GPT Image 2 prompts

A cinematic cyber-noir portrait of a {argument name="subject" default="rugged young man"} with {argument name="hair style" default="thick dreadlocks tied back"}, sharp beard, and an intense expression, standing in a {argument name="setting" default="narrow old European street"} at dusk. High-contrast lighting with vibrant teal-blue shadows and glowing orange highlights illuminating his face and jacket. Detailed comic-book and graphic novel style, dramatic ink lines, painterly textures, ultra-detailed hair strands, moody urban atmosphere,distant gothic rooftops fading into mist, dynamic perspective, poster art aesthetic, bold color grading, modern digital illustration, 4K, highly stylized, atmospheric and edgy.

A dreamy cinematic portrait of a {argument name="subject" default="young woman"} holding a bouquet of {argument name="flowers" default="soft pink and white roses"} near a sunlit window, long dark flowing hair partially covering her face, soft natural golden-hour lighting, warm cozy indoor atmosphere, {argument name="outfit" default="delicate pastel pink outfit"}, romantic and aesthetic mood, gentle expression, shallow depth of field, soft shadows, vintage tones, photorealistic, dreamy composition, warm sunlight streaming through curtains, high detail, soft focus, elegant feminine portrait.

Use the {argument name="subject" default="uploaded girl image"} as the subject. Recreate the exact pose, framing, studio setup, and warm monochrome aesthetic of the reference image. Replace the giant teddy bear with an {argument name="object" default="oversized plush panda"}. The girl is seated on the floor holding a smartphone, wearing a black turtleneck and beige outfit. Soft cinematic studio lighting, luxury editorial photography, ultra-realistic details, warm beige background, premium fashion-magazine look. Add the word "{argument name="text" default="NOOR"}" in large elegant aesthetic cursive typography across the background, subtly blended into the backdrop. Hyper-realistic, 8K, soft shadows, clean minimalist composition, highly detailed panda fur, professional studio portrait.

Design a high-quality poster providing information about the weather in {argument name="location" default="Istanbul"} on {argument name="date" default="June 2, 2026"}.

A stunning ultra-realistic {argument name="subject" default="luxury coffee advertisement"} featuring a {argument name="character" default="beautiful young woman"} holding an {argument name="item" default="elegant black coffee cup"}, sitting beside a cozy café window on a rainy evening. Golden city lights reflected on wet streets, cinematic bokeh, premium fashion styling, natural smile, rich coffee atmosphere, luxury branding aesthetic, warm amber lighting, photorealistic skin texture, high-end commercial photography, shallow depth of field, luxury lifestyle magazine quality, sophisticated urban environment, coffee beans and premium packaging in foreground, dramatic reflections, ultra detailed, 8K, HDR, masterpiece advertising campaign, viral social media marketing image, professional luxury brand promotion.

Please turn this {argument name="source" default="Google Map"} screenshot into a beautiful {argument name="output style" default="illustration"} while faithfully reproducing the buildings.

Transform the reference photo into an authentic 1990s anime character portrait. Preserve the person's recognizable facial features, bone structure, expression, and identity, while adapting them to classic 90s anime aesthetics.

Inspired by the cool, rebellious male

Goal: Create a gritty black-and-white comic storyboard page for a pirate game scene, like a scanned manga/graphic novel production panel.

Canvas: Vertical comic page crop, grayscale only, with a dark surrounding workspace/background and a white page border. Add a small production label at the top left reading {argument name="panel code" default="A6 - 4"}. The main visible panel occupies almost the entire canvas, with a thick white gutter and the top edges of two smaller panels barely visible at the bottom.

Main panel layout: A rain-lashed pirate ship deck during a violent thunderstorm. Center the composition on one tall pirate captain in a dramatic low-angle pose, standing full body from thighs upward, fist thrust toward the viewer. His face is obscured by a soft rectangular gray blur/censor box. He has long wet curly hair, a heavy black captain’s coat with many buttons, layered shirt and sash, belts, chains, pistols or tools at the waist, and ragged wet fabric. Behind him is a rowdy crew of exactly 5 visible pirates: 1 shouting pirate on the left midground partly covered by a gray blur box, 1 hunched pirate at the lower left partly covered by a gray blur box, 1 shouting pirate just right of center behind the captain, 1 hat-wearing pirate near the right side, and 1 bearded pirate at the far right holding a flintlock pistol with a gray blur box over part of his torso/face area. Add exactly 1 large parrot or raven-like bird flying in the upper right with wings spread. Include ship rigging, wooden beams, ropes, lantern glow on the left, rain streaks, sea spray, lightning, and chaotic crowd energy.

