GPT Image 2 prompts

Goal: Create a dark tactical firearm character-sheet style technical poster for the {argument name="rifle model" default="FN SCAR-L"}, presented as a clean military product infographic.

Canvas: Vertical 2:3 poster on a matte charcoal-black textured background, with worn tan stencil typography and thin muted-gray grid lines. Use realistic studio-rendered firearm illustrations in flat side profile, colored coyote tan and black, with subtle scratches, edge wear, and soft shadows.

Header: At the top left, place a huge distressed stencil title reading “FN SCAR-L”. Beneath it, add three smaller lines: “[SPECIAL OPERATIONS FORCES COMBAT ASSAULT RIFLE – LIGHT]”, “{argument name="caliber line" default="5.56x45mm NATO / 7.62x51mm NATO"}”, and “{argument name="tagline" default="THE MODULAR BATTLE RIFLE FOR THE MODERN WARRIOR"}”. At the top right, include a circular FN monogram-style logo and the text “FN HERSTAL”.

Main visual layout: Show exactly 4 firearm views/assemblies. 1) A large full-length left-facing side profile rifle spanning the upper half, dominant and highly detailed. 2) A smaller full rifle side profile beneath it on the right, used for numbered feature callouts. 3) A partially disassembled upper receiver and barrel assembly in the lower middle, with barrel components separated horizontally. 4) A lower receiver/stock assembly below it, with magazine removed and a small loose pin or screw nearby. Add one cropped rifle detail close-up at the bottom right inside the caliber conversion box.

Feature callouts: Include exactly 6 numbered callouts with small square labels and fine leader lines pointing to rifle parts. The numbered list at left must read: 1 “MIL-STD-1913 TOP RAIL” with subtext “For optics and accessories”; 2 “SIDE FOLDING STOCK” with subtext “Adjustable length of pull”; 3 “AMBIDEXTROUS CONTROLS” with subtext “Selector, magazine release, and bolt catch”; 4 “FREE-FLOATING HANDGUARD” with subtext “With accessory rails at 3, 6, and 9 o’clock”; 5 “QUICK-CHANGE BARREL” with subtext “Caliber conversion in minutes”; 6 “SHORT-STROKE GAS PISTON” with subtext “Reliable in adverse conditions”.

Specification table: Place a boxed table in the lower left titled “SPECIFICATIONS [5.56x45mm]”. Include exactly 10 rows: Caliber: 5.56x45mm NATO; Operation: Gas-operated, short-stroke piston; Magazine capacity: 30 rounds (STANAG); Overall length: 35.25 in (895 mm); Barrel length: 16.25 in (413 mm); Weight (unloaded): 7.50 lb (3.40 kg); Rate of fire: 625–650 rounds/min; Muzzle velocity: ~2,750 ft/s (838 m/s); Effective range: 500 m; Sights: Iron sights, foldable (front & rear), optic ready (MIL-STD-1913 rail).

Bottom information panels: Add two boxed text panels along the bottom. Left panel: a short paragraph explaining that the FN SCAR-L is a modular assault rifle designed for U.S. Special Operations Command and chambered in 5.56x45mm NATO, emphasizing reliability, accuracy, adaptability, and lightweight battle-proven use. Right panel title: “CALIBER CONVERSION”, followed by text explaining quick conversion using FN’s quick-change barrel system and exactly 2 bullet calibers: “5.56x45mm NATO (SCAR-L)” and “7.62x51mm NATO (SCAR-H)”.

Visual constraints: Keep all text in English, crisp and legible; use a professional weapon manual / defense catalog aesthetic; avoid people, hands, firing effects, flags, or clutter; maintain precise orthographic side views and a subdued black, tan, and gray color palette.

Goal: Create a realistic military technical character-sheet style infographic for the {argument name="rifle model" default="M2010 ESR"}, showing a tan enhanced sniper rifle with labeled components, specifications, measurement guides, and detail close-ups.

Canvas: Vertical 2:3 poster on a dark charcoal textured background, high-resolution, realistic 3D product-render look, muted tactical colors, tan and black rifle parts, beige military stencil typography.

Layout: At the top left, place a large title reading “{argument name="headline text" default="M2010 ESR"}” with the subtitle “ENHANCED SNIPER RIFLE” underneath. Below it add a short description: “The M2010 Enhanced Sniper Rifle (ESR) is a 7.62×51mm NATO precision bolt-action rifle adopted by the U.S. Army in 2010 as the upgrade to the M24 Sniper Weapon System.” The upper half is dominated by a large three-quarter view of the rifle angled from lower left to upper right, with scope, bipod, long barrel, muzzle brake, adjustable stock, pistol grip, magazine, and free-float handguard visible.

Callouts and text: Include exactly 3 component callouts around the large rifle: “MIL-STD-1913 Accessory Rail,” “5 or 10 Round Detachable Magazine,” and “Adjustable Stock (Length of Pull & Cheek Weld).” Add a “FEATURES” block with exactly 6 bullet points: “Accuracy and effective range improved over M24,” “Modular design for easier maintenance and upgrades,” “Adjustable stock for length of pull and cheek weld,” “Match-grade barrel for enhanced precision,” “Free-float handguard with accessory rail,” and “Compatible with sound suppressors.”

Specifications panel: On the right side of the upper rifle, include a boxed table titled “SPECIFICATIONS” with exactly 9 rows: Caliber: 7.62×51mm NATO; Action: Bolt-action; Overall Length: 48.5 in (1,230 mm); Barrel Length: 24 in (610 mm); Weight: 14.8 lb (6.7 kg) w/ scope & bipod; Magazine Capacity: 5 or 10 rounds; Effective Range: 800–1,000 m; Maximum Range: 1,500 m+; Sights: Leupold Mark 4 6.5–20×50mm.

Middle diagram: Add a second full side-profile view of the same rifle across the center, perfectly horizontal, smaller than the top render. Add measurement lines showing “48.5 in (1,230 mm)” above the rifle and “8.6 in (218 mm)” vertically near the stock.

Detail section: At the bottom, create a grid titled “DETAIL VIEWS” with exactly 6 close-up panels in two rows of three. The panels are: 1) “Muzzle Brake” with subtext “Reduces recoil and muzzle rise”; 2) “Free-Float Handguard” with subtext “Improves accuracy by minimizing barrel contact”; 3) “Bolt Action” with subtext “Smooth, reliable, and proven design”; 4) “Adjustable Stock” with subtext “Customizable for shooter fit and comfort”; 5) “Detachable Magazine” with subtext “5 or 10 round options for operational flexibility”; 6) “Leupold Mark 4 Scope” with subtext “6.5–20× magnification for long-range precision.”

Visual style: Photorealistic tactical catalog rendering, subtle rim lighting, crisp labels and thin pointer lines, slight vignette, professional defense-industry spec sheet aesthetic. Keep all text legible, use only the listed text, and do not add people, logos, flags, watermarks, or extra weapons.

Goal: Create a nostalgic instant-film / Cheki-style anime illustration of {argument name="character name" default="a working girl"} looking dejected at her desk after receiving criticism on her AutoCAD work, with the concept of a delicate outfit worn by someone sturdily living everyday working life.

