Thiết kế ứng dụng / web

A DETAILED 3D INSTALLATION OF THE UPLOADED LOGO MOUNTED ON A SOLID CONCRETE WALL. ONLY THE LOGO ITSELF IS VACUUM-SEALED AND WRAPPED TIGHTLY IN A LAYER OF GLOSSY COLOR PLASTIC SHRINK-WRAP, FOLLOWING THE EXACT CONTOUR OF EVERY LETTER AND SHAPE. THE PLASTIC FILM TIGHTLY HUGS THE EDGES OF THE LOGO, CREATING REALISTIC CRINKLES AND SHARP SPECULAR HIGHLIGHTS ON THE LOGO'S SURFACE. THE BACKGROUND WALL IS CLEAN CONCRETE, FREE OF ANY PLASTIC WRAPPING OR FRAMES. STUDIO SPOTLIGHTS CAST SOFT, REALISTIC DROP SHADOWS: FROM THE WRAPPED LOGO ONTO THE WALL. RATIO 1080X1350 VERTICAL.

Hand-drawn concept sketch of [subject], rough construction lines, dynamic perspective, grayscale pencil rendering, industrial designer sketchbook style, creative brainstorming aesthetic

Goal: Create a realistic full-body and portrait reference character sheet for a modern young East Asian man, useful as a video/animation model reference. Canvas: Wide horizontal studio sheet, approximately 16:9 aspect ratio, light neutral gray background, thin white divider lines separating panels, clean editorial fashion catalog look. Layout: Use exactly 5 visible views arranged in two main zones. Left column contains 2 stacked portrait crops: top portrait facing camera with a neutral expression, bottom portrait looking slightly off to the viewer’s right. Right wide zone contains 3 full-body outfit references: front view, side profile view, and back view, evenly spaced across the gray studio background. Keep the front full-body view partially cropped at the neck/head like a clothing reference, while the side and back views show the upper body to just above the head/neck area. Subject details: The same young Japanese man appears in every panel, early-to-mid 20s, slim build, average height, pale-to-light skin, clean-shaven face, calm serious expression, natural uneven skin texture, realistic facial proportions. Hair is black, medium length, softly wavy, center-parted with loose curtain bangs and slightly tousled volume around the ears. Outfit: Minimal casual menswear: dark navy overshirt worn unbuttoned, white crew-neck T-shirt underneath, loose dark charcoal pleated trousers, black leather shoes. Hands are in trouser pockets for all full-body views. Clothing should have realistic cotton texture, folds, seams, collar, buttons, cuffs, and subtle studio shadows. Photography style: High-resolution photorealistic studio photography, soft diffused lighting from the front-left, accurate anatomy, natural posture, muted tones, crisp detail, shallow but not extreme depth of field. No text, no labels, no logos, no watermark, no props, no extra people.

<subject> in this style=<transformation> - Preserve the subject’s recognizable structure, but reinterpret it as a stylized editorial composition where drawing and miniature object-world logic coexist. - Convert some forms into sparse grainy line illustration while converting other forms into soft pastel miniatures or still-life objects. - Reinterpret texture-rich or detailed areas as small clusters of plants, props, vessels, terrain fragments, or decorative micro-world motifs. - Convert smooth surfaces into matte pastel color fields, while using graphic linework to describe edges, volume, and texture selectively. - Reduce complexity into a limited color system with one dominant graphic ink tone and a few quiet pastel support colors. - Replace deep realism with poster-like clarity, paper-print tactility, and curated tabletop arrangement. - Allow the subject to feel partially drawn and partially physically staged, creating a deliberate ambiguity between illustration and object. - Use negative space and sparse composition to keep the image clean, modern, and highly readable. </transformation> <shape_language> - Favor simple containers, rounded objects, leaves, stems, pebbles, tabletop silhouettes, miniature clusters, and clean abstract forms. - Use a mix of flat contour areas and soft low-relief sculptural accents. - Keep repetition rhythmic and decorative. - Let small details feel intentional and editorial rather than busy. </shape_language> <rendering_rules> - Use visibly grainy printed or crayon-like marks for lines and texture accents. - Keep volumetric forms soft, simplified, and pastel. - Limit the palette aggressively for a designed, print-friendly look. - Use shallow spatial depth with subtle diorama hints. - Make the final result feel like an art print, a magazine illustration, and a miniature set all at once. </rendering_rules> <finish> - Graphic limited-palette editorial hybrid. - Printmaking texture fused with soft pastel mini-world staging. - Smart, modern, collectible, and quietly distinctive. </finish> <avoid> - Avoid photoreal rendering. - Avoid deep cinematic perspective or strong realism. - Avoid overstuffing with too many tiny details. - Avoid using too many colors; keep it controlled and designed. </avoid>