Text and effects: Include exactly 3 speech balloons and exactly 3 sound-effect captions. Speech balloon 1, upper left, says “AWAY, ME SEA-DOGS! TO WRECKER’S REACH!!”. Speech balloon 2, right center, says “AYE! TO THE HEARTSTONE!!!”. Speech balloon 3 is only partially visible in the bottom cropped panel and begins “YOU’D BETTER BE,”. Sound effect 1, large bold tilted letters on the left, says “AAAYE!”. Sound effect 2, upper right near lightning, says “HEARTSTONE!”. Sound effect 3, near the bird, says “BRAWK!”. Use hand-lettered comic typography, white speech balloons with black outlines, bold outlined sound effects, and dynamic placement.

Visual style: High-detail monochrome ink illustration, dense crosshatching, wet glossy blacks, harsh contrast, splashy white highlights, scratchy rain lines, cinematic fantasy pirate action, dramatic perspective, heavy texture, energetic comic-book rendering, no color except grayscale.

Customizable details: The pirate destination can be {argument name="destination name" default="Wrecker’s Reach"}; the magical object or rallying cry can be {argument name="relic name" default="Heartstone"}; the captain’s command can be {argument name="captain command" default="AWAY, ME SEA-DOGS! TO WRECKER’S REACH!!"}; the crew response can be {argument name="crew response" default="AYE! TO THE HEARTSTONE!!!"}.

Constraints: Keep the image as a single dramatic comic page crop, preserve exactly 5 visible crew pirates plus the central captain and exactly 1 bird, preserve exactly 4 soft gray censor/placeholder rectangles over characters, preserve the thick white comic borders, and do not add extra readable text beyond the specified labels, balloons, and sound effects.

Goal: Create a cinematic futuristic fashion advertising poster for a fictional urban performance brand called {argument name="brand name" default="SNITCH"}, with the tagline {argument name="brand tagline" default="MOVE WITH INTENT."} and the main campaign headline {argument name="headline text" default="OUTRUN ORDINARY."}.

Canvas: Vertical poster, 4:5 aspect ratio, high-resolution luxury streetwear campaign look. The scene is set at dawn in a dense glass-and-steel megacity resembling Mumbai/BKC, with wet reflective pavement, glowing skyscraper windows, warm sunrise flare between towers, and a sleek elevated pedestrian bridge leading into the city.

Main subject: One young male urban runner/model walking toward the camera at center, full body visible, confident forward motion, head slightly turned upward to the left. His face is intentionally anonymized with a soft rectangular blur over the facial features. He has short curly dark hair. Outfit: dark charcoal technical bomber jacket, dark layered performance shirt, black crossbody sling bag, loose black cargo pants with zipper pockets, and gray high-top tactical sneakers. Pose: left hand gripping the jacket or bag strap near the chest, right arm relaxed, one foot stepping forward.

Layout and typography: Place a large white distressed condensed sans-serif headline in the lower-left foreground reading {argument name="headline text" default="OUTRUN ORDINARY."}. Beneath it, add the subheading "For the ones building tomorrow." Add a thin outlined button below: "ENTER THE NEXT MOVE" with a small diagonal arrow icon. Use premium techwear typography, all caps, tracking, micro-labels, HUD-style brackets, white text with subtle transparency.

Branding and visible text elements: Top left contains a geometric monogram logo plus the word {argument name="brand name" default="SNITCH"} in wide-spaced white letters, with a small trademark mark. Under it: {argument name="brand tagline" default="MOVE WITH INTENT."}. Top right contains "$S25" and three stacked product words: "PERFORMANCE", "LUXURY", "CORD SET". Add one large billboard on the right tower showing the brand monogram, "SNITCH", and the slogan "THE CITY. NEVER STOPS. NEITHER DO YOU." Add another blue billboard reading "NEXT MOVE". Add a smaller left-side billboard reading "BUILD BEYOND LIMITS".

HUD data panels: Include exactly 5 left-side data lines inside subtle corner-bracket framing: 1) "TIME 05:47 AM", 2) "ELEVATION +370M", 3) "DISTRICT BKC // MUMBAI", 4) "COORDINATES 19.0596° N 72.8697° E", 5) "FOCUS BUILD ELEVATE INSPIRE". Include exactly 5 right-lower status labels stacked vertically: 1) "TRAFFIC FLOW INTENSE", 2) "MINDSET RELENTLESS", 3) "MOVEMENT FORWARD", 4) "STATUS IN PROGRESS", 5) a small downward minimal marker/icon.

Footer elements: Lower left includes exactly 3 small circular icons/logos followed by the line "SNITCH // URBAN PERFORMANCE LAB". Below that, add a barcode with the tiny code "SS25-NEXTMOVE-2025".

Visual style: Photorealistic cinematic advertising, premium streetwear campaign, moody dark teal and charcoal palette with warm gold sunrise highlights, wet asphalt reflections, strong depth, lens flare, realistic city scale, sharp foreground subject, crisp typography integrated into the environment, subtle sci-fi interface overlays. Lighting should rim-light the model and emphasize reflections on the bridge railing and floor.

