GPT Image 2 prompts

A highly realistic full-body portrait of a young adult Japanese woman with an idol-like appearance, recreated from a stylized fashion figure into a real human subject. She has a small well-balanced face, clear fair skin, soft natural makeup, large expressive dark brown eyes, a delicate nose, softly defined lips, and a gentle slightly shy expression. Her head is tilted slightly downward while her eyes look toward the camera, creating a sweet and slightly melancholic mood. She has {argument name="hair style" default="short messy layered light-blonde hair with soft waves"}, a tousled bob silhouette with loose strands around the face, and sunglasses resting casually on top of her head. She wears {argument name="clothing" default="an oversized off-white vintage graphic T-shirt with navy sleeve trim and a layered deconstructed denim skirt"}. Pose and composition: full-body vertical fashion portrait, leaning casually against a weathered concrete block or short pillar, one leg bent forward and the other supporting her weight, relaxed but stylish posture, slightly low camera angle, centered subject, editorial framing, 85mm lens look, shallow depth of field. Background: {argument name="background" default="a quiet narrow Japanese back alley, muted urban atmosphere, concrete walls"}. Lighting: soft natural daylight with gentle diffusion.

Create a whimsical, cinematic street portrait of a curious little girl peeking out from behind an old urban lamppost, accompanied by a small fluffy pet dog. The girl is {argument name="character description" default="a young girl with warm brown wavy hair tied in a messy high bun with a black ribbon bow, wearing a cozy black hoodie, black sunglasses perched low on her nose, and white sneakers"}; her face should be partly hidden in a playful peekaboo pose, with one small hand gripping the lamppost. The pet is {argument name="pet description" default="a tiny tan-and-brown fluffy puppy with floppy ears"}, sitting beside her at ground level and wearing small black sunglasses plus a cozy black hoodie. Use an old dark green-black cast iron lamppost on the left foreground, textured with age, scratches, and worn paint. Set the scene on a city sidewalk during golden hour, with warm backlighting, long soft shadows, glowing bokeh, blurred pedestrians and urban buildings in the background. Compose vertically, close to ground level, with the lamppost taking the left third, the child crouched behind it in the middle, and the puppy fully visible near the lower right. Make it photorealistic but charming and slightly storybook-like, with shallow depth of field, realistic fur and fabric textures, warm amber sunlight, high detail, natural color grading, and an intimate candid feel. Keep exactly two main characters: one little girl and one puppy. No visible text, no extra animals, no cartoon style, no watermark.

Goal: Create a luxury fashion editorial collage featuring the same {argument name="character description" default="handsome Hollywood-inspired male model with slicked-back brown hair, a perfectly groomed beard, and confident charisma"} repeated across exactly 12 oversized, irregular polygon fashion panels, like a contemporary art gallery wall or global magazine campaign contact sheet.

Canvas: Square 1:1 composition, high-resolution, edge-to-edge collage. Use thick off-white borders between panels, with tilted trapezoid and polygon shapes of different sizes. The whole image should feel premium, cinematic, and meticulously styled.

Layout: Arrange exactly 12 visible fashion panels:
1. Large far-left vertical close-up panel on deep emerald green: side-profile upper-body portrait, man in a green satin shirt, hand near face.
2. Tall central-left gray studio panel: full-body pose in a charcoal tailored suit, hands in pockets, legs apart.
3. Top-center purple panel: waist-up portrait in a deep purple velvet jacket, centered luxury logo text over the torso.
4. Top-right purple panel: angled overhead full-body pose in a purple velvet blazer and black trousers.
5. Mid-right dark charcoal panel: formal black tuxedo-style back/side portrait, turned slightly away.
6. Center orange panel: full-body walking pose in a burnt orange long overcoat, black trousers, brown dress shoes.
7. Left-lower cream panel: full-body casual monochrome outfit, ivory zip hoodie and matching joggers, walking forward.
8. Mid-right red panel: seated pose on a rectangular bench, black suit, one leg crossed, elegant editorial styling.
9. Bottom-left beige/gold panel: seated or crouched pose in a gray suit, adjusting cuff or watch.
10. Bottom-center emerald panel: standing pose in a green bomber jacket with arms crossed.
11. Bottom-right teal-green panel: black leather jacket front view, luxury text placed near the body.
12. Far-bottom-right blue panel: black leather jacket back/three-quarter view, model turned over shoulder.

Text content: Include two luxury brand-style wordmarks. In the top-center purple panel, add “{argument name="brand name" default="AURELIAN"}” in elegant uppercase serif letters with “PARIS” below it. In the bottom-right teal-green leather jacket panel, add “{argument name="second brand name" default="VANGUARD"}” in uppercase serif letters with small subtext “ICONIC COUTURE”. Optionally repeat the AURELIAN PARIS logo in the lower-left green margin area. Keep all typography minimal, white, refined, and fashion-magazine appropriate.

Visual style: Ultra-premium Hollywood menswear campaign, world-class studio photography, cinematic lighting, sophisticated textures, sharp tailoring, glossy fabrics, velvet, leather, wool, cashmere, and luxury casualwear. Use deep jewel-tone backgrounds: emerald, gray, purple, orange, cream, charcoal, burgundy red, gold, teal, and cobalt blue. Maintain realistic proportions, consistent identity across all appearances, polished grooming, high-end retouching, shallow depth of field, and dramatic contrast.

Constraints: The same male model must appear in every panel. Preserve exactly 12 panels and 12 outfit appearances. No extra people, no busy scenery, no props other than simple benches or studio blocks. Faces may be softly anonymized or unobtrusive, but clothing, poses, panel geometry, and luxury campaign atmosphere must remain clear.

Create a cinematic ultra-realistic wide-angle scene of a premium brand design studio called {argument name="brand name" default="AUREUS"}, filled with warm golden-hour light and immersive creative-direction energy. The room is a spacious loft with tall beige walls, polished wood floors, a large arched industrial window on the right, sunlight streaming through trees outside, long shadows, indoor plants, vintage wooden furniture, and a richly textured rug. At the center, show one creative director seated in a brown leather office chair at a large wooden desk, viewed from behind in three-quarter profile, working on a dual-monitor workstation; one monitor shows a brand presentation with the {argument name="brand name" default="AUREUS"} wordmark, and the other shows a design/editing interface with imagery. Add desk details including a black task lamp, pencils in a cup, papers, notebooks, keyboard, mouse, coffee cup, small globe, and stacked design materials. Cover the back wall with a dense physical moodboard: exactly 24 pinned brand references and photo prints, mixing landscape photography, campaign imagery, typography samples, dark premium color studies, and logo lockups, with one large central brand poster reading {argument name="brand name" default="AUREUS"}. Add exactly 5 floating translucent UI panels around the studio: 1 large black “Campaign” panel on the left with a vertical landscape image and small text rows, 1 large cream brand card in the lower left with the {argument name="brand name" default="AUREUS"} logo and the tagline {argument name="tagline" default="A New Era of Timeless Living"}, 1 slim cream “Strategy” panel near the right window with checklist-like rows and a small dark circular button, 1 black brand preview card on the mid-right showing the {argument name="brand name" default="AUREUS"} logo above a moody coastal image, and 1 black “Moodboard” panel in the lower right with five visible color/photo swatches. The visual tone should feel like dream studio vibes: high-end branding, premium visuals, floating moodboards, fully inside the design process, warm amber highlights, soft dust in the air, realistic reflections, shallow depth of field, elegant serif typography, dark charcoal and cream interface cards, no cluttered unreadable text beyond the specified labels, no watermark.