Goal: Create a realistic screenshot of a minimalist AI chat interface showing a Japanese conversation where a user asks for a 16-bit pixel-art conversion and the assistant responds with a generated image. Canvas: Portrait screenshot, 768 x 896 px, white background, clean web-app UI with lots of empty space. No browser chrome, no visible sidebars. Use soft gray dividers, rounded cards, and small interface icons. Layout: Top-right user area contains exactly 2 discrete elements: 1 small square anime avatar, and 1 rounded white chat bubble. The avatar is an anime horse-girl face with brown hair, white forelock, purple eyes, horse ears, and a purple ribbon. The user bubble text is 16bitのピクセルアートにしてください. Assistant message block 1: Place the Manus logo and the bold name “manus” on the left. Below it, show exactly 1 small rounded notice pill containing a sparkle icon and Japanese text 今回は Manus が負担します—クレジットは消費されません. Then show the assistant reply line: 承知しました。元の表情と配色を保ちながら、16bit風のピクセルアートに仕上げます。. Under that, show exactly 2 checked task rows with gray circular check icons: 1) “編集要件を整理する” and 2) “16bit風ピクセルアートを生成する”. Beneath the tasks, show exactly 2 small action icons: a copy icon and a curved-arrow/regenerate-style icon. Assistant message block 2: Lower on the page, repeat the Manus logo and bold name “manus”. Under it, show a single Japanese sentence: 16bitゲーム機時代をイメージした、くっきりしたドット絵調に変換しました。. Below this sentence, align a large generated image preview to the left. Generated image preview: The preview is a square-ish pixel-art portrait cropped by the screenshot bottom edge. It shows a cheerful anime horse-girl character in crisp 16-bit game style: brown bob-length hair with a large white forelock, tall horse ears, a purple bow on the left ear, purple sparkling eyes, small blush marks, open smiling mouth, and white hair ties near both sides. Use chunky pixel edges, visible dithering, limited retro palette, and dark outline pixels. Background inside the pixel-art image is a warm indoor room with beige wall, window light, and horizontal furniture/trim, all rendered in pixel art. The visible generated preview should occupy the lower-left portion and be cut off at the bottom of the screenshot. Additional UI details: Add a small floating circular down-arrow button at the bottom-right corner. Keep the overall look like a Japanese Manus chat page, with generous margins, black body text, subtle gray UI accents, and no true transparency checkerboard. Count of visible repeated UI/content elements: 1 user avatar, 1 user message bubble, 2 Manus headers, 1 notice pill, 2 checked task rows, 2 small action icons, 1 generated image preview, and 1 floating down-arrow button. Constraints: Preserve the Japanese text exactly where given in the default parameter values. Do not add extra messages, extra avatars, extra buttons, watermarks, browser tabs, or side navigation.