Constraints: Use exactly 1 central human figure, exactly 3 prominent billboards, exactly 5 left HUD data lines, exactly 5 right-lower status labels, exactly 3 footer icons, and no extra main characters. Keep the poster polished, aspirational, and motivational.

Create a dramatic vertical anime fantasy key visual of {argument name="character name" default="Hotaru"}, a powerful young female warrior standing full-body in a ruined futuristic city at night. She has very long flowing {argument name="hair color" default="silver-white"} hair whipping outward in the storm, an athletic elegant build, and wears an ornate white, teal, and gold magical battle dress with asymmetrical armored bodice, layered short skirt, thigh-high dark stockings, white high-heeled boots with teal crystal accents, delicate gloves, and jewel-like metallic details. Her face should be intentionally obscured by a soft square blur or neutral censor block, while the rest of the image remains highly detailed. Pose her in a commanding wide-legged stance from a low-angle perspective, one arm extended outward to the left as if unleashing power and the other arm lowered with a clenched fist. Behind her, place a gigantic translucent crystalline winged crest or phoenix-like armor halo made of sharp emerald glass shards, centered behind her head and shoulders. Surround the character with roaring circular streams of {argument name="energy color" default="neon emerald green"} magical fire, lightning-like particles, sparks, and glowing plasma arcs forming a vortex around her body. The background is a dark post-apocalyptic sci-fi city with broken towers, ruined scaffolding, distant silhouettes, and reflective cracked metal ground, all backlit by intense green light. Use cinematic composition, high contrast, volumetric glow, rim lighting, deep shadows, floating embers, dynamic motion, ultra-detailed anime illustration style, glossy fantasy armor rendering, sharp linework, painterly effects, 3:4 vertical poster framing, epic scale, no text, no watermark.

Using REFERENCE_0 as a rough composition and camera-position guide, turn the sketch into a cinematic steampunk scene: two brass mechanical hands in the foreground hold a worn, sepia-toned old photograph at the same angle and position indicated by the reference. Set the outer scene inside a vintage train carriage with dark leather seats, warm wood trim, a window on the left, and moody amber lighting. Inside the photograph, show an antique street scene in front of a clockwork repair shop: exactly 2 main subjects, a kneeling adult inventor with a blurred/obscured face and a small round-bodied robot standing beside them. Add aged photo details including creases, scuffed corners, yellowed borders, and faded monochrome-brown coloration. Keep the overall look photorealistic, cinematic, detailed, and nostalgic, with shallow depth of field outside the photograph. Customizable details: {argument name="outer setting" default="vintage train carriage"}, {argument name="photo subject setting" default="clockwork repair shop street"}, {argument name="main human role" default="adult inventor"}, {argument name="companion character" default="small round-bodied robot"}, {argument name="color mood" default="warm sepia steampunk"}. Constraints: preserve the reference composition of hands framing a tilted rectangular photo; include exactly 2 visible mechanical hands holding exactly 1 photograph; include exactly 2 main figures inside the photo; avoid modern objects, clean digital UI, or extra text beyond shop signage.

Create a vertical 2:3 Japanese ukiyo-e inspired anime illustration of {argument name="character name" default="Oda Nobunaga"} as a stylish mature warlord, shown from the waist up in a dignified three-quarter portrait. He has a tall messy black topknot with loose strands and a gold hair ornament, one visible ear, and his face is intentionally obscured by a centered opaque tan rectangle covering the full facial area from forehead to chin. Dress him in ornate Sengoku-era armor and layered kimono: a high black collar with gold trim, dark navy and black lamellar chest armor with red cords, two large gold chrysanthemum-like crests on the chest, red brocade shoulder panels with floral patterns, cream sleeves with purple and gold cloud motifs, dangling red tassels, and two decorated sword hilts crossing at the lower front. The background is parchment cream with an orange-red setting sun disk behind the figure, fine black ink landscape lines, stylized white cloud streaks, purple-gray ink wash clouds, halftone dot textures, and a small Japanese castle in the lower left distance. Frame the composition with thick black Art Nouveau-style rounded corner borders and spiral ornaments: exactly 6 prominent black spiral/circle ornaments, with 2 at the upper left, 2 at the upper right, and 2 at the lower corners. Use detailed linework, muted antique colors, gold accents, print texture, manga-meets-traditional-Japanese-poster aesthetics, and dramatic heroic posture. Add a small signature and date in the lower right reading {argument name="date text" default="June 2, 2026"} and {argument name="signature text" default="Oyagi"}. No other text, no watermark, keep the central tan face-cover rectangle perfectly flat and untextured.

