, a standing orange tabby cat in a small bow tie, blocking the rabbit’s path and gesturing insistently; 4) {argument name="barista character" default="a gentle brown-haired café girl"}, a smiling barista behind the counter with wavy hair, a cream blouse, star earrings, and a star-pattern apron, offering freshly brewed coffee and pastries. Make the rabbit look like he wants to say he has no time, while the cat waiter clearly refuses to let him leave, and Alice is enchanted by the coffee and baked sweets. Add exactly five readable sign areas: 1) airplane/fuselage logo “STARDUST CAFÉ,” 2) wing phrase “Come for Coffee, Stay for Wonder,” 3) main menu chalkboard listing “Milky Way Latte,” “Starry Night Mocha,” “Desert Rose Tea,” “Moonlight Cocoa,” “Comet Caramel Macchiato,” and “Wonderland Brew,” 4) wooden directional signposts reading “TEA PARTY,” “DESSERT,” and “THIS WAY,” and 5) front chalkboard reading “STARDUST CAFÉ / Coffee, Tea & Curious Adventures Await.” Surround the scene with desert cacti, roses, wildflowers, old suitcases, small crates, a welcome sign, potted plants, and a blue starry rug under the rabbit and cat. Use a warm golden-pink sunset horizon fading into deep violet-blue galaxy sky, with mountains in the distance, dust, sparkles, soft rim light, intricate linework, cozy fairy-tale atmosphere, 5k detail, no modern objects, no photorealism, no extra characters, and no watermark.

Create a vertical cyberpunk anime portrait of a young female cyborg in a neon night city, shown in an extreme close-up three-quarter view from the shoulders up, facing slightly right while looking directly at the viewer. She has long messy windblown {argument name="hair color" default="hot pink"} hair with many thin strands crossing her face, luminous magenta eyes, glossy lips, pale skin, and a calm intense expression. The left side of her face is exposed cybernetic machinery: polished dark chrome facial plates, thin black mechanical seams, small cables, and a glowing circular cybernetic lens around her left eye emitting bright pink light. Add a small kanji tattoo under the cybernetic eye reading {argument name="cheek tattoo" default="夢"}. She wears a high-collared black glossy leather cyberpunk jacket with metal ridges and a shoulder patch showing a pink demonic oni mask emblem. Her visible left ear has multiple piercings and dangling metal earrings, including one rectangular tag with vertical Japanese text reading {argument name="earring text" default="未来都市"}. The background is a blurred futuristic city at night with violet, blue, and magenta neon signs, including one vertical glowing sign that also reads {argument name="neon sign text" default="未来都市"}; use shallow depth of field and strong bokeh. Lighting should be dramatic and cinematic, with magenta rim light on the hair, cool blue shadows, glossy reflections on metal and leather, and a slightly gritty textured digital-paint finish. Composition should crop tightly at the top and sides, with the face dominating the frame, the cybernetic eye near the left-center focal point, and no extra characters or readable text beyond the specified tattoo, earring, and neon sign.

Create a dramatic monochrome manga/anime close-up illustration of {argument name="character name" default="Misaki, the Green Lady"}, a beautiful young woman with long straight {argument name="hair color" default="muted sage green"} hair and green eyes, wearing thin black oval glasses. She is shown from the chest up in a confident three-quarter pose, touching the bridge of her glasses with two fingers as if adjusting them, mouth slightly open mid-speech, expression composed and sharp. Her outfit is a refined white Victorian-inspired blouse with ruffled cuffs, ruffled front placket, small black buttons, a black ribbon bow at the collar, and a high neck detail; include exactly 6 visible outfit/accessory elements: glasses, black collar ribbon, ruffled cuffs, ruffled chest trim, black front buttons, and white blouse. Use a high-contrast black-and-white manga page style with selective green color only in the hair, eyes, and a green heart. Background: explosive radial speed lines, halftone textures, ink splatter, heavy black impact strokes, and manga sound-effect marks on the left and lower foreground. On the right side, place one large jagged speech bubble with vertical Japanese text reading {argument name="speech bubble text" default="真実は\nいつもひとつ\nですわ\n💚"}; keep the green heart at the bottom of the text. Composition should be dynamic and intense, with the character occupying the left and center, the speech bubble occupying the right third, crisp line art, screentone shading, glossy hair highlights, and no extra characters, no watermark, no borders beyond the manga-style effects.