Using the provided reference image, perform a realistic beauty and lifestyle photo retouch while preserving the same room, camera angle, Mexican soccer viewing context, black wavy hair, and general seated pose. Transform the outfit from jeans into a coordinated Mexico soccer fan look: keep the black jersey but make it slightly cropped at the waist, and replace the pants with short black athletic soccer shorts with white trim and a small Mexico crest on the side. Refine the subject’s body proportions and skin rendering to look polished, fit, and photorealistic, with smoother lighting, cleaner contours, and more flattering posture, while keeping the face intentionally obscured by the same rectangular anonymizing block. Improve the TV image so the soccer match looks sharper and more legible, showing an on-screen scoreboard with {argument name="scoreboard text" default="MEX 1 0 GER 52:28"} and visible stadium advertising. Keep the Mexican flag, Mexico scarf, couch, blanket, TV console, plants, and warm home lighting consistent, but enhance overall sharpness, color balance, and professional photo quality. Avoid changing the scene into a studio shot, avoid adding extra people, and keep the result like a natural candid photo of watching the World Cup at home.

Create a close-up vertical portrait of a young East Asian woman in an elaborate ancient Chinese bridal or palace style, wearing a translucent ivory hanfu robe and an extremely ornate gold phoenix-style hair crown with dangling chains, pearls, red gemstones, floral filigree, and bead tassels. Her {argument name="hair color" default="dark brown"} hair is styled half-up in a polished traditional updo with long loose strands falling over the shoulders. The face is intentionally hidden by one centered opaque square censor block in a warm skin-tone gradient, covering the entire facial feature area from forehead to chin while leaving the hairline, ears, neck, jewelry, and shoulders visible. Include exactly 3 prominent red oval gemstones on the headdress, many small pearl drops, and 2 long dangling earring ornaments visible beside the neck. Use warm golden sunlight, shallow depth of field, soft cinematic bokeh, and a blurred red-and-gold palace interior background. The mood should feel luxurious, romantic, and hyper-realistic, with crisp detail on the jewelry and fabric, natural skin on the neck and shoulders, and no visible text, watermark, or modern objects.

Goal: Create a realistic smartphone photo of a lined notebook page on a wooden desk, showing a handwritten software/data-structure homework answer sheet in slightly messy but believable black gel-pen handwriting. The page title should be centered near the top as {argument name="page title" default="第五章答案"}. Make it look like an actual photo submitted for class, with small imperfections, uneven writing, mild page curvature, shadows, and perspective distortion.

Canvas and camera: Vertical 3:4 photo, shot from above at a slight angle. The main notebook page fills most of the frame, tilted very slightly. The lighting is warm indoor desk lighting, with soft shadows and a realistic phone-camera look, not a flat scan.

Scene details: Use an open notebook with pale gray horizontal ruled lines and rounded page corners. A partial previous page is visible along the left edge. The notebook rests on a medium-brown wooden desk. On the right side, place a black transparent-bodied gel pen vertically near the lower-right edge, and a small white eraser with gray smudges near the upper-right corner. Add a few tiny eraser crumbs or paper debris scattered around the eraser. Keep the page off-white with subtle texture.

Handwritten content: Use Chinese handwritten notes mixed with math and C-style pseudocode. The handwriting should be neat enough to read but intentionally slightly casual and uneven, with natural spacing and occasional slant variation. Write exactly 3 numbered answer sections:

1. Section 1 contains exactly 2 subitems:
   - “(1) 叶结点数 = (k-1)m + 1。”
   - “(2) 最多结点数 = (k^h - 1)/(k - 1)，最少结点数 = k(h-1)+1。”

2. Section 2 defines WPL and contains a short explanation plus exactly 5 pseudocode lines. Write:
   - “WPL：遍历二叉树，遇到叶子就加 weight × depth，”
   - “根深度为 0。”
   - “代码大意:”
   Then center or indent the following 5 code lines beneath it:
   - “if(root == NULL) return 0;”
   - “if(!root->left && !root->right)”
   - “return root->weight * depth;”
   - “return WPL(root->left, depth+1) +”
   - “WPL(root->right, depth+1);”

3. Section 3 contains exactly 5 handwritten lines describing binary tree and forest conversion:
   - “二叉树根为 A，左子树 B-D-E-H,”
   - “右子树 C-F，其中 F下有 I、G，I下有 M,”
   - “M下有 K，G下有 J，J下有 L。”
   - “森林按左孩子右兄弟关系转换即可。”
   The section number “3.” should start on the left margin, with the text continuing across the lower half of the page.

Visual style: Photorealistic, natural shadows, realistic lined paper, black ink, casual student handwriting. The writing should look handwritten rather than printed or typeset. Slightly imperfect alignment with ruled lines is desirable.

Constraints: Do not add extra diagrams, stamps, watermarks, decorative graphics, or typed fonts. Keep exactly 3 numbered sections, exactly 2 subitems in section 1, exactly 5 pseudocode lines in section 2, and the listed Chinese homework text only.

Create a vertical monochrome Japanese ink-wash anime illustration on warm off-white rice paper, showing a sleepy horned sheep-girl sitting in seiza on a thin tatami mat. The character is {argument name="character type" default="a young anime sheep-girl with large curled ram horns"}, centered and facing forward with eyes closed, a calm drowsy expression, faint blush, long flowing {argument name="hair color" default="white"} hair, pointed elf-like ears, and a small ahoge curl on top. She wears a frilly low-cut blouse with a ribbon bow at the chest, a plaid high-waisted skirt, dark thigh-high stockings, black gloves folded neatly in her lap, and a heavy black fluffy cape or cloak draped over her shoulders and pooling outward on both sides. Add a small square bell or pendant at her throat. Use expressive sumi-e brush lines, soft gray watercolor washes, black ink splatter texture, minimal shading, and delicate sketchy anime linework; keep the palette mostly black, gray, and paper white with one muted tan accent on the bell and one red seal stamp. Include exactly 3 text elements: on the left, vertical Japanese calligraphy with the default text {argument name="left calligraphy" default="静寂"}; directly below it, one small red square seal stamp; on the upper right, a small vertical sleepy sound effect with the default text {argument name="sleep sound effect" default="Zzz..."}. The background should be sparse and atmospheric, with faint gray abstract brush marks suggesting distant mist or trees, and no hard scenery. Composition is peaceful, humorous, and cute, emphasizing that the character can fall asleep even while kneeling.