Goal: Create a cozy Japanese children's TV theater still showing a puppet-style fantasy scene titled {argument name="show title" default="エルフィと機械の花畑"}, with a gentle storybook atmosphere and visible broadcast graphics.

Canvas: 4:3 landscape frame, warm cinematic lighting, shallow depth of field, detailed handcrafted miniature set.

Main scene: Inside a rustic cottage kitchen with clay-plaster walls, exposed wood beams, a wooden window on the left looking out to a bright yellow flower field and blue sky, shelves and hanging herbs in the background, and a small wooden table in the foreground. Two marionette characters sit at the table drinking tea: 1) a delicate elf girl puppet on the left with pale mint-white bobbed hair, long pointed ears, a simple cream linen dress, slim wooden puppet arms, and visible suspension strings attached above; her face area is softly blurred/obscured as if anonymized, 2) a large humanoid robot puppet on the right made of dark riveted metal plates, exposed cables, red lens eyes, bulky shoulder armor, articulated hands, and visible marionette rods/strings. The elf holds a small cup with both hands; the robot raises one hand as if speaking. On the table include exactly 4 visible food and tea items: a white teapot with blue floral pattern, one small cup in the elf's hands, one plate of pastries, and a rectangular cloth or tray under the teapot.

Broadcast overlay layout: Add exactly 5 graphic/text areas: 1) top-left digital clock reading {argument name="broadcast time" default="7:32"} in large white outlined numerals, 2) below it a small parchment weather card with Japanese text 「人形の国」 and 「きょうの天気」 plus a red sun icon, 3) top-right colorful logo reading {argument name="program logo" default="ぬくもり キッズシアター"} with flowers and rounded playful lettering, 4) bottom-left large wooden signboard title with flowers and vines reading 「エルフィと機械の花畑」 and subtitle 「〜やさしさがつなぐ未来〜」, 5) lower-right wooden next-story box reading 「▶つぎのおはなし」 and 「『とべない小鳥と風の魔法』」.

Additional bottom text: Across the very bottom, add the white Japanese tagline {argument name="tagline" default="やさしい心は、どんな未来も変えていける。"}. Near the lower-right above the bottom edge, add sponsor text 「＜提供＞」, 「木のおもちゃ工房」, and 「ひだまりの森」 with small flower decorations.

Visual style: Hyper-detailed stop-motion puppetry mixed with fantasy miniature photography, warm amber interior light, nostalgic NHK-style children's morning program, polished gpt-image-2 realism, tactile wood, cloth, metal, and paper textures. Keep all marionette strings visible and delicate. Use soft bloom around the window and cozy shadows in the cottage.

Constraints: Use exactly 2 seated characters, exactly 5 broadcast graphic/text areas, and exactly 4 table items as listed. Keep the scene wholesome and gentle, no extra logos beyond the described fake TV graphics, no modern appliances, no watermark.

Create a vertical close-up anime illustration of {argument name="character description" default="a teenage girl with long dark brown-black hair in twin braids"} against a clean white background. Frame her from the top of the head to the upper chest, with the head very large in the composition and slightly cropped near the top. She has glossy, messy, layered hair with loose strands around the sides, a visible left ear, and two thick braids falling over both shoulders. Her face is intentionally hidden by a large centered soft-edged square censor block in a muted brown-to-warm-gray gradient, covering the entire facial area from the forehead down to the chin while leaving the hair, ear, neck, collar, and hand visible. Show her wearing {argument name="outfit" default="a white school blouse with a pointed collar and a black ribbon bow tie"}, with subtle sketchy folds and pale shading. One hand is raised near the lower face, fingers partly tucked under the chin and disappearing behind the censor block. Use a refined Japanese anime style with delicate black linework, semi-realistic hair highlights, soft watercolor-like shading, muted colors, and a clean minimal background. No text, no watermark, no extra objects.

Create a vertical anime-style cyberpunk portrait set inside a glowing public aquarium, featuring {argument name="character description" default="a slim androgynous young man with messy silver-white layered hair"} seated in three-quarter profile facing right, with the face deliberately obscured by one large opaque muted lavender square centered over the head. The character wears {argument name="outfit" default="a glossy black high-collar futuristic jacket with zipper details, straps, and thin cyan neon light strips on the sleeves and back"}. Behind him is a floor-to-ceiling reef tank filled with saturated deep blue water, magenta and violet reflections, bright aquarium glass highlights, and dense coral scenery in pink, orange, red, lavender, and peach. Include exactly 22 visible fish: 5 orange-and-white clownfish clustered mostly along the left and lower-left reef, 2 bright blue tang fish on the left side, 1 larger orange fish near the center-left, 1 small yellow fish near the lower center, and 13 small pink-orange fish scattered across the upper and middle water. Include exactly 9 prominent coral clusters: 2 branching orange coral clusters on the left, 3 pink branching coral clusters along the lower half, 1 pale lavender coral cluster at the bottom-left, 1 glowing purple coral cluster near the top-left, 1 pale blue-white coral cluster behind the character, and 1 small orange coral cluster near the lower center. Use dramatic rim lighting from the aquarium, high contrast, luminous cyan and violet highlights on the jacket and hair, detailed strand-by-strand hair rendering, glossy glass reflections, rich saturated colors, cinematic depth, and an ultra-polished Japanese anime illustration style. Keep the composition portrait-oriented {argument name="aspect ratio" default="3:4"}, with the character occupying the right half and the reef tank filling the left and background. No text, no watermark, no photorealism.