Create a highly detailed vertical cinematic concept art scene set in a snowy cyberpunk Japanese mountain city at night. The viewpoint is from street level, looking up a wet stone stairway through a massive red wooden torii-style gate with a snow-covered roof beam and a central wooden sign reading {argument name="gate sign text" default="山形道"}. Two large glowing paper lanterns hang on the left and right sides of the gate, each also marked {argument name="lantern text" default="山形道"}, and exactly two smaller square lanterns hang beneath the gate beam. In the foreground, place exactly two young travelers seen mostly from behind, both wearing long weathered sci-fi kimono coats, layered armor, backpacks, belts, pouches, boots, and subtle cybernetic equipment; the left traveler is slightly taller with short dark hair and turns toward the right traveler, while the right traveler has a shorter bob haircut and looks back at the left traveler. Around them, include exactly four armed armored guards clearly visible near the gate area, plus one robed civilian with a straw hat at the far left and a small street vendor stall at the right. The city rises vertically in dense stacked wooden-and-metal buildings, pipes, catwalks, bridges, balconies, windows, neon signs, and hanging paper signs, blending Edo-period architecture with dystopian industrial science fiction. Use many warm amber window lights and lantern glows against cold blue-gray snow, mist, and towering dark structures. Snow falls through the air, rooftops and railings are dusted with snow, and the pavement is glossy with puddles reflecting orange lantern light. Mood: mysterious, immersive, lived-in, grand scale, quiet before an adventure. Style: ultra-detailed anime-realistic digital painting, intricate environmental design, cinematic lighting, atmospheric depth, sharp foreground detail, deep perspective, 9:16 vertical composition, no modern cars, no English signage, no watermark.

Create a dark cinematic horror close-up of {argument name="character name" default="an ominous priestess"} staring directly into the camera inside a dim cathedral or crypt. The subject is centered and symmetrical, an elderly pale woman with gaunt features, cracked dry skin, deep shadowed eye sockets, and glowing red eyes with small eerie reflections. She wears a tall black rectangular headdress rising out of frame and an off-white wrapped hood or veil around her head and shoulders, with heavy folds of fabric framing her face. Place three visible candle flames: one large blurred flame in the extreme foreground at the bottom center partially obscuring her chin, and two smaller blurred flames behind her on the left and right at eye height, creating a ritualistic triangular composition. Use {argument name="lighting style" default="very low-key warm candlelight"}, shallow depth of field, strong background blur, heavy film grain, muted brown-black shadows, and a tense occult atmosphere. Frame as a widescreen 16:9 movie still, face centered, intimate and unsettling, photorealistic, with anamorphic cinematic softness, no text, no watermark.

Create a polished anime-style portrait of a beautiful young woman with {argument name="hair color" default="warm chestnut brown"} hair styled in a high messy bun with loose wispy strands and soft straight bangs. She has large warm brown eyes, long lashes, delicate makeup, rosy cheeks, glossy lips, and a gentle confident smile while looking directly at the viewer. Show her from mid-thigh upward in a softly sensual but elegant pose, leaning slightly forward with her arms behind her back. She wears a fitted cream ribbed wrap crop top with short sleeves, a deep V neckline, scalloped trim, and a tied bow at the side, paired with a high-waisted dusty pink skirt. Add a fine gold necklace with a small pendant. The background is a minimal bright room with an off-white wall and a sheer curtain on the left, lit by warm natural window light that creates soft highlights on her face, hair, and clothing. Use highly detailed semi-realistic anime rendering, smooth skin, refined linework, glossy hair strands, soft shadows, pastel neutral colors, vertical 2:3 composition, shallow depth of field, clean elegant atmosphere, no text, no watermark.

Using the provided reference image as the character base, redraw it as a polished anime-style bust illustration while preserving the same character identity, head angle, gentle smile, blue hair, blue eyes, and animal ears. Clean up the rough sketch into refined thin line art, correct the facial proportions and symmetry, make the eyes more detailed and expressive, and render the coloring as soft watercolor-like cel shading with pale blush and airy highlights. Remove any stray marks or partial shapes from the original reference background and place the character on a plain clean white background. Keep the composition as a close-up head-and-neck portrait with no text, no accessories, and no extra background elements. Optional customization: character name {argument name="character name" default="blue-haired kemomimi girl"}, finish style {argument name="finish style" default="soft watercolor cel anime"}, background {argument name="background" default="plain white"}.