Canvas: Square 1:1 image, designed as an aged Polaroid print with a thick cream paper border, rounded corners, dust, scratches, stains, vignette darkening, and a sepia faded-photo treatment.

Main photo area: Show exactly 1 young anime-style woman in side profile, seated at a desk in a dim office or studio. Her face is mostly hidden by a plain square censor block in muted beige-brown, centered over her face. She has long wavy {argument name="hair color" default="blonde"} hair in a high ponytail with loose curls, braided details, small floral hair ornaments, dangling earrings, and a very large bow headband. Her outfit is a frilly lavender-blue and pink lace blouse or dress with ribbons, puffed sleeves, and ornate trim, intentionally cute and refined despite the work setting. Her posture should feel slumped, disappointed, and quietly hurt.

Work environment: Include exactly 1 computer monitor on the right, angled toward her, showing a pale CAD drafting interface with architectural line drawings, toolbars, and technical rectangles. Include exactly 1 warm desk lamp in the upper right casting a soft cone of light. Background should be softly blurred with window-like shapes and warm dusty atmosphere.

Border design: The Polaroid border is hand-decorated with exactly 24 doodles in colored pen: 8 hearts, 4 stars, 3 arrows, 3 dotted swirl trails, 3 sparkle clusters, and 3 small circles. Use pink, purple, blue, and mustard-yellow ink. Make the doodles casual, uneven, and handwritten, as if drawn on the photo after printing.

Text content: On the lower white border, add handwritten Japanese text in dark blue ink with a slightly messy personal diary style. Use exactly 3 text elements: a two-line main caption reading {argument name="main handwritten caption" default="AutoCAD が出来たら\n何でも出来る…そう思っていた"}, a smaller third line reading {argument name="sub caption" default="時期もありました"}, and a date at the bottom right reading {argument name="date" default="2026年5月19日"}. Keep the writing legible but natural, like marker or ballpoint pen on old paper.

Visual style: Highly detailed soft anime illustration, delicate linework, warm sepia lighting, muted pastel palette, grainy analog film texture, slight overexposure around the lamp, emotional slice-of-life mood, cute but melancholic.

Constraints: Do not add extra characters, extra monitors, logos, watermarks, or modern UI overlays. Keep the square censor block flat and opaque. Preserve the handmade instant-photo frame and aged paper texture.

Goal: Create an elegant widescreen PPT courseware cover slide about {argument name="topic" default="the past and present of blueberries"}, blending Chinese editorial typography with soft watercolor botanical illustration.

Canvas: 16:9 landscape presentation slide, warm off-white handmade paper texture background with subtle fibers, faint stains, and minimal ink speckles.

Layout: Left third is dominated by a large Chinese title in dark blue-gray: {argument name="Chinese title" default="蓝莓的\n前世今生"}. Place the title in two stacked lines with generous spacing, using a refined Song/Ming-style serif font. Above it, small uppercase text reads “PPT COURSEWARE” preceded by a tiny gold circular icon. Under the title, add a short thin gold line and the English subtitle in small serif capitals: {argument name="English subtitle" default="THE PAST AND PRESENT OF BLUEBERRIES"}. Bottom left shows the year {argument name="year" default="2026"} in large serif numerals, with a thin horizontal line beneath, then five small Chinese category labels separated by vertical dividers: “起源 | 演化 | 传播 | 栽培 | 科学 | 生活”.

Main imagery: Use exactly 6 visible blueberry-related watercolor elements across the slide: 1 large soft blue blueberry bloom near the upper center with a dark star-shaped calyx, 1 translucent blue leaf angled diagonally below-left of it, 1 small blueberry bloom near lower left labeled “origin”, 1 clustered pair of blurry blueberries near lower middle-right with one visible calyx, 1 circular blueberry cross-section on the right labeled “cut” with pale pulp and small seed dots, and 1 small fuzzy blueberry wash near the lower right. Keep all fruit elements desaturated indigo, smoky blue, and gray-blue, with feathered watercolor edges.

Technical annotation details: Add fine gold and blue-gray diagram marks: tiny crosses, dotted guide lines, small circles, target rings, and thin leader lines. Include exactly 3 small numeric labels: “01” near the upper-center blueberry, “02” near the middle-right cluster, and “01” as a page number at the bottom right. Add the labels “origin” and “cut” in very small serif text.

Right margin: Add a narrow vertical text column on the far right with small Chinese copy: {argument name="vertical quote" default="一颗小小的蓝色果实，如何跨越时间与空间，与人类的文明相遇、相知、相伴。"}. Beside or below it, add very small vertical English text in uppercase, decorative and partially editorial, about blueberries traversing time and space.

Visual style: Minimal premium academic presentation design, Chinese new-literati aesthetic, botanical science notes, high negative space, soft ink-wash watercolor, muted navy-blue and gray palette with restrained antique-gold accents, crisp typography but organic painted fruit. No people, no photorealistic fruit, no bright colors, no watermark.

Goal: Create an elegant Tang-dynasty inspired courseware/PPT cover slide about {argument name="main title" default="杨贵妃 与 荔枝"}, with a poetic Chinese historical aesthetic centered on lychees and a veiled court beauty.

Canvas: 16:9 horizontal presentation cover, warm ivory handmade-rice-paper background with subtle fibers, stains, faint scratches, and ink-wash aging. Overall palette is muted cream, dusty rose, pale peach, faded red, olive gray, and ink gray.

Layout: Place the main title large on the left-center in refined Chinese calligraphy-style serif lettering. Put the small subtitle directly underneath with a decorative gold flourish at the far left. Keep generous negative space on the left and lower middle. On the right half, show a soft translucent back-view silhouette of a Tang-dynasty noblewoman with bare shoulders, high rounded hair buns, ornate gold dangling hair jewelry, and large pink-red peony flowers in her hair; her face area is intentionally covered by a plain pale beige rectangular privacy block. Add floating lychees around her and foreground depth blur.

Text content: Include exactly these visible text areas: 1) top-left small cloud-shaped seal containing “盛唐 / 风物”; 2) main title “杨贵妃 与 荔枝”; 3) subtitle “一颗果子的盛唐记忆”; 4) lower-left English stack “Tang Dynasty / Cultural Story / Courseware”; 5) lower-left Chinese label “PPT · 课件”; 6) right-side vertical poem text “长安回望绣堆，山顶千门次第开。——杜牧《过华清宫》”; 7) small tag near the peeled lychee reading “贵妃鲜” with a red seal mark; 8) bottom-right page number “01” with a small gold ornamental underline.

Subject details: Include exactly 4 visible lychee elements: one small blurred whole lychee floating near the upper center, one sharp whole lychee on the right side, one large peeled lychee in the lower-right center with translucent white flesh exposed and red bumpy skin opened around it, and one large heavily blurred partial lychee in the lower foreground. Add a few pale green lychee leaves connected to the peeled fruit. Make the lychees painterly but realistic, with bumpy red shells and soft natural highlights.

Visual style: Minimal luxury Chinese editorial design, museum-courseware cover, soft ink wash, translucent layering, shallow depth of field, delicate gold accents, understated red stamp marks, balanced asymmetrical composition, no modern UI elements.