Create a cinematic ultra-detailed fantasy ballroom scene titled {argument name="theme" default="The Ball"}: a full-body royal couple standing arm-in-arm in the center of an opulent candlelit palace ballroom. The woman wears a shimmering silver-gray embroidered ball gown with a wide floor-length skirt, long fitted sleeves, a low elegant neckline, delicate jewelry, and a flowing cape-like train; her curly blonde hair is styled in an elaborate updo. The man is tall with long dark wavy hair, wearing an ornate white ceremonial military-style suit with gold embroidery, gold chains, medals, a high collar, fitted trousers, white boots, a white cloak draped from his shoulders, and a decorative sword at his side. Their faces are intentionally obscured by plain soft brown rectangular masks with no facial features. Surround them with exactly 8 visible background guests in lavish dark and metallic formal gowns and court attire, all with similarly obscured rectangular faces where visible. The ballroom is grand and baroque, with towering gilded walls, blue-and-gold drapery, framed portraits, crystal chandeliers, many glowing candles, and a glossy black-and-cream marble floor reflecting the couple and candlelight. Use warm golden lighting, deep shadows, rich textures, realistic fabric embroidery, sparkling beadwork, cinematic depth of field, symmetrical composition, vertical portrait framing, and high-fashion historical fantasy realism. Avoid modern objects, readable text, logos, or watermarks.

Create a vertical 2:3 cinematic anime-style illustration of a misty early-summer Japanese shrine approach in the rainy season, centered on ancient mossy stone steps climbing through a lush forest toward a wooden torii gate and a traditional shrine building half-hidden in pale morning haze. The scene is filled with blooming hydrangeas in blue, purple, lavender, and pink on both sides of the stairs, wet leaves, puddled stone, soft reflections, and scattered petals. A young girl, {argument name="character name" default="a nameless shrine visitor"}, stands near the middle of the staircase viewed from behind and slightly turned, wearing a white blouse, a lavender skirt, white socks, dark shoes, and a small shoulder bag; her long {argument name="hair color" default="black"} hair is decorated with small flowers. She holds a transparent umbrella with a faint floral pattern, catching delicate highlights from the sunbeams. Include exactly 3 black cats: one cat walking up the steps ahead of the girl with its tail raised, one cat standing near her feet on the stair below, and one cat sitting in the lower right foreground wearing a subtle red collar and looking toward her. Frame the path with old stone lanterns and engraved stone pillars, warm lantern glows, damp stone textures, and tall trees arching overhead. Use {argument name="time of day" default="soft morning after rain"} lighting with volumetric god rays filtering through the canopy, cool blue-green shadows, sparkling moisture, dreamy atmospheric depth, high detail, painterly realism, serene nostalgic mood, no modern buildings, no readable signage emphasized, no watermark.

Using REFERENCE_0 as the character base, create a consistent set of exactly 8 Japanese educational slide images about {argument name="topic" default="Claudeに『部下AI』を持たせる"}. Keep the character design, outfit, mask, hairstyle, and cyberpunk details consistent with the reference, but redraw the character in new poses and expressions for each slide.

Overall format: Each slide should look like a hand-drawn notebook/sketchbook page with a cream paper background, black rounded outer frame, left-side spiral binding, marker-style Japanese handwriting, red emphasis marks, arrows, check marks, small icons, and simple workflow diagrams. Use a warm analog note-taking style rather than a clean corporate deck. Maintain the same slide design language across all 8 images.

Character placement: Place the character as a presenter/mascot on the right side or lower-right area of most slides. Vary the pose naturally: surprised with both hands open, pointing at diagrams, holding a clipboard, sitting while reviewing, making an OK/check gesture, or explaining with one raised finger. Preserve the reference character’s identity and clothing.