Goal: Create a flashy Japanese social-media sales banner advertising an AI appearance-consulting prompt package, designed like a high-converting X/Twitter promo thumbnail.

Canvas: Wide horizontal 16:9 banner, ultra-saturated, glossy, cinematic ad design. Background is an explosive gold-and-magenta money burst with flying yen coins, banknotes, spark trails, confetti, and intense lens flares.

Layout: Left third features a cute mascot character in a white hoodie and white pants, running or leaning forward with both hands extended toward the viewer. The character has a round white helmet-like head, black X-shaped eyes, and a big smiling black mouth, similar to a friendly faceless mascot. Above the mascot is one white speech bubble reading 「今日から」 and a large red exclamation mark pair nearby. Bottom-left has a black-and-gold laurel badge reading 「累計 5億」.

Center: The main focal point is huge 3D metallic-gold typography with dramatic bevels, shadows, and glowing highlights. At the top center, place a purple ribbon label reading 「1件あたり」. Below it, show the main price stack: 「¥10,000〜」 on the first line and 「¥20,000」 on the second line, extremely large and gold. Under the price, add a gold ribbon reading 「今日から売れる」.

Product elements: Include exactly 3 boxed product/package mockups around the center: 1) a dark red box near the left-center labeled 「提案資料 一式」 with a small document preview, 2) a smaller dark red box lower-left of center labeled 「プロンプト集」 with a chat bubble icon, and 3) a dark red box on the right labeled 「Before/After 画像」 showing two female profile photos side by side with small English labels “BEFORE” and “AFTER”.

Right side: Add a tilted red rubber-stamp graphic in the upper-right that says “SOLD OUT” in English, with a transparent white stamp background and distressed texture.

Bottom typography: Across the bottom, use enormous red-to-gold 3D Japanese lettering with white highlights and dark drop shadows reading 「1〜2万円／今日から売れる」. Just above that, on a white-and-gold banner, place purple text reading 「AI 外見コンサル完全プロンプト」.

Visual style: Maximum-impact Japanese internet marketing thumbnail, luxury gold, red, purple, and magenta palette, high contrast, exaggerated 3D text, glowing outlines, sparkle effects, dynamic motion, dense composition, celebratory “money explosion” atmosphere.

Text customization: Use the exact visible Japanese copy unless changed: {argument name="main price text" default="¥10,000〜 ¥20,000"}, {argument name="bottom headline" default="1〜2万円／今日から売れる"}, {argument name="product name" default="AI 外見コンサル完全プロンプト"}, {argument name="top ribbon text" default="1件あたり"}, {argument name="stamp text" default="SOLD OUT"}.

Constraints: Keep all text crisp and readable, preserve the exact count of 3 product boxes, 1 mascot, 1 speech bubble, 1 badge, and 1 sold-out stamp. Do not add extra logos, watermarks, or unrelated characters.

Create a vertical dark romantic gothic anime portrait of an androgynous young man with {argument name="hair color" default="silver-white"} long layered hair, soft bangs covering part of the face, and a loose low side ponytail tied with a black ribbon. He has pale skin, delicate features, downcast tired eyes with long eyelashes, and a melancholic, dreamy expression. Show him in a close waist-up composition, wearing a refined white Victorian blouse with billowy sleeves, buttoned cuffs, and a narrow black necktie. His slender gloved hands are raised to his mouth, gently holding and partly biting a cluster of deep {argument name="petal color" default="blood-red"} flower petals; include exactly three loose red petals falling or resting on his clothing. Place him behind a sheer black lace veil entering from the left side, with ornate floral lace texture partially obscuring his face and torso. The background is an antique oval mirror or carved frame with baroque ornamentation, surrounded by flowers: exactly three white orchids on the right side and multiple dark red peony-like blossoms behind him. Use a muted monochrome sepia-gray palette with selective deep red accents on the petals and flowers, soft diffused lighting, high-detail linework, elegant manga illustration style, delicate shading, moody gothic atmosphere, intricate fabric folds, ornate frame details, shallow depth, no text, no watermark.