Constraints: Keep the page clean and airy, maintain the exact count of 4 lychee elements, keep all typography readable, use no extra text beyond the specified labels, no watermark, no photorealistic harsh lighting.

Goal: Create a high-fashion menswear concept board inspired by {argument name="inspiration source" default="deep sea jellyfish"}, combining a central couture runway look with fashion design research panels, technical tailoring sketches, material tests, color studies, and handwritten atelier annotations.

Canvas: Portrait-oriented editorial fashion board on warm off-white archival paper, approximately 4:5 ratio, with faint grid marks, thin drafting lines, crop marks, pin-registration marks, subtle paper grain, and a luxury atelier mood. Use black, deep navy, icy cyan, translucent blue, silver, and white.

Main subject: Place one full-body male model centered and slightly forward, wearing {argument name="collection look name" default="Look 07 - Abyssalis"}. The face should be softly blurred or obscured, while the outfit remains sharp. The garment is futuristic couture menswear: a glossy black structured long coat or suit base with a high stand collar, layered liquid-black tailoring, wide sculptural shoulders, transparent jellyfish-like cape panels, flowing organza veils, fine dangling filament strands, reflective wet-look surfaces, and scattered bioluminescent cyan-blue bead lights. The silhouette should feel like a deep-sea creature translated into formal tailoring, with long translucent tentacle ribbons trailing from shoulders, sleeves, and hem to the floor. Add black boots visible beneath the layered coat.

Left research column: Include exactly 8 stacked research sections with small labels and visual materials:
1. Header reading “INSPIRATION : DEEP SEA JELLYFISH” with subheading “BIOLUMINESCENT STRUCTURE.”
2. “ANATOMI STUDY” section with exactly 2 jellyfish anatomy sketches, one blue-toned and one monochrome line sketch, plus small handwritten notes about a transparent bell, gelatinous structure, and radial symmetry.
3. “TENTACLE DYNAMICS” section with exactly 1 side-view jellyfish drawing showing long streaming tentacles, annotated with fluid motion, graceful undulation, and movement.
4. “BIOLUMINESCENCE ANALYSIS” section with exactly 2 glowing jellyfish reference images and exactly 5 vertical color bars beside them, from black and deep navy through teal to pale icy blue.
5. “MATERIAL EXPERIMENTS” section with exactly 3 square translucent material samples, annotated as fine membrane, iridescent organza, sheer tulle, liquid satin, and fiber optic thread.
6. “COLOR PALETTE” section with exactly 7 rectangular swatches: black, midnight navy, dark blue, teal, cyan, pale blue, and nearly white translucent blue, annotated as deep sea, ice blue, bioluminescent, and pearl white.
7. “FABRIC SWATCHES & TESTS” section with exactly 5 overlapping physical swatches in black vinyl/satin and translucent blue organza, some with glowing blue thread or bead-like highlights.
8. “LIGHT REACTION TEST” section with exactly 1 small dark photo panel showing bright blue fiber-optic streaks, plus a note about reaction, bioluminescent thread, subtle glow, and in darkness. At the bottom include atelier notes reading “ATELIER NOTES,” “PROJECT : ABYSSALIS,” “LOOK : 07,” “SEASON : SS27,” and “DEVELOPMENT STAGE : 4,” followed by a designer signature line.

Right technical column: Include exactly 6 design-development sections:
1. Top heading “TECHNICAL TAILORING SKETCH” with subheading “LOOK 07 - ABYSSALIS,” showing exactly 2 tall technical fashion sketches, front and back views, of the same long translucent layered coat.
2. A bullet-note block with exactly 5 bullets: bioluminescent tailoring; transparent marine layering; liquid silhouette engineering; floating organza extension; deep-sea inspired main couture; fluid structural movement.
3. A construction-detail area with exactly 4 smaller line drawings: structured inner coat with enlarged organic hood; transparent layered cape with floating silhouette; tentacle extensions as sheer organza; flowing wide-leg pants with liquid cut lines.
4. “EMBELLISHMENT MAP” section with exactly 1 front torso sketch covered in blue dot placements and exactly 3 legend dots labeled high density, medium density, and low density; add a note for fiber optic embroidery placement.
5. “DETAIL & TEXTURE CLOSE-UP” section with exactly 3 rectangular close-up images showing black glossy fabric with blue glowing bead lines, transparent crinkled organza, and watery translucent blue membrane texture.
6. Sparse handwritten annotations around the sketches, using elegant thin cursive, while keeping the text secondary and not overcrowded.

Visual style: Photorealistic central model blended with delicate pencil fashion sketches, scientific specimen diagrams, luxury couture board presentation, translucent layering, iridescent highlights, cyan bioluminescent specks, black liquid satin, and fine technical drafting aesthetics. The board should feel like an elite fashion house concept sheet for futuristic couture menswear, directly inspired by deep-sea jellyfish anatomy.

Text constraints: Use the visible English headings and atelier labels as described. Keep typography small, refined, and editorial; mix uppercase serif labels with handwritten cursive notes. Do not add logos, watermarks, extra models, or unrelated sea creatures. The final image should read as one complete fashion concept board for {argument name="garment theme" default="bioluminescent jellyfish couture menswear"}.

Goal: Create a highly detailed vertical museum-style infographic poster about {argument name="monument name" default="自由女神像"}, combining Chinese primary typography with small English subtitles, as if from an architectural history atlas.

Canvas: Tall portrait poster, approximately 2:3 aspect ratio, warm ivory paper background with faint grid lines, thin technical drafting rules, subtle aging texture, and precise editorial margins. Use a refined beige, charcoal, muted bronze, oxidized copper green, and faded gray palette.

Main title area: At the top center place the large black Chinese headline “自由女神像”, with the English title “STATUE OF LIBERTY” beneath it in spaced serif capitals. Add a small bilingual motto line below: “自由照亮世界，人类追求自由与希望的永恒象征” and “FREEDOM ILLUMINATES THE WORLD, AN ETERNAL SYMBOL OF HUMAN PURSUIT OF LIBERTY AND HOPE.”

Central subject: Dominating the middle, show a full-height Statue of Liberty technical cutaway. The left half should be the exterior oxidized copper statue with flowing robes, raised torch, crown spikes, and tablet; the right half should be sliced open to reveal dark iron lattice scaffolding, internal stairs, cross-bracing, and pedestal structure. Render it as realistic engraved architectural illustration with soft shadows and precise linework.

Left information column: Add a narrow sidebar titled “基本信息 / BASIC INFORMATION” with exactly 8 icon-and-text entries: official name, location, dedication date, designer, sculptor, total height, weight, and symbolism. Below it add “材料与工艺 / MATERIALS & CRAFTSMANSHIP” with exactly 4 stacked material swatches: copper sheets, wrought iron frame, granite base, and concrete foundation. Add a small “关键数据 / KEY DATA” block with exactly 6 compact numeric rows.

Left of central statue: Add “结构解析 / STRUCTURAL ANALYSIS” with exactly 4 circular magnified callout diagrams connected to the statue: torch structure, head structure, internal skeleton, and copper fixing method. Each circle contains a cropped technical detail, short Chinese label, English subtitle, and two short annotation lines.