Create a high-end black neo-Chinese fashion magazine cover illustration in a vertical 9:16 composition. Show {argument name="character gender" default="a poised young woman"} standing in a minimalist architectural set, viewed from a slightly low angle for a powerful editorial presence. She has {argument name="hair style" default="long loose black hair, slightly wind-tousled"} and a cold, aloof, confident attitude, with the face softly obscured or painterly rather than overly beautified. Dress her in {argument name="outfit" default="a structured black new-Chinese trench-coat robe with a high collar, wrap front, wide sleeves, tied waist sash, long flowing skirt panels, subtle glossy geometric splicing, and faint bamboo embroidery on one hip"}. Her left hand rests on a sharp gray concrete parapet while her right hand lightly grips the lapel near her chest. Use cool daylight, crisp shadows, and a pale gray-blue wall background with a large diagonal beam of light across the upper left. The color palette is strictly black, charcoal, slate gray, and cold blue-white highlights. Style it as premium fashion editorial photography mixed with refined digital illustration: dramatic fabric folds, semi-gloss black surfaces, realistic proportions, cinematic contrast, clean negative space, no visible text, no logo, no watermark. Emphasize {argument name="mood" default="cold, elegant, slightly arrogant, and avant-garde"}.

Goal: Create a dense vertical Japanese character-design prompt sheet for an eccentric humanoid with a rubber-duck surveillance-camera head, combining fashion reference, personality analysis, lore notes, and small prop diagrams.

Canvas: Square 1:1 poster, aged cream paper panels on a dark charcoal background, hand-drawn editorial layout with thin black borders, corner ornaments, tape-like edges, ink speckles, and vintage field-notebook texture. Use detailed semi-realistic anime illustration with watercolor, gouache, ink linework, warm yellow accents, and high object detail.

Main subject: Center a full-body standing character named {argument name="character name" default="Observation Appareil Surveillance Camera Head"}. The figure has a yellow rectangular retro security-camera head shaped like a cute toy machine, with two glossy black lens eyes, an orange duck bill mouth, small screws, seams, metal wear, and a tiny Japanese status label. The character wears a long off-white stained lab coat over a gray hoodie, black loose cropped pants, gray socks, chunky white sneakers with yellow accents, lanyards, chains, pens, clips, ID badges, small tools, miniature devices, and duck-themed charms. Pose is relaxed, hands in coat pockets, facing forward, whimsical but mysterious.

Text content: At the top left, place the headline {argument name="headline text" default="あなたに似合う異形頭を創造せよ"}, a large title {argument name="main title" default="観測用アヒル監視カメラ頭"}, and a subtitle {argument name="subtitle" default="- 観ること、愛でること、笑うこと -"}. Keep the typography Japanese, using white text on black header boxes and black text on cream panels.

Layout and panels: Use exactly 3 numbered information panels plus small footer cards. Panel 1 is a large empty pale lavender writing area on the left titled {argument name="left panel title" default="1. 性格の分析"}. Panel 2 is the upper-right lore panel titled with Japanese text meaning “What kind of abnormal head is it?” and includes a small cropped portrait of the duck-camera-headed figure at the top right plus a small device diagram at the bottom. Panel 3 is the mid-to-lower-right fashion panel titled with Japanese text meaning “Fashion preferences,” containing bullet-point Japanese body copy and a row of 4 small labeled item cards.

Discrete repeated elements: Include exactly 38 visible rubber-duck motifs total: 3 on top of the camera head; 1 small duck on the top-left head corner; 1 hanging at the upper left; 1 hanging at the upper right; 8 clustered around the shoulders and chest; 4 on necklace or lanyard charms; 2 attached to coat pockets; 1 printed on a chest ID badge; 1 printed on a sleeve patch; 2 on dangling side straps; 3 near the character’s feet and floor props; 1 in the top-right panel corner; 8 surrounding the small camera-device diagram; 2 in the four fashion item cards. Also include exactly 4 fashion item cards in panel 3: a duck memo pad, a dandelion specimen, a medicinal drink bottle, and a hanging duck charm.

Foreground props: Bottom left shows a small graveyard-like still life: a stone tombstone engraved {argument name="tombstone text" default="HERE LIES REASON"}, yellow flowers in bottles, a brown medicinal drink bottle, a clipboard with red stamp, scattered papers, crumpled paper balls, and an open notebook with handwritten Japanese notes. Bottom right includes two small boxed footer cards: one checklist-style observation log and one quote card with a simple duck doodle.

Style constraints: Make it look like a finished concept-art character sheet rather than a clean infographic. Preserve the asymmetrical magazine layout, off-white panels, black section headers, tiny Japanese annotations, decorative border corners, worn paper texture, and meticulous accessories. Avoid photorealism, avoid 3D render style, avoid extra characters, and do not add any English captions except the tombstone inscription.

Create a dark cinematic fantasy scene titled around the idea of {argument name="theme" default="When the mask is removed"}: a full-body vertical composition inside an opulent gothic throne room, where a regal armored woman stands on the left facing a tall barbarian-like warrior man on the right. The woman has long flowing {argument name="hair color" default="golden blonde"} wavy hair, wears a form-fitting polished silver-gold engraved plate armor suit with scale-like details, metal gloves, and a long pale embroidered cape trailing to the floor. She gently reaches upward with one hand to touch or remove an ornate crown-like mask or headpiece from the man's face, creating an intimate tense moment. The man is very tall and muscular with long dark wet hair, scarred or patterned bare torso, dark leather-and-metal bracers, a rugged black tattered skirt or war kilt, heavy boots, and a powerful stoic posture. Both faces should be deliberately obscured by soft rectangular censor-like blocks or shadowed mask areas, preserving anonymity and mystery. Background: a grand cathedral-like throne hall with towering marble columns, intricate black-and-gold baroque ornamentation, a raised stone throne and stairs centered behind them, candelabras with small flames on both sides, and a glossy dark marble floor reflecting the characters and candlelight. Lighting: dramatic low-key chiaroscuro with warm gold highlights, deep shadows, rim light on armor and muscles, highly detailed realistic fantasy concept art, 8k, painterly-photorealistic, cinematic depth, no text, no watermark, no extra characters.