Goal: Create a tall Korean infographic poster summarizing Anthropic’s lesson on building effective human-agent teams, page {argument name="page number" default="1/2"}, with a clean beige editorial tech style and numbered lesson sections.

Canvas: Vertical 2:3 poster, off-white warm background, thin rounded cards, orange numbered circles, black Korean typography, subtle gray dividers, small line-art illustrations, and Anthropic-like minimal branding. Use a dense but readable layout similar to a blog-summary infographic.

Header: Large bold title in Korean: “Anthropic의 교훈: 효과적인 Human-Agent Team 만들기”. Under it, small subtitle: “한 사람이 한 AI와 일하던 시대에서, 사람과 에이전트가 같은 워크스페이스에서 함께 일하는 팀의 시대로”. Metadata row: “원문 요약 | Claude Tag | 2026.06.24”. Top-right corner has a small rounded page badge “1/2” and an illustration of a hand placing blocks on three orange drawers; the drawers show simple icons for humans, a robot, and a target.

Layout: Use exactly 5 main numbered sections on this page, stacked in a clear grid. Section 1 and 2 sit side by side below the header. Section 3 is a wide band split between three participation cards and a mock chat screenshot. Section 4 is a full-width lesson card. Section 5 is another full-width lesson card. Bottom has a navigation strip, source strip, and footer branding.

Section 1: Number badge “1” and title “무엇이 달라졌나?”. Show a comparison diagram with exactly 2 panels: left panel labeled “싱글플레이어”, showing one human icon ↔ one AI robot icon with caption “1명 인간 ↔ 1개 AI”; right panel labeled “멀티플레이어”, showing multiple human silhouettes + multiple robot icons + a target icon with caption “여러 사람 + 여러 에이전트 + 공동 목표”. Add a bottom callout: “이제 AI 협업은 개인 작업이 아니라 팀 작업에 가깝다”.

Section 2: Number badge “2” and title “멀티플레이어 에이전트란?”. Include exactly 3 definition rows with icons: robot icon row “여러 인간과 동시에 일하는 AI 에이전트”; brain icon row “자체 기억(memory), 스킬(skills), 자격증명(credentials)을 보유”; Slack logo-inspired colored icon row “Slack 같은 협업 공간에 상주하며 팀 목표를 지원”. Bottom example pill: “예: 팀 채널에서 데이터 분석을 함께 진행”.

Section 3: Number badge “3” and title “팀에 생산적으로 참여하려면”. Left side contains exactly 3 rounded cards: card 1 with head and sticky-note icon, title “지속적 메모리”, description “목표를 기억하고 실행을 조정”; card 2 with shield and key icon, title “인간과 분리된 자격증명”, description “안전한 권한과 예측 가능한 가드레일”; card 3 with database and magnifying-glass icon, title “넓고 지속적인 정보 접근”, description “조직이 원하는 방식을 이해하고 행동”. Below those cards add a star callout: “기술만으로는 부족하다 → 팀의 운영 방식과 공통 규범도 필요”. Right side shows a Slack-like chat screenshot titled “# data-analysis” with a purple sidebar and exactly 5 visible chat messages from these senders: “민지 (분석 리드)”, “Claude (Data Analyst)”, “호준 (PM)”, “Claude (Data Analyst)”, and a small final status/message area. Include tiny avatars and timestamps, but keep text mostly legible and compact.

Section 4: Number badge “4” and title “Lesson 1. 공개적으로 일하고, 에이전트에 넓은 맥락을 제공하라”. Left illustration: exactly 3 people sitting with laptops around a dotted shared workspace diagram containing document, chat, and code icons. Middle bullet area explains that agents should work in visible spaces like Slack, Jira, documents, and repositories; private DMs can hide context; and agents need access to recorded conversations and artifacts. Add two side cards: card titled “왜 중요한가?” with orange star icon and bullet points about memory not being automatic, missing context lowering output quality, and shared context improving recommendations; card titled “Anthropic 식 운영 방식” with gear icon and bullets about shared channels, openly sharing work products, moving useful DMs into shared space, and documenting easy-to-read summaries. Bottom lock callout: “민감한 대화는 DM, Claude.ai, Claude Code와처럼 별도 공간에서 다룰 수 있다”.