Right analytical panels: Add “尺寸图与比例 / DIMENSIONS & PROPORTIONS” with exactly 5 diagrams: full statue height comparison, statue-only height, pedestal height, top-view star-shaped base diagram, and base-width measurement. Add “结构系统分解 / STRUCTURAL SYSTEM BREAKDOWN” with exactly 5 stacked exploded mini illustrations: copper exterior, iron framework, internal spiral/stair system, granite pedestal, and concrete foundation.

Far-right sidebar: Add “环境与耐久性 / ENVIRONMENT & DURABILITY” with exactly 4 icon entries: wind resistance, structural design, temperature range, and lightning safety. Add “方向与视野 / ORIENTATION & VIEWSHED” with a compass rose and a pale New York Harbor map, including radial sight lines. Add “全球影响力 / GLOBAL IMPACT” with exactly 4 icon entries: symbol of immigration, annual visitors, global freedom icon, and cultural influence.

Bottom middle band: Add “历史时间线 / HISTORY TIMELINE” on the lower left with exactly 8 dated milestones from concept, design, construction, shipment, assembly, dedication, restoration, and modern preservation. Across the lower center add “建造过程与运输安装 / CONSTRUCTION & ASSEMBLY PROCESS” with exactly 7 numbered sepia photo cards labeled 01 to 07: selected fabrication, disassembly and crating, maritime transportation, pedestal construction, statue assembly, final completion, and public opening. Each card has a small archival image, number, Chinese label, and English subtitle.

Bottom scenic footer: The entire bottom quarter should be a cinematic panoramic scene of the Statue of Liberty on Liberty Island at golden-hour sunset, with New York City skyline in the background, calm harbor water, ferries, warm clouds, and atmospheric haze. Overlay white text on the left titled “自由的灯塔 / THE BEACON OF FREEDOM” followed by a short paragraph about welcoming people from across the world. On the right overlay “设计理念 / DESIGN PHILOSOPHY” with exactly 4 icon entries: Neoclassical style, Inspired by Roman Libertas, Symbol of Enlightenment, and A Gift Beyond Borders. Add a centered small bilingual dedication line near the bottom: “法国人民赠予美国人民 / A GIFT FROM THE PEOPLE OF FRANCE TO THE PEOPLE OF THE UNITED STATES.”

Visual style: Premium educational infographic, dense but orderly, realistic architectural engraving mixed with archival photography, thin serif typography, subtle Chinese editorial layout, crisp icons, accurate proportions, clean grid alignment, and no modern UI elements. Ensure all labels are legible and the poster feels like a professional historical architecture reference sheet.

Goal: Create a black-and-white manga page about two schoolgirls arriving at a school gate under cherry blossoms, with one character partly highlighted in pale cyan and the other character’s face obscured by gray blur blocks.

Canvas: Vertical manga page, approximately 2:3 aspect ratio, high-resolution monochrome screentone style with thick black panel borders, clean line art, speed lines, petals, and dramatic perspective. Use grayscale shading with only one accent color: faint cyan glow on the glasses-haired girl.

Layout: Use exactly 5 visible manga panels. Panel 1 is a large top-left scene showing a twin-tailed schoolgirl in a sailor uniform standing on a cherry blossom path with the school gate far behind her; her face is covered by a soft-edged gray rectangular blur. Panel 2 is a narrow inset at the top right showing a cropped long-haired girl in sailor uniform holding a school bag in front of her body. Panel 3 is another top-right inset beside it showing the same long-haired girl wearing a beret and glowing cyan glasses, touching the bridge of her glasses with a confident gesture. Panel 4 is a wide middle panel showing the twin-tailed girl close up, raising one hand in greeting; her face is again covered by a larger gray rectangular blur. Panel 5 is a large bottom panel showing a dramatic tilted view of an ornate school gate opening in bright light, with the long-haired girl running or turning toward the viewer and the cyan-glasses girl standing farther back on the right.

Characters: The twin-tailed girl has messy medium-length twin ponytails tied with ribbons, a sailor school uniform with bow, pleated skirt, and shy energetic body language. The long-haired girl has very long flowing hair, sailor uniform, pleated skirt, school bag in one inset, and a composed expression in the bottom panel. The second version of the long-haired girl wears a round beret, glasses glowing pale cyan, sailor uniform, dark skirt, knee socks, and has a mysterious tech-like aura.

Text content: Include exactly 3 Japanese speech/sound text elements: a speech bubble in panel 1 reading {argument name="first speech bubble" default="あ見えたー校門"}, a speech bubble in panel 4 reading {argument name="second speech bubble" default="沙羅さーん、葵ちゃん！"}, and large stylized gate-opening sound effects in panel 5 reading {argument name="sound effect text" default="ザァァァ"}. Add small handwritten manga sound effects near the girls if needed, but keep them minimal and visually secondary.

Setting and style: A spring school entrance lined with dense cherry blossom trees, falling petals, stone path, ornate iron gate, bright backlight from the gate, detailed manga hatching, screentones, speed lines, and cinematic perspective. The overall style should resemble a polished Japanese shoujo manga page, with expressive poses, delicate hair rendering, dramatic lighting, and clean page composition.

Constraints: Keep the page monochrome except for the cyan glow on the glasses character. Keep the blurred gray face blocks visible on the twin-tailed girl in panels 1 and 4. Do not add extra panels, extra main characters, watermarks, logos, or English text. Use {argument name="school uniform style" default="Japanese sailor uniforms with bow ties and pleated skirts"} and a {argument name="season" default="spring cherry blossom season"} atmosphere.

Goal: Create a cute pastel Japanese personality-result infographic for an “animal character donut diagnosis,” with a magical dessert café atmosphere and a cat-themed result.

Canvas: Vertical 3:4 poster, soft pink and lavender palette, glossy kawaii style, dreamy bokeh background of donuts and macarons, sparkles, tiny flowers, ribbons, moons, stars, and pearl-like highlights throughout.

Main title area: At the top, use large rounded bubble lettering with a white outline and pink-purple gradient reading {argument name="headline text" default="どうぶつキャラドーナツ診断"}. Under it, smaller warm yellow text reads “あなたには”, then a huge central result word reading {argument name="result animal" default="ネコ"}, followed by smaller text “の気配が漂っています”. Decorate the title with exactly 6 prominent small motifs: 2 pink flowers, 2 small stars, 1 butterfly, and 1 ribbon.

Layout: Split composition with text cards on the left and hero visuals on the right. The left column contains exactly 5 translucent rounded white cards with soft pink borders and tiny icon decorations. The right side contains exactly 2 main visual elements: one circular anime portrait frame above and one cat-shaped donut on a plate below.