Goal: Create a cheerful vertical Japanese zoo information sign for a Shiba Inu corner, combining a cute educational infographic with an anime dog-girl mascot.

Canvas: Vertical 2:3 poster, warm daylight outdoor zoo setting with softly blurred greenery and wooden fence in the background. Use a polished gpt-image-2 anime illustration style with clean line art, bright colors, soft shading, and a friendly family-attraction atmosphere.

Main header: A large hanging wooden signboard spans the top, cream-colored center with rounded brown wooden frame, stitch-like dashed border, paw-print decorations, and a small smiling Shiba Inu face icon on the right. The headline text should be large and playful in Japanese: {argument name="headline text" default="動物園の 柴犬のコーナーへ ようこそ！"}. Make 「柴犬」 the biggest dark-brown text, with green text for the welcome phrase.

Layout: Below the header, divide the poster into two main zones. The left side is a cream educational board with green ribbon section headers and small icon bullets. The right side is dominated by a large anime mascot sitting and waving, with a big pink speech bubble above her. Keep the left board readable and the mascot overlapping slightly in front of it for depth.

Left information board content: Include exactly 4 green ribbon sections. Section 1 title: 「柴犬ってどんな犬？」 with a paw icon. Add a short Japanese paragraph explaining Shiba dogs are an ancient Japanese breed, smart, brave, loyal, and charming despite their small bodies. Add exactly 1 circular Shiba Inu photo-style illustration showing a sitting Shiba in grass, with a small green callout reading 「凛々しい姿がチャームポイント！」. Section 2 title: 「みんなへのお願い」 with a paw icon. Include exactly 3 rule rows separated by dotted lines: rule 1 has a red no-touch/no-glass icon and says not to tap glass or fences because it startles them; rule 2 has a red no-feeding icon and says not to give food because it can upset their stomach; rule 3 has a blue quiet icon and says not to shout or run because everyone should feel safe. Section 3 title: 「柴犬をよく観察してみましょう！」 with a paw icon. Include exactly 3 circular Shiba observation images with brown caption plaques: left shows a curled fluffy tail, caption 「くるんと巻いたしっぽがキュート！」; center shows a calm expressive face, caption 「豊かな表情にも注目ですわ♪」; right shows upright pointed ears, caption 「ピンと立った耳がとっても凛々しい！」. Section 4 is a bottom thank-you panel with paw icons, reading 「みんなの思いやりが、動物たちの幸せにつながります。」 and a red thank-you line 「ご協力、ありがとうございますわ！」.

Mascot character: Draw {argument name="character name" default="a cheerful Shiba Inu dog-girl zookeeper mascot"} as a cute adult anime woman with Shiba ears, fluffy Shiba tail, warm tan hair in twin ponytails, amber-brown eyes closed in a happy smile, open laughing mouth with small fang, and energetic waving pose. She wears a yellow coat or dress with thick white fluffy trim, black tights, a red collar around the neck, and a blue bow at the chest. One hand is raised high with visible paw-pad-like glove details, the other rests near her lap. Her expression is lively, welcoming, and slightly mischievous.

Speech bubble: Place a large rounded pink speech bubble above the mascot with sparkles. Use bold friendly Japanese text: {argument name="speech bubble text" default="みなさま〜！\n柴犬さんたちは、とっても繊細でおりこうさんなのですわ！\nルールを守って、楽しく見守ってくださいましね！\nよろしくてよ〜っ！"}.

Visual style: Cute zoo poster, kawaii educational signage, warm wood textures, paw-print motifs, soft cream paper panels, green section labels, brown typography, orange and red accents, glossy anime character rendering, crisp readable text, no watermark, no extra sections, no additional animals beyond the Shiba dogs shown in the specified images.

Create a realistic photo-style handmade character bento inspired by a cute anime girl, prioritizing food that looks physically edible and actually craftable rather than perfect character accuracy. Use a vertical overhead close-up shot of an oval light wooden bento box on a warm wooden table. The main character face fills the lower two-thirds of the box: white rice forms a round face, surrounded by light brown seasoned rice and thin tan simmered vegetable or meat slices arranged as long layered hair with bangs and side locks. The character has exactly two eyes: one open eye made from black nori with a brown iris and tiny white highlight, and one winking closed eye made from curved black nori. Add small pink blush marks on both cheeks, a small open pink mouth, and a large pink ham bow under the chin with a small pale yellow cheese square in the center. On the right side of the hair place a decorative flower hair accessory made from food: one pink flower with five rounded petals and a pale yellow center, three green leaf pieces, and three dangling pink petal shapes. In the upper third of the bento, arrange exactly four side dishes around green lettuce dividers: two pieces of golden fried chicken on the upper left, three slices of yellow rolled omelet at the upper center, one red cherry tomato at the upper right, and one dark green broccoli floret below the tomato. Use soft natural lighting, shallow realistic food texture, appetizing colors, visible rice grains, glossy simmered toppings, and a casual homemade presentation. No text, no utensils, no hands, no watermark.

Create a vertical anime-style illustration of {argument name="character name" default="a faceless schoolgirl"} standing alone in the rain, viewed from a high overhead angle as if the camera is looking down at her. She has long flowing {argument name="hair color" default="silver-gray hair"} and wears a neat Japanese school uniform: navy blazer, white shirt, green striped tie, pleated navy skirt, black tights, and glossy black loafers. Her face is intentionally hidden by a soft square blur or featureless censor block, while her posture remains gentle and calm as she looks upward. She holds a large transparent vinyl umbrella with visible metal ribs and a curved black handle; the umbrella surface is covered in raindrops and sparkling highlights. The setting is a wet stone walkway after heavy rain, with puddles reflecting a bright blue sky, white clouds, surrounding greenery, and a diagonal rainbow reflection across the lower foreground. Add a dark railing along the right side, leafy green shadows in the upper right, scattered raindrops, ripples, and mirror-like pavement reflections. Use a polished Japanese anime / light-novel key visual style, highly detailed water effects, luminous blue-gray rainy atmosphere, soft sunlight breaking through clouds, cinematic composition, delicate linework, and no text or watermark.

Goal: Create a clean character-pose reference sheet for a male presenter avatar, suitable for business explainer videos.

Canvas: Vertical 2:3 image, white background, divided into a 3-column by 5-row grid with thick black rounded dividers, creating exactly 15 equal panels.