Generate exactly these 8 slide topics:
1. Opening / overall overview: Title text {argument name="headline text" default="Claudeに『部下AI』を持たせる"}; subtitle {argument name="subtitle text" default="15分で仕事を自動化する5つのテンプレート"}; main message: 「部下AI」が開発の常識を変えた; include the five automation examples: コードレビュー, テスト作成, ドキュメント生成, 全部15分で自動化, and a simple checklist-like layout.
2. What is a sub-agent: Explain 「サブエージェントとは」 as a reusable AI role or specialist assistant. Include small role cards and arrows showing a main Claude agent delegating tasks.
3. Why it is needed: Explain 「なぜ必要か」 with a before/after comparison, emphasizing that repeatedly asking AI for the same tasks is inefficient and can be turned into templates.
4. Code reviewer: Title 「コードレビュー」; show a GitHub/code review style flow with files, comments, bugs, and a check mark.
5. Test writer: Title 「テストライター」; show a test-generation workflow with steps, file names, unit tests, and pass/fail style icons.
6. Security scanner: Title 「セキュリティスキャナー」; show risk levels CRITICAL, HIGH, MEDIUM, LOW and emphasize that the agent detects issues but does not automatically fix them.
7. Cost optimization: Title 「コスト削減の鍵：モデルの使い分け」; show a comparison that cheaper/smaller models can handle simple checks and that costs can become much cheaper, with bold red savings notes.
8. Start today / summary: Title 「今日から始める一歩」; show a simple 3-step process: create a template, place it in a Claude agents folder or workflow, ask for review/automation, then mark it complete with a big check.

Text style: Use Japanese text throughout. Make headlines large and bold, with red underlines and handwritten annotations. Keep diagrams simple and readable. Use only the listed 8 slide topics and do not add extra slides.

Constraints: Keep the same notebook layout, cream color palette, black border, left spiral binding, and consistent character identity across every image. Avoid photorealism, avoid 3D rendering, avoid watermarks, and do not change the character into a different person.

Goal: Create a wide horizontal triptych showing the same fantasy sky-kingdom scene rendered in three progressive art stages, like a concept art evolution from rough sketch to finished illustration.

Canvas: Very wide 3-panel image, approximately 2:1 aspect ratio, divided into exactly 3 equal vertical panels with no visible borders beyond the panel edges.

Layout: Each panel shows the same composition from a dramatic low, tilted perspective: a young fantasy knight stands in the lower foreground on a stone bridge or stairway, looking toward the viewer, holding or surrounded by a long flowing pale mint cloak that streams leftward in the wind. Behind the knight is a grand floating castle city suspended in a bright blue sky, with cables, stone columns, ruins, greenery, and clouds creating strong depth and motion lines.

Panel count and descriptions: Use exactly 3 panels. Panel 1, left: loose rough concept sketch with pale blue sky wash, visible construction lines, scribbly black pencil strokes, unfinished architecture, motion streaks, and minimal muted color. Panel 2, center: clean anime-style colored illustration with strong ink outlines, cel-shaded armor, saturated blue sky, fluffy clouds, green terraces, stone bridge details, and a floating castle in the distance. Panel 3, right: highly polished final fantasy illustration with sparkling sunlight, dense architectural detail, vivid blue sky, bright white clouds, lush green vines and trees, glowing highlights, painterly rendering, and cinematic clarity.

Subject details: The foreground character is {argument name="character type" default="a young blond fantasy knight"}, wearing ornate silver-and-gold armor with leather straps, boots, gloves, and a large flowing pale mint-white cape. The face may be softly simplified or obscured, but the body should face forward in a heroic pose with the cape sweeping dramatically across the scene. Keep the character consistently placed in all 3 panels: lower center foreground, slightly larger in the right panel, with the same costume and pose.

Environment details: The setting is {argument name="fantasy location" default="a floating castle city above the clouds"}. Include suspended stone platforms, carved pillars, bridge railings, hanging cables or ropes stretching diagonally across the sky, fragments of ancient ruins, patches of green grass, ivy, trees, and many white clouds. The background castle should be tall, ornate, vertical, and luminous, with towers and terraces stacked upward.

Visual style: Japanese fantasy anime concept art, dynamic perspective, wind-swept movement, adventurous mood, luminous sky palette. Main colors are {argument name="main color palette" default="sky blue, cloud white, pale mint, stone gray, gold, and fresh green"}. The right panel should feel magical and camera-facing with sparkling highlights and refined detail, while the left panel remains visibly unfinished.

Constraints: No captions, no logos, no watermarks, no text overlays. Maintain exactly 3 vertical stages of the same scene. Emphasize the progression from rough sketch to clean color to final polished fantasy key art.

Goal: Create a vertical Japanese landing page first-view hero image for an at-home English program for busy working mothers, with a warm parent-child learning photo on the right and large sales copy on the left.

Canvas: Vertical smartphone LP/FV design, 675×1200 px or 9:16 ratio, cream-white background, soft pastel palette, airy spacing, friendly family-oriented mood.