Section 5: Number badge “5” and title “Lesson 2. 모든 인간과 에이전트에 역할과 도구를 명확히 부여하라”. Include a short Korean bullet list about assigning clear roles, tools, credentials, and expectations. Below left, create exactly 2 small tables: “Human (사람)” table with 4 rows of human teammates and roles; “Agents (에이전트)” table with 4 rows: “Claude (Data Analyst)”, “Claude (Design Critic)”, “Claude (Research Synthesizer)”, “Claude (QA Tester)”. On the right, a green bordered checklist card titled “잘 작동하는 팀의 모습” with exactly 5 checklist items: “합의된 Task Set”, “사람과 에이전트가 같은 스레드에서 함께 작업”, “각 역할의 범위와 책임을 문서화”, “개인은 AI가 나타나는 대신, 팀 전체가 같은 맥락과 같은 숫자를 공유”, and one concise final item about shared tools and permissions.

Bottom navigation: Add an orange arrow icon and text: “다음 페이지: Lesson 3 복잡성 설정 · Lesson 4 신뢰 구축 · 체크리스트 · 결론”, with a small “1/2” badge on the right. Source strip contains two icon-labeled items: “출처 URL: https://claude.com/blog/building-effective-human-agent-teams” and “원문 제목: Building effective human-agent teams”. Bottom-right large wordmark “ANTHROPIC”. Bottom-left copyright “© 2026 Gonnecter”. Add a tiny colorful abstract logo at the bottom-right corner.

Visual style: Modern Korean business infographic, crisp vector line art, warm cream background, orange accents, muted purple for chat mockup, green for success checklist, precise alignment, high information density, readable Korean text, no photorealism, no shadows beyond very subtle card depth.

Constraints: Keep exactly 5 numbered main sections, exactly 2 comparison panels, exactly 3 definition rows, exactly 3 participation cards, exactly 5 chat messages, exactly 2 role tables, and exactly 5 checklist items. Do not add extra pages, extra lesson numbers, watermarks, or unrelated icons.

Create a realistic vertical concert photo of a cheerful young Japanese idol singer named {argument name="character name" default="a smiling idol singer"} performing on a small live-house stage. She is shown from mid-thigh upward in a dynamic close perspective, leaning toward the camera, smiling with her mouth open while singing, holding exactly 1 black handheld microphone in her right hand and reaching her left arm toward the viewer with a playful pointing gesture. She wears a glossy white Chinese-inspired idol dress, a fitted sleeveless qipao-style mini dress with pale floral brocade texture, gold piping, black frog closures, black trim, and a front slit; use {argument name="outfit color" default="ivory white with gold and black accents"}. Her hair is {argument name="hair color" default="dark brown"}, styled with straight bangs and exactly 2 small messy side buns, decorated with exactly 2 white floral hair ornaments and trailing white ribbons. The stage background is dim and industrial, with dark ceiling beams, concrete walls, shallow depth of field, and exactly 2 bright circular stage lights glowing behind her, creating warm rim light and bokeh. Use a high-quality DSLR concert photography look, natural skin tones, sharp focus on her face and microphone, slight motion energy in the extended hand, 3:4 vertical composition, intimate idol performance atmosphere, no text, no watermark.

Create a delicate vertical anime-style illustration of {argument name="character name" default="Celestin"}, an ethereal angelic young man with very pale skin and soft, messy {argument name="hair color" default="snow-white"} hair falling over his eyes, shown in close-up three-quarter profile as he gently kisses the back of someone’s hand. His expression is calm, reverent, and slightly melancholic, with long eyelashes and downcast eyes. He wears layered white ceremonial clothing with translucent fabric, pale blue accents, ornate silver jewelry, a dark high collar with metallic trim, dangling earrings, and blue crystal-feather ornaments near one ear. Large soft white wings and drifting feathers surround him, filling the background with a dreamy heavenly atmosphere. Use a luminous watercolor and pencil-sketch aesthetic, very soft pastel whites, icy blues, pearly grays, high-key lighting, airy brush textures, sparkling snow-like particles, and a romantic shoujo manga composition. The foreground hand should be gracefully held in both of his hands as he kisses it, with elegant slender fingers. Add handwritten glowing white text near the lower right reading {argument name="thank you message" default="Thank you for 400♡"}, plus a small cute doodle of a winged chibi face or angel mascot beside it. Canvas should be vertical 9:16, intimate cropped portrait, no hard outlines, no modern objects, no watermark.