Left column text cards: Card 1 is titled “理由は” and contains a short explanation saying the person has a mysterious refined mood, coolness but also sweetness, and a cat-like presence. Card 2 is titled “ネコの特徴” and lists exactly 5 bullet points: free-spirited and goes at their own pace; cautious but affectionate; aesthetically sensitive and elegant; sweet only with trusted people; observant and clever. Card 3 is titled “相性が良い動物” and contains exactly 2 compatibility animals with simple line icons: “オオカミ” with a note about respecting independence and forming strong bonds, and “ウサギ” with a note about a gentle soothing personality. Card 4 is titled “運勢” and lists exactly 5 fortune lines: overall luck, love luck, work luck, lucky item, and lucky color. Card 5 is titled “ネコドーナツについて” and explains that the donut is coated in cocoa and white chocolate, has shiny purple glaze, fluffy ears and tail, and is a special donut packed with cat charm.

Right circular portrait: Create an ornate round frame filled with sparkles, pearls, flowers, crescent moon charms, and a small ribbon. Inside is a mysterious anime-style catlike girl with {argument name="hair color" default="silvery lavender hair"}, dark gothic clothing, purple gemstone accessories, luminous highlights, and a dreamy night-sky glow. Cover the center of her face with a plain opaque gray square censor block, matching the generated composition.

Cat donut hero: In the lower right foreground, show one large cat-shaped donut on a delicate white plate with gold trim. The donut has a cream-and-lavender icing design, purple glaze dripping around the bottom, glitter sugar, exactly 2 upright cat ears, exactly 1 curled purple tail on the right, exactly 1 purple bow on the upper right ear, exactly 2 large glossy purple eyes, exactly 3 whisker lines on each cheek, a tiny smiling mouth, small blush cheeks, and scattered edible pearls. Add a small gold fork at the lower right of the plate.

Visual style: Ultra-cute kawaii dessert illustration, glossy 3D-anime hybrid, high detail, soft glow, pastel pink background, magical sparkles, clean readable infographic typography, rounded Japanese fonts, white text boxes, delicate decorative borders, feminine and sweet. Use {argument name="main color palette" default="pastel pink, lavender, cream, and gold"}.

Constraints: Keep all Japanese text legible, do not add extra cards or extra main characters, maintain exactly 5 left-column cards, exactly 2 compatibility animals, exactly 1 cat donut, and a polished social-media diagnosis result poster look.

Create a highly detailed cinematic cyberpunk anime illustration of {argument name="vehicle type" default="an enormous retrofuturistic air bike"} flying through a dense vertical megacity at night. The scene shows exactly two riders seated side by side on top of the machine: rider 1 is a shorter schoolgirl with short dark purple hair, wearing a navy sailor-style school uniform with a red scarf, white knee socks, and dark shoes, holding the handlebars; rider 2 is a taller schoolgirl with long straight black hair, wearing a dark blazer school uniform with a red bow, plaid skirt, black knee socks, and dark shoes, sitting close behind/alongside her. Their faces are softly obscured or out of focus, while the costumes, hair, and vehicle are sharp. The air bike is massive, horizontal, and packed with exposed brass and gunmetal machinery: pipes, rivets, pressure tanks, antennae, cables, gauges, vents, mechanical arms, small panels, decals, and glowing displays. Include exactly four visible purple-blue glowing levitation/engine rings: one large circular thruster at the front-left nose, one large underside ring near the front center, one underside ring near the middle-rear, and one large rear-right circular thruster. Add bright electric arcs around the levitation rings and magenta/cyan light spilling onto the metal. The front has a round warm headlamp, red indicator lights, a weathered rectangular cockpit housing, and a dish-like antenna frame above it. Use a dark futuristic Japanese city background filled with towering skyscrapers, deep perspective, rain-haze atmosphere, and many neon signs in pink, violet, cyan, and blue; signs may contain stylized Japanese kanji/kana-like text, including a prominent large pink sign high on the right and smaller vertical signboards scattered throughout. Overall mood: dense, gritty, ornate, maximalist, steampunk-meets-cyberpunk, nocturnal, dramatic low-angle three-quarter view, the vehicle dominating the frame and hovering above the city. Color palette: blackened metal, bronze, deep navy, magenta neon, violet plasma, cyan highlights, and warm amber lamps. Render in ultra-detailed anime concept art style with crisp mechanical linework, cinematic lighting, volumetric fog, and no watermark. Customize the main neon signage phrase as {argument name="main neon sign text" default="八咫烏"}, the rider uniforms as {argument name="school uniform style" default="dark Japanese school uniforms with red accents"}, the city mood as {argument name="city atmosphere" default="rainy neon night megacity"}, and the lighting accent color as {argument name="plasma glow color" default="purple-blue"}.

Create a highly realistic cinematic sports photograph of a nighttime professional baseball game in a packed stadium, shot from a low wide-angle position just behind home plate on the third-base side. The scene shows exactly 7 visible on-field figures: 1 right-handed batter in a gray away uniform and black helmet raising a wooden bat, 1 catcher in navy protective gear crouched behind the plate, 1 black-uniformed home plate umpire leaning in behind the catcher, 1 pitcher in a navy jersey and white pants mid-delivery on the mound, and 3 distant fielders spread across the infield/outfield. Use {argument name="team colors" default="gray away uniforms versus navy home uniforms"}. The ball should be visible in flight between pitcher and batter. Include the dirt batter's box, white chalk lines, green grass, pitcher's mound, outfield wall, and a huge crowd filling the grandstands under bright stadium floodlights. Place a large electronic scoreboard in the upper right with inning numbers, scores, and dense Japanese-style team/player text, glowing realistically but not needing to be fully legible. Mood: dramatic, tense, heartwarming live-game atmosphere, like a decisive late-inning pitch. Use natural stadium lighting, deep night sky, photorealistic detail, slight motion energy, crisp foreground, and a wide 16:9 composition. Customize the batter as {argument name="main batter description" default="a powerful right-handed hitter in gray with black sleeves"}, the stadium as {argument name="stadium location" default="a large Japanese baseball stadium"}, the scoreboard score as {argument name="scoreboard score" default="1 to 5"}, and the camera style as {argument name="camera style" default="professional sports photojournalism with cinematic realism"}. Avoid logos, watermarks, readable brand names, extra players beyond the specified 7, cartoon styling, or empty seats.

Create a colorful single-panel anime travel meme postcard set in Kyoto. Canvas: landscape 4:3 composition with a thin black rounded border, like the first panel of a four-panel comic. In the top-left corner, place a small white rounded square label containing the black number “1”. Background: a bright spring Kyoto street scene with traditional wooden machiya houses, stone-paved road receding into the distance, a large multi-tiered pagoda near the center-right, Kyoto Tower visible far in the distance, blue sky, mountains, and pink cherry blossoms overhead with petals drifting through the air. Foreground: on the right half, draw an elegant anime-style young woman wearing a pale yellow floral kimono with a dark navy obi, posed like a welcoming guide with one hand extended toward the city. She has vivid {argument name="hair color" default="blue"} shoulder-length wavy hair and a decorative flower hair ornament with dangling gold tassels. Her face is intentionally covered by a large opaque square censor block with a vertical blue-gray to beige gradient, centered over the face and obscuring all facial features. Text: on the upper-left and center-left, add large hand-brushed Japanese calligraphy reading {argument name="headline text" default="京都へ おこしやす"}, with “京都へ” in dark red and “おこしやす” in dark navy, plus a red swoosh underline and two small flower icons. Style: polished GPT-image anime illustration, clean line art, saturated spring colors, soft sunlight, high detail in architecture and kimono pattern, cheerful tourism-poster mood. Constraints: include exactly one numbered panel label, exactly one foreground character, exactly one face-covering square, and do not add extra captions, watermarks, speech bubbles, or additional characters.