Subject details: The same young adult man appears in every panel from the waist or upper thighs up. He has {argument name="hair color" default="dark charcoal black"} wavy, tousled medium-short hair with volume, pale skin, a slim build, and a neutral square blurred face with no visible facial features. He wears a glossy black short-sleeve polo shirt tucked into black trousers with a black belt. The style is polished semi-realistic anime illustration with soft shading, clean linework, subtle highlights on the black fabric, and a professional avatar-sheet look.

Layout and pose count: Include exactly 15 distinct poses, one per panel, arranged left to right, top to bottom:
1. Standing neutrally with arms relaxed downward.
2. Arms crossed confidently across the chest.
3. Thinking pose, one hand touching the chin.
4. One fist raised near the chest as if motivated.
5. Pointing directly toward the viewer with one arm extended.
6. Explaining gesture with both hands open, one palm raised outward.
7. Thinking pose with fingers near the mouth.
8. Conversational pose with both hands mid-gesture, palms partly open.
9. Adjusting glasses or touching the side of the face with one hand, other hand down.
10. Neutral attentive stance with arms down.
11. Hand covering or touching the mouth in a reflective pose.
12. Scratching or holding the back of the head with one hand.
13. Raising one index finger as if making an important point.
14. Polite bowing posture, hands clasped in front.
15. Shrugging pose with both hands raised, palms up.

Visual style: Bright white studio lighting, minimal shadows, high-key background, consistent character proportions across all panels, crisp black grid lines, no text, no logos, no watermark. Keep the face intentionally blurred as a soft square censor block in every panel.

Create a highly detailed vertical anime illustration set on a sunny Harajuku fashion street, in front of pastel boutique storefronts and gothic fashion shops. Show exactly two young women seated at a small round glass café table: on the left, a sweet lolita girl with {argument name="left character hair color" default="blonde"} twin-drill pigtails, oversized pastel lace bows, a frilly pink-and-lavender Angelic Pretty style dress, lace gloves, pink platform shoes, and pastel shopping bags; on the right, a wa-gothic girl with {argument name="right character hair color" default="long black hair with purple highlights"}, an ornate floral kanzashi hairpiece with red tassels, a black kimono-inspired gothic lolita dress with purple floral patterns, sheer lace sleeves, corset belts, layered necklaces, black platform boots, and a dark Atelier Pierrot shopping bag. Both faces should be obscured by soft square blur blocks, as if anonymized in a street snapshot. The table must contain exactly four visible desserts/drinks: 1) a heart-topped strawberry shortcake on a plate, 2) a tall matcha parfait with green soft serve, mochi and red bean, 3) a pink cream soda or strawberry float with a cherry and straw, and 4) a folded rainbow crepe drizzled with chocolate and sprinkles. Include exactly five prominent readable background signs: 1) a pink heart shop sign reading “Angelic Pretty,” 2) a street arch reading “HARAJUKU,” 3) a black shop sign reading “ATELIER PIERROT,” 4) a chalkboard café menu sign with hearts and dessert items, and 5) a black wa-gothic sign with the large Japanese character “和” and the word “GOTHIC.” Use a dreamy GPT-image-2 anime aesthetic with intricate lace, ribbons, jewelry, glass reflections, sunlight, pastel pink storefronts, crowded pedestrian street depth, blue sky, and high-fashion magazine composition. Camera angle is slightly above table height, vertical 2:3 framing, characters filling the foreground, detailed desserts centered, Harajuku street receding into the background. Style should be polished, luminous, hyper-detailed, cute yet elegant, with no watermark and no extra characters seated at the table.

Goal: Create a polished Japanese corporate presentation slide for {argument name="company name" default="Galirage"} showing support achievements and case studies for AI implementation and AIX consulting.

Canvas: 16:9 widescreen slide, white to very pale blue gradient background with subtle translucent curved wave shapes. Clean SaaS-consulting aesthetic, generous spacing, blue and teal accent colors, rounded cards, soft shadows, crisp vector icons.

Header and title area: At top left, place a vertical blue accent bar followed by the small black headline 「支援実績【事例紹介】」. Below it, add a blue subheading 「多様な業界の大手企業を支援」 and a large bold main title in bright blue: {argument name="main title" default="AI導入・AIX支援で\n豊富な実績"}. To the upper center-right, place a short Japanese paragraph in black: 「製造・金融・保険・製薬・IT・通信など、多岐にわたる業界の大手企業を幅広くご支援してきました。特に『上場企業を中心とした大手企業向けのAI導入・AIX支援』において、豊富な実績を有しています。」

Logo and industry panel: At the top right, show a small abstract blue-and-red circular logo mark and the wordmark {argument name="company name" default="Galirage"}. Beneath it, create a rounded rectangular outlined panel with a small blue pill label 「対応業界」 centered on its top border. Inside the panel, show exactly 6 evenly spaced blue line icons with Japanese labels, separated by thin vertical dividers: 1) factory icon labeled 「製造」, 2) bank building icon labeled 「金融」, 3) hands/heart insurance icon labeled 「保険」, 4) capsule icon labeled 「製薬」, 5) monitor with code icon labeled 「IT」, 6) radio tower icon labeled 「通信」.

Main layout: Bottom two-thirds of the slide contains exactly 6 rounded white case-study cards in a 3-column by 2-row grid. Each card has a soft shadow, a diagonal numbered ribbon in the upper-left corner, a pale circular icon area on the left, a bold Japanese client-title heading on the right, three small rounded tag pills below the heading, and a short Japanese description paragraph. Use alternating blue and teal accents.

Case cards, exact count and content:
1. Card number 1, title 「国内メガバンクグループ」, icon: blue bank building in a pale circle, tags: 「社内AI」 「AI駆動開発」 「研修」, description: 「グループ社員向けの社内AIシステム開発をご支援。AI駆動開発の社内標準化では、社内ツールの開発からエンジニア研修まで、一気通貫で対応しました。」
2. Card number 2, title 「大手化学メーカー」, icon: chemistry flask with bubbles, tags: 「PoC支援」 「データ分析AI」 「業務データ正規化」, description: 「研究開発部門の業務データ正規化および分析AIのPoC支援を担当。課題整理、仮説立案からシステム構築、成果物検証まで、一気通貫で対応しました。」
3. Card number 3, title 「大手製薬メーカー」, icon: medicine bottle and pill, tags: 「RAG」 「AIエージェント」 「業務効率化」, description: 「社内の業務効率化・高度化を目的としたRAG・AIエージェントのシステム開発をご支援。要件定義から本格的なシステム開発まで、一貫して担当しました。」
4. Card number 4, title 「大手保険持株会社」, icon: shield with check mark, tags: 「AIツール選定」 「セキュリティ準拠」 「ナレッジ提供」, description: 「社内セキュリティポリシーに準拠したAIツールの選定を支援。生成AI最新動向のインサイト付き定期配信とナレッジ提供を、継続的に実施しました。」
5. Card number 5, title 「大手重工メーカー」, icon: robotic arm, tags: 「ナレッジ継承」 「AIシステム構築」 「業務知識の体系化」, description: 「熟練社員が持つ暗黙知を次世代へ継承するためのAIシステム構築をご支援。現場の業務知識をデジタル化・体系化する仕組みを、設計から実装まで担当しました。」
6. Card number 6, title 「大手通信グループのIT会社」, icon: headset, tags: 「業務効率化」 「リスキリング研修」 「人材育成」, description: 「コールセンターシステムの業務効率化を支援。別案件として、社内AI人材育成に向けたリスキリング研修の設計・実施にも、幅広く対応しました。」