Visual style: Clean Japanese web advertising design, rounded shapes, hand-drawn decorative accents, pastel pink, brown, yellow, mint green, and sky blue. Use a bright lifestyle-photo look blended into a graphic LP layout. Keep typography large, rounded, and highly readable; emphasize key words in vivid pink.

Main image area: On the right half, show a Japanese mother and young child sitting at a wooden table doing English flashcard study at home. The mother wears a simple cream blouse and sits beside the child; the child wears a yellow striped top and holds an illustrated apple card labeled “apple”. Place a tablet on the table at the far right with cute stickers and text “Let’s enjoy English!”. Add four visible English flashcards on the table: 1) apple held by the child, 2) cat card, 3) dog card, 4) ball card. Obscure both faces with soft rectangular skin-tone privacy blurs while keeping hair, clothes, hands, and activity visible. Background includes a plant, shelf, soft daylight curtains, and pastel triangular bunting flags across the upper right.

Header: Top left logo with a small house icon and the brand text “おうち英語きっず”, with “おうち” in pink, “英語” in dark brown, and “きっず” in blue.

Text content: Add a dashed pink speech bubble near the top left containing “忙しいママでも続けやすい” with a small heart. Large main headline on the left: “1日10分、 おうちで英語が 好きになる。” Make “10”, “英語”, and “好き” bright pink, the rest dark brown, and underline the headline with a yellow brush stroke. Supporting copy below: “30代ワーキングマザーのための 子どもおうち英語講座”. Add small body text: “英語が苦手でも大丈夫。 動画・ワーク・声かけ例つきで、 親子で楽しく続けられます。”

Benefit cards: Include exactly three rounded feature cards in one row near the lower middle: 1) pink clock icon with “1回10分”, 2) green heart icon with “ママの英語力不要”, 3) blue house icon with “自宅でOK”.

CTA: Near the bottom, add one large rounded hot-pink button with white text “無料体験レッスンを見る” and a white circular arrow icon on the right. Under it, add a small smartphone icon and text “スマホでOK・今すぐチェック” with small pink attention marks.

Decorations: Add small hearts, spark lines, stars, dots, scalloped cloud-like cream shapes along the bottom edge, and a cute smiling cloud icon at bottom right. Keep everything soft, cheerful, trustworthy, and conversion-focused.

Customizable details: Use the brand name {argument name="brand name" default="おうち英語きっず"}, the main headline {argument name="main headline" default="1日10分、おうちで英語が好きになる。"}, the target audience line {argument name="target audience" default="30代ワーキングマザーのための 子どもおうち英語講座"}, the CTA button text {argument name="CTA text" default="無料体験レッスンを見る"}, and the tablet sticker phrase {argument name="tablet phrase" default="Let’s enjoy English!"}.

Constraints: Use Japanese text exactly as specified where visible, maintain a professional LP hero layout, include exactly three benefit cards and exactly four flashcards, no extra people, no watermark, no browser chrome.

Create a moody, grainy analog-film style portrait of exactly one young person sitting on outdoor stone stairs in front of a dark building entrance. The subject has long {argument name="hair color" default="black"} hair falling over the shoulders, wears an oversized dark navy zip jacket and a pastel tie-dye T-shirt with pale pink and blue tones, and sits casually with arms folded near the knees. Place exactly one large opaque square censor block over the entire face, centered from forehead to chin, in a warm brown gradient that hides all facial features. Show exactly three broad gray stone stair treads behind and beneath the subject, with a dark doorway and weathered wall details in the background. Use a square 1:1 crop, close medium framing from upper thighs to above the head, soft dim natural light, muted colors, shallow depth of field, subtle underexposure, visible film grain, and a candid street-photography feeling. The setting should feel like {argument name="setting mood" default="quiet urban evening"}; keep the pose relaxed and introspective, with no text, no logos, and no extra people.

Goal: Create a lavish fantasy anime rock-band concert illustration featuring {argument name="band name" default="a magical girl rock band"} performing on a glowing gothic crystal stage.

Canvas: Wide 16:9 cinematic composition, ultra-detailed, high-resolution anime key visual, dramatic concert lighting, deep violet and electric blue color palette with pink highlights.

Layout: Place the lead vocalist in the center foreground on a circular glowing stage sigil, surrounded by exactly 6 other band members and 1 small dog mascot. Arrange the performers in a full-stage ensemble: 2 musicians on the far left, 1 left-center bassist, 1 center vocalist, 1 drummer in the rear center, 1 right-center violinist, and 1 far-right keyboardist. Use strong depth, sparkling atmosphere, and bright purple audience glow sticks along the bottom edge.