Create a vertical storybook-style watercolor illustration of a soft plush doll named {argument name="character name" default="Mimi"} sitting on an antique wooden writing desk in a cozy old library. The doll is centered and shown full body, with a large round chibi head, tiny seated body, velvety fabric texture, pale cream plush skin, short bobbed {argument name="hair color" default="pastel pink"} hair with blunt bangs, 2 large blue embroidered-looking eyes, a small neutral mouth, faint blush, and 2 cat ears with pink inner fabric. Dress the doll in a black-and-white maid outfit: white ruffled headband, black long-sleeve dress, white ruffled apron, cream frill trim at cuffs and hem, black plush shoes, and a purple ribbon bow at the collar. The scene should feel warm, nostalgic, and tactile, emphasizing fuzzy fabric and hand-sewn plush seams. The background is a richly detailed antique study: a glowing sunlit window on the left, dark wooden bookshelves filled with old books on the right, a framed vintage painting above the doll, and stacks of aged books on the desk. Include exactly 5 prominent foreground desk items: 1 handwritten parchment sheet, 1 black ink bottle, 1 white feather quill, 1 lit candle in a brass holder, and 1 small ornate brass object. Use golden afternoon light streaming from the left window, soft shadows, sepia browns, cream highlights, and delicate watercolor pencil linework. Composition should be vertical 9:16, intimate close-up, with the plush doll occupying the middle of the frame and the desk objects framing the lower foreground. Add a subtle illustrated texture like handmade paper, high detail, cozy fantasy atmosphere, no modern objects, no visible text, no watermark.

Using REFERENCE_0 as the creature design base, turn the rough sketch into a polished collectible toy/garage-kit render. Preserve the monster’s core silhouette and features from the sketch, but convert it into a glossy translucent green vinyl figure with soft sculpted details, rounded toy-like proportions, pearly off-white claws/teeth/belly highlights, and black-and-white glossy eyes.

Create a clean product turnaround sheet with exactly 3 vertical panels separated by white gutters: 1) front view, 2) left side profile, 3) right side / rear three-quarter profile. Keep the figure standing upright in the same general pose, with the horns, many eyes, small wings, multiple arms, segmented belly, chunky legs, tail, claws, bumps, scales, and kaiju-toy texture translated into a finished 3D sculpt.

Style: {argument name="toy material" default="translucent emerald green glossy vinyl"}, studio product photography, soft even lighting, subtle shadows, neutral light gray background, high detail, playful Japanese sofubi monster aesthetic. Do not include text, logos, hands, tools, or sketch lines. The final image should look like rendered photos of a physical prototype ready for 3D modeling or sculpting reference.

Create a square full-bleed digital illustration on a plain solid white background showing an anime twin transformation layout. The main subject is a cute minimalist chibi anime girl occupying the center and right side of the canvas from head to upper torso, with an oversized head, tiny shoulders, very large glossy dark brown eyes, soft peach skin, rosy blushing cheeks, a small nose highlight, and a tiny slightly parted neutral mouth. She has {argument name="hair color" default="dark brown almost black"} messy short wavy bob hair with loose flyaway strands, parted near the center, with one thick loose side braid falling over her left shoulder. Add exactly 1 small pink five-petal flower clip attached near the lower part of the braid. She wears {argument name="top color" default="black"} minimalist spaghetti-strap camisole top with exactly 2 thin shoulder straps and a shallow V neckline. The expression should feel shy, calm, and doll-like, with delicate clean line art, soft watercolor-like shading, subtle blush gradients, and polished anime rendering. In the upper-left corner, include exactly 1 small rounded-rectangle inset photo frame with a thin dark border, showing a realistic portrait of the same young woman as the source look: short tousled dark wavy hair, natural makeup, black spaghetti-strap top, seated in a car interior, facing the camera with a neutral expression. The inset should be about one quarter of the canvas width, placed with a small margin from the top and left edges. Keep the background completely white, with no text, no watermark, no extra accessories, and no additional characters. Use a clean high-resolution social-media avatar style that makes the realistic inset and the larger anime version visually match in hairstyle, outfit, and mood.