Create a photorealistic vertical fashion portrait of a young woman standing in a quiet underground train station corridor in the morning. She is centered slightly right of frame, full body from head to mid-thigh visible, wearing a fitted ribbed {argument name="dress color" default="white"} long-sleeve turtleneck mini sweater dress. Her long softly waved {argument name="hair color" default="rose-gold blonde"} hair falls over both shoulders, parted near the center. Her face is intentionally obscured by a smooth rectangular blur block matching warm skin tones, with no facial features visible. She stands beside a glossy white square-tiled wall on the right, lightly touching the wall with her right hand while her left arm hangs relaxed. The corridor recedes deeply into the background with wet reflective floor tiles, cool fluorescent ceiling lights, a blurred overhead station sign, and exactly 3 distant commuter silhouettes in the background. Use a realistic full-frame camera look with shallow depth of field, cool blue-gray subway lighting, subtle reflections, cinematic urban atmosphere, natural body proportions, high detail in fabric ribs and tiles, and a small pale sparkle/star watermark-like mark near the bottom-right corner. No readable text, no extra people in the foreground, no accessories, no bags, no visible face.

Create a photorealistic wide-angle night baseball game scene from a low camera position just behind home plate, as if shot with a professional sports camera. The main subject is a right-handed batter in a {argument name="batting team uniform" default="white and gray away uniform with black helmet"} standing in the left foreground, bat raised and ready to swing. Directly behind him are exactly 2 crouched figures: a catcher in dark navy protective gear and an umpire in black uniform leaning forward. On the field show exactly 4 additional visible players: the pitcher in a {argument name="pitching team uniform" default="dark navy jersey and white pants"} mid-delivery on the mound, one infielder near second base, one fielder on the far right, and one distant fielder near the outfield. The baseball should be visible in midair traveling from the pitcher toward the plate. Set the scene in a packed stadium at night with bright floodlights, green outfield walls, a brown dirt infield, crisp white chalk batter's box lines, and a lush green grass field. In the upper right background include a large illuminated Japanese-style scoreboard filled with inning numbers, team rows, colorful statistics, and dense small Japanese text, without needing perfectly readable characters. The stands should be crowded with thousands of spectators wearing mostly light clothing, creating a realistic textured crowd. Use dramatic stadium lighting, deep shadows, high ISO sports photography realism, natural motion tension, sharp foreground detail, slightly softer distant crowd, cinematic 16:9 composition, no watermark, no added captions.

Goal: Create a wide Japanese social-media comparison thumbnail about how to use three image-generation AI tools, with a clean tech infographic look and a presenter holding example phone screens.

Canvas: 16:9 horizontal banner, dark navy blue background with subtle dotted patterns and small sparkle accents. Use a large rounded white headline panel across the upper left and center, leaving the right side for the presenter. Bright, high-contrast, polished YouTube/X thumbnail style with crisp vector text, soft shadows, rounded cards, and glossy 3D icons.

Main headline area: In the large white rounded rectangle, place bold Japanese headline text: {argument name="headline text" default="画像生成AI 3ツール\nどう使い分ける？"}. Make “3ツール” extra large and dark blue, with the rest in heavy black. Under it, add a pale gray rounded sub-box containing two lines: “全部最強！！” and “用途で分かれる時代に、それぞれの強みを整理。”

Top decorative icons: Include exactly 2 floating 3D icons near the top right of the headline panel: 1 white artist palette with 5 colored paint dots, and 1 rounded square AI app icon with the letters “AI” in purple-blue neon.

Comparison section: Place exactly 3 rounded white tool cards aligned horizontally across the lower left/middle, each with a colored border, an icon, a title, and two labeled rows. Card 1 has a green border and chat bubble icon, title “GPT image 2”, label “強み” with text “会話で細かく修正”, label “用途” with text “文字入り画像・図解”. Card 2 has a yellow border and star icon, title “Nanobanana 2”, label “強み” with text “日本語サムネ一発仕上げ”, label “用途” with text “SNS用サムネ・一貫キャラ”. Card 3 has a purple border and simple sailboat line icon, title “Midjourney”, label “強み” with text “雰囲気重視のビジュアル”, label “用途” with text “アート系・写実的ドラマ”. Use green/yellow/purple pill badges for the labels and keep all Japanese text bold and readable.

Presenter and phones: On the right side, show a muscular casually dressed male presenter in a fitted white T-shirt and black pants, cropped from head to thighs, standing in front of the navy background. His face should be intentionally covered by a plain skin-tone rectangular blur/censor block. He holds exactly 3 smartphones fanned in his hand. Phone 1 is labeled “GPT image 2” and shows an investment beginner infographic with charts and Japanese text; phone 2 is labeled “Nanobanana 2” and shows a Japanese morning-routine thumbnail with a person and bold yellow text; phone 3 is labeled “Midjourney” and shows a cinematic fantasy landscape with mountains and dramatic lighting. The phones should overlap slightly and have glossy black bezels.

Bottom banner: Add a long rounded pale yellow banner across the bottom with a lightbulb icon on the left and bold Japanese text: {argument name="bottom message" default="1つに絞らんでええ。3つ揃えて、プロの作業環境に。"}. Emphasize “3つ揃えて” in dark blue with a yellow underline.

Visual style: Japanese creator thumbnail, modern infographic, clean rounded rectangles, thick bold typography, high readability, dark blue/white/yellow/green/purple palette, subtle glow around cards, professional but friendly. Avoid extra cards, extra phones, watermarks, or unreadable small text.

Using REFERENCE_0 as the layout and gag concept, transform the ASCII-art scene into a polished full-color Japanese manga/cartoon illustration. Keep the same basic composition: a cat on top of an Ehime mikan cardboard box, striking a ramen bowl with chopsticks, with sound effects on the left and a speech bubble on the right.

Style: Convert the monochrome text-art into a lively hand-drawn anime illustration with clean ink lines, warm colors, expressive motion lines, comic impact bursts, and a white background. Make the cat cute but angry/frustrated, with flushed cheeks, open shouting mouth, and an orange sitting on its head.

Additions and details: Add a large ramen bowl in front of the cat, two chopsticks held like drumsticks, yellow impact sparks where the chopsticks hit the bowl, and a realistic cardboard fruit crate texture. Add exactly 5 loose whole oranges around the scene: 1 on the cat’s head, 1 in front of the box near center, 1 at bottom left, 1 at bottom center, and 1 at bottom right. Also show 2 printed oranges on the front of the cardboard box.

Legal-complaint theme additions: Add stacks of documents and complaint paperwork around the box, including a large pile at lower left labeled 「訴状」 with a red stamp, loose sheets in the foreground stamped 「提出済」, and a thick dark blue legal book at lower right labeled 「民事訴訟法」 with a sticky note reading 「いつまで待たせる気なの？」. Include scattered papers, tabs, envelopes, and crumpled paper balls to make the scene feel chaotic.