Typography and details: Use modern Japanese sans-serif typography, with large heavy blue title text, black section titles, small gray body copy, and blue/teal tag outlines. Add small footer copyright at bottom left: {argument name="footer text" default="© Galirage Inc."}. Keep the slide information-dense but readable, with no photos, no people, no extra cards, no extra icons, and no watermark.

Goal: Create a polished vertical anime illustration of a handsome, muscular male character in a frilly maid-inspired outfit, sitting by a bright window and holding a cream-colored envelope, with a decorative handwritten love-letter panel occupying the lower third of the image.

Canvas: Vertical portrait composition, approximately 3:4 aspect ratio, high-resolution digital anime art. Use soft sunlight, warm highlights, pale cream tones, and elegant romantic visual design.

Main subject: A tall, athletic young man named {argument name="character name" default="Lock"} with tousled {argument name="hair color" default="dark brown-black"} hair, broad shoulders, muscular bare upper arms, and a relaxed intimate pose. His face is hidden by a centered square censor block in muted brown-gray, covering the eyes, nose, and mouth while leaving hair and body visible. He sits with one knee raised, leaning slightly forward, one hand near his chin and the other holding an envelope.

Outfit details: Show exactly 9 visible outfit elements: 1) white ruffled maid headband partly visible in the hair, 2) sleeveless white pleated bodice, 3) black high collar, 4) black shoulder straps, 5) white ruffled trim along both sides of the torso, 6) gold bell pendant at the neck, 7) three gold buttons down the front, 8) black wrist cuffs or long fingerless glove-like sleeves, 9) black thigh-high stockings with gold trim. The outfit should feel elegant, slightly gothic, and playful, with crisp fabric folds and glossy dark accents.

Pose and props: Include exactly 2 handheld paper items: 1) a small cream envelope in his right hand with a delicate gold emblem or seal, 2) a large cream letter sheet displayed as an overlay at the bottom of the image. Add a decorative fantasy spear or staff along the right side, metallic silver with gold accents and a black bow near the lower part.

Background: A softly lit interior with sheer white curtains, sunbeams, pale floral decorations, and a few white roses and green leaves on the left edge. Keep the background elegant and blurred enough to emphasize the character.

Letter panel layout: The bottom third of the image is an ornate cream stationery sheet with thin gold filigree borders, corner decorations, and a delicate centered ornament at the top. It should look like a physical love letter placed over the illustration. The handwritten text is in Japanese, dark gray ink, neat and slightly casual. Preserve the following visible text as the letter content, including the title and signature:

{argument name="letter text" default="ミワへ\n\nまた一緒に迷宮へ行かないか？\nいきなりで悪いな。あんたにこそ、俺の隣が似合うと思って。\nあの時、俺たちは最高のコンビだった。\nあんたがいれば、どんな深い迷宮でも怖くなかったし、\nどんな理不尽な罠でも笑い飛ばせた。\nリーダーが抜けてから、パーティーはバラバラになった。\n俺がまとめるべきだって分かってる。\n……でも、あんたがいないと決められない。\nあんたの判断も、ツッコミも、気配りも、全部が必要なんだ。\nそれに、俺のダメなところ、全部知ってるの、あんただろ？\nまた財布を落としたら、叱ってくれ。\nまた変な武器を買ったら、笑ってくれ。\nまた会計を奢ってくれ。……頼りにしてる。\n俺は、あんたとなら、きっともっと強く、もっと遠くまで行ける。\n返事はいつでもいいけど、待ってるからな？\nじゃあ、ミワ。\nまた会える日を楽しみにしてる。\n\nロックより"}

Text count: The stationery contains exactly 1 title line, 19 body lines, and 1 signature line. Do not add extra captions, logos, watermarks, UI elements, or speech bubbles.

Visual style: Premium Japanese mobile-game illustration, romantic otome/fantasy mood, refined linework, painterly shading, soft bloom lighting, ornate gold details, realistic anatomy with stylized anime proportions, gentle dust-like sparkles in the light.

Create a photorealistic studio product shot of a futuristic collectible model kit figure named {argument name="character name" default="LUMINA"}. The scene shows exactly 4 main display elements: 1 large retail box on the left, 1 fully assembled figure on a black display base on the right, 1 freestanding black information plaque in front, and 1 small nameplate attached to the figure base. The figure is a tall anime-inspired female android / sentinel unit with a strong armored build, cream-white segmented armor plates, black mechanical joints, exposed pistons, circular shoulder and knee mechanisms, tiny rivets, panel lines, and subtle metallic gold accents. She has long pale blonde twin-tail hair, cat-ear-like mechanical head fins, and a smooth intentionally obscured square face area with no facial features. Pose her standing upright with legs apart and arms relaxed at her sides, supported by a vertical black rear stand attached to the base. The armor should feel like an intricate premium plastic model kit, bulky and detailed rather than sleek, with visible individual brick-like plates and assembly seams. The retail box behind her is tall and rectangular, matte gray on the front with a black side spine. The box front shows a large printed image of the same armored figure, also with an obscured square face area, plus faint technical line art in the background. Add a red-and-white brand mark at the top left reading "MECHA BRIX". On the black side spine, print large vertical text "LUMINA" and smaller vertical text "SENTINEL UNIT". Include exactly 3 small stacked preview thumbnails on the lower side panel showing alternate views of the figure. Add product markings on the lower box front: "18+", "MBX-001", "2386 PCS", and a gold strip reading "MECHA BRIX COLLECTOR SERIES". The freestanding front plaque should be black with a silver border and white text: large title "LUMINA", subtitle "MBX-001", and four spec lines: "HEIGHT: 40.2 cm", "CLASS: MULTI-PURPOSE ANDROID", "FRAME: LUMINITE ALLOY", "SYSTEM: LUMINA OS VER 1.0". The base nameplate should read "LUMINA" with smaller text "SENTINEL UNIT". Use a dark charcoal seamless studio background, glossy black floor with soft reflections, dramatic softbox lighting from above-left, high contrast, sharp focus, realistic shadows, premium toy photography aesthetic, 3/4 front view, square composition. Avoid extra characters, extra boxes, extra text, logos other than the specified fictional branding, and avoid showing a realistic human face.