Subject details: Show exactly 7 anime-style female band members: 1 far-left keyboardist seated/standing at a futuristic keyboard console; 1 left guitarist in black gothic idol fashion with cat-ear accessories holding an electric guitar; 1 left-center bassist in dark ornate clothing holding a bass guitar; 1 center lead singer with {argument name="lead singer hair color" default="blonde hair"}, wearing a pink plaid cropped idol top, short layered skirt, belts, ribbons, thigh straps, and tall lace-up platform boots, singing into a pink microphone on a stand with one hand extended toward the viewer; 1 rear-center drummer seated at a full drum kit with a glowing emblem on the bass drum; 1 right-center violinist in a red-and-black ornate stage outfit playing violin; 1 far-right keyboardist in a blue-and-white fantasy dress playing a glowing synthesizer. Add exactly 1 small Cavalier King Charles Spaniel mascot sitting at the lead singer’s feet, wearing a cute stage harness or decorative vest.

Stage and environment: The venue is a grand magical cathedral-concert hall made of crystalline arches, gothic spires, translucent purple prisms, floating shards, moonlit windows, and holographic musical displays. Behind the vocalist, create a huge circular stained-glass-like magic portal or ornate clockwork mandala glowing in lavender. Include two visible moons in the night sky, starry cosmic background, floating crystal decorations, fairy-like sparkles, and radiant stage beams.

Visual style: Extremely detailed Japanese anime illustration, idol-rock fantasy aesthetic, glossy lighting, intricate costumes, jewel-like textures, luminous neon outlines, cinematic backlight, volumetric haze, high contrast, sharp focus, elaborate accessories, magical particles, energetic live-concert mood.

Color and mood: Dominant {argument name="main color palette" default="purple, indigo, magenta, and icy blue"}; glamorous, powerful, celebratory, and dreamlike.

Constraints: Use exactly 7 human performers and exactly 1 dog mascot. No visible lyrics or readable poster text. Avoid extra band members. Keep the composition symmetrical around the lead vocalist, with the drummer clearly behind her and the dog clearly in the front foreground.

Create a close-up anime-style portrait of {argument name="character description" default="a bespectacled young woman with a deadpan, half-lidded stare"} facing forward against a plain white background. She has long straight {argument name="hair color" default="black-brown"} hair with blunt bangs, glossy sketchy highlights, and hair falling past both shoulders; both ears are slightly visible through the hair. Only the upper torso is shown, wearing a crisp {argument name="shirt color" default="white"} collared shirt with a dark tie or inner placket and a dark brown choker around the neck. Thin eyeglass frames are partially visible at the left and right edges of the face. Render in polished Japanese manga/anime illustration style with clean ink outlines, dense hatching in the hair, soft cel shading, subtle skin tones, and a vertical portrait crop. Place one large opaque square censor block centered over the entire face, from the forehead down to just above the chin, colored with a muted brown-to-taupe vertical gradient, hiding the eyes, nose, mouth, and most of the glasses while leaving the hair, ears, neck, collar, and shoulders visible. No text, no watermark, no extra objects.

Create a dreamy ultra-detailed 16:9 wide-angle photorealistic desk setup in a pastel pink kawaii gaming room. The scene centers on an ultrawide curved monitor showing a soft anime girl wallpaper: {argument name="character name" default="a pink-haired bunny maid anime girl"} with big glossy eyes, frilly white lace headband, red bows, floppy bunny ears, strawberry accents, tiny hearts, ribbons, ornate corner decorations, and a blush-pink background. Use {argument name="main color palette" default="white, blush pink, strawberry red, and soft lavender RGB light"} throughout the room. On the desk, include exactly 10 visible desktop items: 1 white game controller on the left, 1 small digital clock display reading {argument name="clock text" default="14:27 / MAY 02 / FRIDAY"}, 1 tiny retro cube screen with a character icon, 1 decorative pink crystal lamp, 1 wide soundbar under the monitor, 1 small smart control tablet with app icons in front of the monitor, 1 coiled pink cable, 1 mechanical white-and-pink keyboard with pink keycaps and strawberry charms, 1 white mouse, and 1 large strawberry-themed desk mat with anime line art. On the right side, place exactly 1 white glass-panel gaming PC tower with visible internal components, pink RGB lighting, 3 circular RGB front fans, 1 rear RGB fan, liquid-cooling tubes, a graphics card labeled {argument name="graphics card label" default="GEFORCE RTX"}, and a small anime figurine sitting on the lower front edge inside/against the case. Behind the monitor, create a soft pink wall with exactly 9 glowing hexagonal light panels arranged asymmetrically. Add white shelves and decor: exactly 2 floating shelves on the right with small plants and anime figures, 1 glowing flower-shaped neon light, and leafy green vines on the left edge. Lighting should be cozy and cinematic with bloom, soft reflections on glossy white surfaces, gentle depth of field, high-end PC battlestation aesthetics, no people in the room, no clutter beyond the listed items, and no watermark. Make the room feel like an immaculate strawberry-themed anime streaming setup, with the monitor and PC as the main focal points.