Text to preserve or recreate: Keep the box label as {argument name="box label" default="愛媛みかん"}. Keep two left-side star sound effects reading 「チン」. Keep the upper-right chant text as {argument name="chant text" default="マチクタビレタ〜\nマチクタビレタ〜"}. Keep the right speech bubble text as {argument name="speech bubble text" default="Grok集団訴訟\nマダ〜"}. Use bold, legible Japanese lettering, not ASCII characters.

Constraints: Do not keep the ASCII-art look; fully reinterpret it as a colorful comic illustration while retaining the original joke, layout, and main Japanese text.

Create a polished anime-style half-body portrait of an extremely handsome adult man standing in an upscale high-rise bar at night, with a dramatic city skyline visible through tall windows behind him. He has {argument name="hair color" default="platinum blond"} layered tousled medium-length hair with long side bangs and sharp luminous highlights, an elegant neck and broad shoulders, and a confident fashionable presence. His face should be mostly obscured by a centered square privacy censor block in a warm beige-to-dark-gray vertical gradient, covering the eyes, nose, and mouth while leaving hair, ears, jawline edges, neck, and upper torso visible. Dress him in a cream-white tailored blazer over an unbuttoned black dress shirt with exactly 3 visible gold buttons, open low enough to reveal a toned chest and collarbones. The background should show exactly 2 main environment zones: on the left, a nighttime skyscraper cityscape with many blue-white window lights and red antenna dots; on the right, a warm luxury bar shelf area with dark wood, golden lighting, bottles, and a glowing rectangular wall sconce. Use cinematic warm indoor lighting mixed with cool blue city light, crisp clean line art, soft cel shading, glossy highlights, a slightly romantic host-club atmosphere, vertical portrait composition cropped at mid-torso, no text, no watermark, no extra characters.

Create a vertical anime illustration of exactly 3 elegant young women in a Taisho romantic cafe setting, a refined Japanese-Western fusion interior filled with warm lantern light, dark carved wood, stained-glass windows, lace curtains, antique furniture, framed paintings, and vases of flowers. The three women wear ornate wa-lolita maid cafe outfits combining kimono sleeves, floral-patterned haori, ruffled white aprons, black lace trim, ribbon bows, and frilled maid headdresses. Character 1 is seated on the left, with very long black hair and a deep red floral kimono-style outfit; Character 2 stands behind them in the center, with short {argument name="hair color" default="lavender"} bob hair, a purple floral kimono jacket, white ruffled blouse, black gloves, and a gentle smile; Character 3 is seated on the lower right, with very long black hair and a teal-blue floral kimono-style outfit. Compose them closely together like a formal cafe portrait, with graceful poses and delicate hands. Use a highly detailed modern anime style with soft painterly shading, intricate linework, muted jewel tones, cream lace, gold accents, and nostalgic early-20th-century Japanese romantic atmosphere. Add an ornate decorative border around the image with large cream and red peonies, flowing gold vines, teal and burgundy patterned panels, and art-nouveau curves. Keep the faces beautiful and expressive, no blur or censoring, no text, no watermark, no extra characters.

Create a serene anime-style vertical portrait of {argument name="character name" default="Ami"}, shown from the chest up in a three-quarter back view, standing outdoors under a vivid blue summer sky with large soft white cumulus clouds. The character has short tousled {argument name="hair color" default="pale blonde"} bobbed hair with fine windblown strands, detailed highlights, and a softly glowing rim light from the sun. Their face is intentionally hidden by a centered opaque square censor block in a warm skin-tone beige, covering the entire facial area while leaving the hair, ear, neck, and silhouette visible. They wear an oversized {argument name="shirt color" default="light sky blue"} T-shirt with loose folds and a white butterfly-wing motif printed across the upper back, resembling delicate angel or butterfly wings. Use clean high-quality Japanese anime illustration style, crisp linework, soft cel shading, airy bright lighting, gentle pastel color palette, realistic fabric creases, and a calm dreamy mood. Composition should be portrait orientation, close framing, character slightly left of center, looking toward the right side of the image, with no text, no watermark, and no extra characters.

Goal: Create a vertical hand-drawn graphic recording infographic in Japanese about {argument name="main topic" default="VPS @ Hetznerで動く Hermes Agentの全貌"}, with a friendly automation-kingdom theme and a central mascot robot.

Canvas: Portrait 2:3 poster, warm cream paper background, watercolor marker texture, dark navy handwritten lettering, loose sketch lines, pastel accent colors, small sparkles, arrows, doodles, and imperfect hand-drawn boxes.

Header: At the very top, draw a sunrise/cloud doodle on the left and a small castle on a hill on the right. Large title text reads “VPS @ Hetznerで動く Hermes Agentの全貌”. Under it, add the subtitle “— 佐藤さんの自動化王国” with a small crown icon.

Central subject: In the middle, draw a cute Hermes-themed robot mascot wearing a winged helmet, headphones, white rounded armor, small wing decorations on boots, and a teal cape. The robot smiles through a dark glass face screen and holds a clipboard labeled “TASK”. Around it, place circular time arrows and labels “0:00”, “6:00”, “12:00”, and “18:00”. Beneath the robot, add a nameplate reading “Hermes Agent ⚚” and a dark teal stats plaque reading “265 skills / 50 jobs / 24/7 running on Hetzner VPS”.

Main schedule layout: Use exactly 3 large schedule panels around the mascot. Left panel is orange and titled “朝（早朝・午前）JST”; it contains exactly 4 task rows with icons: 1) newspaper icon, “16:30 — ai-news-morning / 新聞 → Slack #実験”; 2) clipboard icon, “17:00 — meeting-briefing / MTG準備（平日のみ）→ Slack”; 3) rising chart icon, “17:07 — market-intel-morning / 株価グラフ → Slack”; 4) blue bird icon, “18:07 — x-engagement-daily / X バズ分析 → Slack”. Upper right panel is green and titled “日中 JST”; it contains exactly 1 task card with checklist icon: “9–21時 / 2h — Slack-task-detector / Slack→Linear 自動タスク化（平日）”. Right panel is blue and titled “夕方・夜 JST”; it contains exactly 6 task rows with icons: 1) moon and chart, “03:00 — market-intel-evening / 市場夜レポ”; 2) pink brain, “03:00 — metacognition-daily / メタ認知振り返り”; 3) envelope alert, “04:07 — slack-unanswered / 未返信 Slack 督促”; 4) candlestick chart, “06:00 — kaichi-trader-daily / 米株 paper trade / ← 今日NEW!!（平日）”; 5) calendar, “06:07 — calendar-preview / 今日の予定プレビュー”; 6) bar chart, “日曜 05:07 — x-weekly-growth / X 週次成長レポ”.