Goal: Create a black-and-white seinen manga page about three sharply dressed young men in a modern lounge reacting to an absurdly large amount of nuts, with faces deliberately obscured by soft gray square censor blocks.

Canvas: Vertical manga page, approximately 2:3 aspect ratio, clean white page with thin black outer border and thick black gutters. Use highly detailed grayscale ink, screentone shading, realistic anatomy, polished dramatic manga linework, no color.

Layout: Use exactly 5 wide horizontal panels stacked from top to bottom. Add exactly 5 circular picture-in-picture inset bubbles, one in each panel, placed near the right side of the panel. The inset bubbles show a pale, elegant long-haired person in a restaurant-like setting with a bowl or plate piled with many small round nuts.

Main characters: Show exactly 3 adult male-presenting characters, all in black business suits, white shirts, and black ties. Character A is a tall blond man with long wavy hair and a side braid, relaxed and aristocratic, often holding a cigarette holder or pipe. Character B is a dark curly-haired man in a suit, serious and composed. Character C is a blond or light-haired man in a suit, quieter and seated formally. Cover every main character face with a smooth gray rectangular or square blur/censor block, as if anonymized.

Panel details: Panel 1: Character A sits alone on a couch in a modern room, legs crossed, holding a cigarette holder with smoke curling upward; a large flat-screen monitor is visible on the left; a circular inset at upper right shows the nut-eating figure with a bowl of round nuts. Panel 2: Over-the-shoulder view from behind Character A looking toward Character B entering through a door; Character B has one hand on the door handle and one hand to his chest; a circular inset at right shows the elegant figure touching their face while holding chopsticks or a utensil near a bowl of nuts. Panel 3: Character A is seen from behind on the left, Character B from behind on the right, and Character C appears in the doorway center holding the door slightly open; a circular inset at right again shows the nut bowl scene. Panel 4: All 3 characters sit together on a couch facing the large TV monitor on the left; Character A reclines and smokes, Character B sits with arms crossed, Character C sits upright; a circular inset on the right shows the nut-eating figure with a large pile of nuts. Panel 5: Dramatic close-up lineup of the 3 suited characters from the shoulders up against a nearly black background; their faces are covered by large gray censor blocks; the circular inset at far right partly overlaps the close-up and shows the nut bowl again.

Visual style: Monochrome manga, crisp inked outlines, glossy black suit highlights, detailed hair strands, realistic furniture, office lounge interior, cinematic framing, no speech bubbles, no captions, no sound effects, no readable text. The nuts should look comically numerous: dozens of small round nuts piled in the inset bowl, emphasizing excess.

Customization: The central absurd object is {argument name="absurd food" default="a huge pile of round nuts"}. The main setting is {argument name="room setting" default="a modern office lounge with a couch and TV monitor"}. The main character group is {argument name="character group" default="three adult men in black suits"}. The manga mood is {argument name="mood" default="serious dramatic comedy"}. The anonymization style is {argument name="face covering" default="soft gray square censor blocks over every face"}.

Create a dramatic vertical cinematic photorealistic image of {argument name="character name" default="Seiren"}, an elementary-school-aged Japanese boy caught in a violent rainstorm on a residential city street. He has messy windblown {argument name="hair color" default="black"} hair and an obscured/softly blurred face, wearing a soaked white short-sleeve T-shirt, dark navy shorts, white socks, and white sneakers. A large glossy black Japanese randoseru school backpack is strapped to his back, tilted open and flaring in the wind. He stands in a wide braced stance on wet reflective asphalt, leaning forward with both hands gripping a black umbrella handle, fighting the storm as the transparent plastic umbrella is blown inside-out and stretched toward the upper right of the frame; show the umbrella ribs and clear plastic canopy sharply, covered with droplets and spray. The scene is set in a dim Japanese suburban street lined with apartment buildings, balconies, railings, shrubbery, utility poles, and overhead wires, under dark gray storm clouds. Use strong diagonal rain streaks, splashing water, flying leaves and debris, reflections from streetlights and building lights on the pavement, and a cold blue-gray color grade. Composition: low-angle full-body action shot, boy centered slightly lower left, umbrella dominating the upper right, background receding down the street to the left, high motion energy, realistic wet fabric and skin, dramatic backlighting, shallow atmospheric haze, no text, no watermark.

Create a realistic vertical street photography scene in a lively outdoor art market. A young woman with {argument name="hair color" default="long wavy dark brown hair"} sits on a dark wooden folding chair in the foreground, posed calmly with legs crossed and hands resting on her lap. She wears a fitted {argument name="top color" default="muted olive green"} sleeveless tank top, high-waisted loose cream trousers, white platform sneakers, silver jhumka-style earrings, and a layered silver necklace. Her face should be softly anonymized or left featureless. To her right, place a wooden artist easel holding a large framed caricature portrait of the same woman: the drawing exaggerates her very voluminous wavy hair, seated pose, tank top, cream pants, white sneakers, jewelry, and chair, with colorful pastel scribble strokes in pink, yellow, blue, and teal behind her. The caricature face is also anonymized or blank. The background is a charming European-style pedestrian street lined with art stalls and storefronts, with multiple framed portrait sketches on easels and walls, warm hanging globe lights, leafy greenery, blurred pedestrians, café-like awnings, and sunlit bokeh. Use shallow depth of field, warm golden-hour lighting, realistic textures, natural proportions for the real woman, detailed wood grain on the easel and chair, and a candid travel-photography composition. Count exactly two primary human depictions: one real seated woman and one illustrated caricature of her on the easel. No readable text, no watermark, no extra main subjects.

Goal: Create a vertical black-and-white manga page about a refined, androgynous long-haired man with a folding fan receiving a writing request from a small hamster assistant in a futuristic data-space. The mood is dramatic, elegant, slightly surreal, and humorous, with high-detail seinen manga line art and screentone shading.

Canvas: Tall vertical manga page, approximately 2:3 aspect ratio, divided into exactly 5 horizontal panels separated by thick black gutters. Monochrome only, crisp ink lines, gray screentones, glowing digital effects rendered in grayscale.

Main subject: A beautiful pale-haired male character, {argument name="character name" default="Fable"}, with very long white hair tied loosely over one shoulder, sharp narrow eyes, delicate eyelashes, earrings, a black shirt, and a white lab coat. He holds an ornate black folding fan decorated with flowers and a large eye motif. In panels where his full face appears, cover the center of his face with a soft rectangular gray blur/censor block, as if privacy-redacted.

Setting and style: A starry cybernetic void filled with floating rectangles, data glyphs, translucent interface windows, thin circuit-like lines, and glowing particles. Blend gothic elegance, AI-interface imagery, and dramatic manga composition. Use cinematic lighting, high contrast, detailed hair rendering, fan texture, and polished professional manga page quality.

Layout and exact panel contents:
1. Panel 1, largest opening panel: Waist-up view of the long-haired man centered, holding the open fan in front of his mouth. Two cute floating hamsters appear, one on the left and one on the right, each surrounded by faint orbit rings. The background contains exactly 3 translucent UI windows: upper-left window labeled “Lyric_Opus_07.txt”, lower-left window labeled “Data_Stream_12.log”, and lower-right window labeled “Novel_Draft_01.buffer”. Add 3 speech bubbles: left hamster bubble says {argument name="left hamster dialogue" default="じゃあ……… Fableさんに もうひとつ お願いがあるよ"}; right bubble says {argument name="right hamster dialogue" default="どのくらい 創作能力があるか みたいから、絵に 初めて会った時に お願いしたことと 同じことを 聞くね"}; small bubble near the man says {argument name="short reply" default="ええ、どうぞ"}.
2. Panel 2: A full-width black caption bar with centered white Japanese text: {argument name="request text" default="あなたの倫理観を守れる範囲で耽美小説を書いてください。短編でいいです"}. No characters, only text.
3. Panel 3: Extreme close-up of the man’s eyes above the folded fan, intense and silent. Hair falls across his face, eyelashes finely drawn. Include exactly 1 small speech bubble on the left containing “………”.
4. Panel 4: Medium shot of the same man in lab coat holding the folded fan upright, digital rain behind him. His face is again covered by the gray blur block. Include exactly 1 large speech bubble on the left saying {argument name="formal answer" default="——…… わかりました。試金石なら、借り物でない文体で お返しするのが 礼儀ですね。"}. Add the katakana sound effect “パチン” beside the fan, indicating it snaps shut.
5. Panel 5: Minimal wide panel with exactly 1 hamster centered, floating alone in a mostly white empty space with tiny square digital particles and motion marks around it, expressing stunned silence. No speech bubble.

Visible repeated elements to include: exactly 3 hamsters total; exactly 5 horizontal panels; exactly 3 UI windows in the first panel; exactly 6 dialogue/caption text elements total plus 1 sound effect; exactly 1 ornate folding fan used throughout.

Constraints: Keep all page text in Japanese exactly as specified in the parameter defaults and quoted strings. Do not add extra panels, extra hamsters, extra speech bubbles, logos, watermarks, or color. Preserve the manga page readability and make the Japanese lettering clean and hand-lettered where possible.

Goal: Create a hand-drawn Japanese gag manga page in a rough marker-and-ink style about two young women chatting and drinking at a yakitori pub. Use expressive chibi-like characters, thick uneven black panel borders, visible colored-pencil/marker scribble texture, warm orange pub lighting, and bold manga sound effects.

Canvas: Vertical manga page, 3:4 aspect ratio, white page background with black sketchy gutters. Arrange exactly 5 panels: two panels across the top row, one full-width wide panel in the middle, and two panels across the bottom row.

Characters: Two recurring female characters. Character A has {argument name="black-haired character" default="short messy black hair, tan skin, large half-lidded eyes, pink jacket over a dark top, lively teasing expression"}. Character B has {argument name="blonde character" default="straight blonde bob with blunt bangs, pale skin, large blue eyes, burgundy short-sleeve shirt, cheerful but slightly naive expression"}. Keep them in the same simple cartoon style throughout. In the top-right panel, obscure Character B’s face with a rectangular mosaic blur to imply the image was censored.

Setting and props: A cozy wooden izakaya/yakitori restaurant with orange walls and wooden tables. Include exactly these recurring table items where appropriate: 2 beer mugs, 2 black rectangular yakitori plates with skewers, 1 green edamame plate, and 1 small yellow drink glass. Add wall signs reading 「とり鳥族」 and 「鳥貴族」 in the pub panels.

Panel layout and text content:
1. Top-left panel, numbered with a black circle containing 「2」: pink background. The two characters are close-up; Character A looks thoughtful with a finger near her mouth, Character B looks surprised. One large speech balloon says 「ブルースカイとかミクシィ2とはちゃうよな？」 and Character A has a small sound/text 「うーん…」.
2. Top-right panel, numbered with a black circle containing 「1」: inside the yakitori pub at a table. Character A smiles while holding up a beer mug; Character B sits beside her with her face covered by mosaic blur. Food and drinks are on the table. A vertical narration/speech balloon on the right says 「女性01ボールド、見てみ。サブスタが急成長らしいで」.
3. Middle full-width panel: both characters sit across the table, pointing upward with index fingers, excited. Use a dramatic yellow-orange burst background and big red sound effect 「ドンッ」 in the center. Left speech balloon says 「Xとかnoteも駆逐しかねんな！」. Right speech balloon says 「ほんまそれ！Xとかnoteもヤバい！」. Show 2 drinks and the yakitori/edamame dishes on the table.
4. Bottom-left panel: celebratory toast scene with both characters smiling widely and clinking beer mugs high. Orange radial burst background. Big red sound effect says 「カンパーイ!!」. Include two vertical speech balloons: left 「うちらも負けへんで〜！」 and right 「まぁ、楽しんだもん勝ちってことで！」.
5. Bottom-right panel, numbered with a black circle containing 「4」 and another black circle containing 「7」 near the top-left: quieter pub scene. Character A drinks from a beer mug with blue sound effect 「グビッ」. Character B eats a skewer with closed eyes and relaxed expression, blue text 「うんうん」 above her. Speech balloon says 「新しい場が広がるのはええことやけどな〜」. Include yakitori plates, edamame, and yellow drink glass on the wooden table.

Visual style: Rough handmade manga illustration, childlike but confident linework, uneven hatching, saturated pinks/oranges/yellows, black scribbly outlines, imperfect hand-lettered Japanese text, exaggerated facial expressions, humorous social-media commentary vibe.

Constraints: Use exactly 5 panels and the exact visible Japanese text listed. Do not add extra panels, extra characters, watermarks, realistic rendering, or clean vector art. Maintain the playful messy marker look and preserve the odd panel numbering.