Background services section: Across the lower third, add a ribbon title “Background Services（常駐24/7）” with a crossed-tools icon and stars. Below it, create exactly 5 vertical service cards, each with a logo-style icon, service list, down arrow, and destination box. Card 1: “Slack:” with multicolor Slack-like dots, services “Conversation Hub / Bridge / MCP Server / Approval Listener”, destination “→Slack DM / 佐藤への日次レポ” with a small face doodle. Card 2: “Discord:” with purple gamepad logo, services “Rate Limit / Reaction Listener”, destination “→Discord / チャンネル”. Card 3: “LINE:” with green LINE icon and small bear/family doodle, services “Log Bot / Webhook Relay / Cloudflare Tunnel”, destination “→LINE / グループログ”. Card 4: “X/Twitter:” with yellow X badge, blue bird holding a pencil, service “Draftflow”, destination “→GitHub / Private repos” with a lock icon. Card 5: “ビジネス:” with red target icon, services “AIFluence / RepoSwipe / AI Office Worker / OwnTracks / ngrok / SSH Portfolio TUI”, destination “→Linear / タスク自動登録” with checklist icon.

Visual style: Japanese sketchnote, analog marker and colored pencil, playful but information-dense, clean enough to read, charming imperfect typography, cream background, navy ink, orange/green/blue/purple/red outlined boxes, small decorative arrows and sparkles. Keep all visible text handwritten and preserve the Japanese/English mixed labels exactly where specified.

Constraints: Use exactly 3 main schedule panels, exactly 11 scheduled task rows total, exactly 5 background service cards, and exactly 5 destination boxes. Do not add extra brands, watermarks, QR codes, photorealism, or typed sans-serif UI styling.

Create a quiet anime-style winter portrait of exactly 1 young woman standing outdoors in a snowy leafless forest. The subject has {argument name="hair color" default="glossy black"} shoulder-length straight bob hair with a center part and detailed highlights, wearing exactly 1 oversized {argument name="sweater color" default="pale gray"} ribbed turtleneck sweater with loose sleeves and fine knit linework. Frame her from mid-torso upward, centered slightly right, facing forward, with her face intentionally hidden by a soft-edged rectangular blur/censor block in muted beige and gray tones. The background is a calm winter park with white snow covering the ground and many evenly spaced bare trees receding in perspective; include long narrow trunks, subtle bark texture, faint shadows on the snow, and cool blue-gray atmospheric haze in the distance. Use clean Japanese anime illustration rendering with delicate ink outlines, soft natural daylight, muted winter colors, realistic fabric folds, and high-detail hair shine. No text, no watermark, no extra people, no animals.

Create a polished anime-style portrait illustration of {argument name="character name" default="Ami"}, shown from mid-thigh to just above the head, standing indoors in a warm traditional Japanese room. She has a short glossy {argument name="hair color" default="black"} bob haircut with a center part and slightly tousled ends, but her face is deliberately hidden by a centered opaque square censor block in a soft brown skin-tone gradient, covering the eyes, nose, and mouth. She wears a cream-white yukata with exactly 11 visible red flower motifs: 1 on the upper right chest, 1 on the left chest near the crossing collar, 1 on the right sleeve, 1 on the far left sleeve edge, 1 on the lower right sleeve cuff, 1 small flower on the right sleeve, 1 on the lower left sleeve, 1 on the lower skirt left, 1 on the lower skirt center, 1 on the lower skirt right, and 1 partial flower near the left shoulder area. The yukata has wide soft sleeves, realistic folds, and a crossed collar. Add a vivid {argument name="obi color" default="red"} obi belt around the waist with a pale braided cord tied in a bow at the front center. Behind her head place a large flat {argument name="background circle color" default="red"} circle like a sun disk, filling much of the upper background. The room background should be dark wood with softly blurred shelves and warm amber lighting, creating a cinematic depth-of-field effect. Use high-quality gpt-image-2 anime rendering, clean linework, detailed fabric shadows, glossy hair highlights, warm skin lighting on the neck and hands, vertical portrait composition, no text, no watermark.

Create a whimsical storybook-style watercolor illustration of a young female alchemist named {argument name="character name" default="Luna"} working at a wooden desk under a night sky motif. The character has long flowing {argument name="hair color" default="black hair with purple ombré ends"}, wears an ornate black, white, and violet fantasy alchemist outfit with gold trim, a purple ribbon brooch at the collar, layered cape, belts, pouches, and star decorations; her face is intentionally obscured by a soft vertical gray blur rectangle with no visible facial features. She holds one purple liquid test tube in her left hand and steadies a golden armillary sphere with her right hand. The desk contains exactly 14 visible alchemy and study objects: 1 golden armillary sphere with star patterns, 1 open botanical book, 1 closed blue book underneath, 1 purple potion flask on the desk, 1 wooden test tube rack, 5 test tubes in the rack containing clear, purple, blue, yellow, and pale liquid, 1 small glass vessel near the rack, 1 lavender triangular cloth runner with gold stars and tassels, 1 small star charm, and 1 wooden tabletop. On the left shelf area show exactly 7 glass laboratory containers: 3 tall stoppered bottles, 1 round flask with pink liquid, 1 small bottle with purple liquid, 1 lidded jar, and 1 narrow vial. In the lower left foreground show exactly 1 cream ceramic pitcher filled with wildflowers, including lavender sprigs, purple pansies, peach flowers, white blossoms, green leaves, and trailing vines. In the background show exactly 1 crescent moon, 8 small golden stars, and 1 soft cluster of pale lavender clouds. Use delicate ink outlines, airy white background, transparent watercolor washes, pastel purple and gold accents, hand-drawn children’s book fantasy texture, light paper grain, and a cozy magical laboratory mood. The overall palette should emphasize {argument name="accent color" default="lavender purple"}, warm gold, cream, and muted brown wood. Vertical composition, centered character and desk, no hard shadows, no photorealism, no extra text, no watermark.

Create a cinematic photorealistic medieval fantasy scene titled {argument name="scene title" default="Audience with the King"}: inside a vast Gothic throne hall, a seated king is seen from behind in the lower right foreground, facing two dignified petitioners standing before him. The two central figures are a female knight in full polished silver plate armor with ornate engravings, chainmail details, gauntlets, boots, sword at her side, and a pale white cloak, and a tall prince or nobleman wearing an immaculate white military-style ceremonial suit with gold embroidery, gold buttons, epaulettes, braided chains, belt, and a white cape. The nobleman has long dark wavy hair; the knight has blonde hair tied up. Count exactly 3 main foreground characters: the king, the armored female knight, and the white-clad nobleman. Count exactly 8 background guards or courtiers, arranged symmetrically in dark armor and formal medieval dress along both sides behind the central pair, holding spears or standing at attention. The king wears a dark navy and gold brocade robe and an ornate golden crown, seated on a black-and-gold carved throne with gilded finials and scrollwork. The hall has towering pointed arches, a bright stained-glass lancet window centered in the background, tall dark blue banners with gold heraldic crests, candlelit candelabras, smoky atmospheric light rays, marble columns, and a glossy black-and-white patterned stone floor reflecting the figures. Use dramatic backlighting from the window, warm candle highlights, deep shadows, shallow cinematic depth of field, ultra-detailed costumes, realistic fabric, metal reflections, and a solemn royal-audience mood. Composition should be vertical 4:5, eye-level from slightly behind the king, with the two petitioners centered and full body visible. No text, no watermark, no modern objects.