Goal: Create a cute hand-painted 2D game UI screenshot for a hamster-powered sunflower seed cannon typing game, with a cheerful rural sunflower-field theme and Japanese interface text. Canvas: Wide 16:9 game screen, bright daytime sky, saturated storybook watercolor/gouache style, thick clean outlines, warm cream-and-wood UI panels, sunflower decorations, playful arcade energy. Layout: Top HUD, center battlefield, right target status board, and bottom input/control panel. Use exactly 3 top HUD elements: left rounded cream badge showing “COMBO 6” with a sunflower icon, center wooden title plaque reading ひまわりの種キャノン decorated with sunflowers and leaves, and right rounded cream badge showing “発射回数 12” with small leaves and a sunflower. The main scene is a rolling green meadow filled with sunflowers, a white fence, distant hills, fluffy clouds, drifting petals, dandelion fluff, and flying sunflower seeds. Subject details: On the left, place exactly 1 adorable round hamster character operating exactly 1 wooden-and-brass sunflower cannon. The hamster is beige and brown with glossy black eyes, tiny paws, and holds a red-topped lever. Add exactly 1 speech bubble above it reading “ピッ！”. The cannon is small, wheeled, decorated with sunflower motifs, with a large sunflower-shaped muzzle firing seeds to the right. Show one large yellow arrow-shaped projectile banner coming out of the cannon reading 私の担当じゃねーんだよ！！！！. Include multiple visible black-and-white sunflower seeds streaking across the scene, plus yellow leaves, sparkles, and motion trails. Right target board: Create exactly 1 large wooden target sign on the right, wobbling from impact, decorated with exactly 3 prominent sunflowers and a small “TARGET” label at the top. The target name should read 理不尽上司. Under it, show exactly 1 horizontal HP bar with red-orange-yellow fill on the left and dark depleted area on the right, labeled “872 / 1500 HP”. Below the bar, show a big comic impact label “-128 DAMAGE!” in red-orange letters with white outline. Add starbursts and seeds striking around the board. Bottom control panel: A large cream wooden bordered panel spans the full width. It contains exactly 1 typing input box on the left with large black text 私の担当じゃねーんだよ！！！！ followed by a vertical cursor. On the right is exactly 1 oversized rounded yellow button labeled “発射” with smaller text “Enterでも発射” beneath it, decorated with sunflowers. Along the bottom left is exactly 1 “POWER” label and exactly 1 long power meter with a green-to-yellow-to-red gradient, plus several seed markers inside the gauge. Visual style: Whimsical Japanese mobile game UI, cozy handmade illustration, polished mockup, high readability, bold Japanese typography, wooden texture, decorative rivets, stitched borders, subtle drop shadows, dynamic comic hit effects, no photorealism. Constraints: Keep all listed UI text legible and positioned as described. Use exactly the visible major interactive/status elements: 3 top HUD elements, 1 hamster, 1 cannon, 1 speech bubble, 1 projectile phrase banner, 1 target board, 1 HP bar, 1 damage label, 1 input box, 1 fire button, and 1 power meter. Do not add extra characters, menus, watermarks, logos, or unrelated UI panels.

16:9 cinematic wide frame, oriental immortal realm hyper-scale environmental scene, refined cinematic matte painting, photorealistic fantasy environment, gray-blue twilight series. An oriental divine civilization floating above an endless sea of clouds, with architectural scales approaching geological features. The entire image revolves around a single spatial theme: a massive royal celestial palace hanging upside down from the sky, its lower part descending like an artificial mountain range covering the heavens. The camera stands inside a high-altitude white jade terrace at an eye level of about 1.4 meters, using a 35mm wide-angle lens, looking straight ahead with a slight upward tilt. No aerial shots, no bird's-eye views. The foreground left is heavily framed by an exceptionally large vermilion lacquer pillar, oily dark wood beams, and dark cyan double-eaves pressing into the top of the frame; the right side uses a semi-transparent cyan jade screen and carved white jade balustrades to form a second framing layer. The middle ground is a very wide but empty royal white jade terrace, its polished jade surface reflecting the cold gray sky, with only 3 tiny immortals (about 1% of frame height) standing quietly at the end. Beyond the terrace, it drops immediately into a massive cloud sea fault. In the distance above, a whole inverted royal palace hangs down from above the clouds. It is not just a building, but an entire sky continent carved into a palace: inverted eaves, giant brackets, white jade platforms, dark wood pillars, and dark cyan glazed roofs form a giant artificial landscape. Only a portion is visible. At least 50% of the inverted architecture is obscured by high-altitude gray-blue mist, cloud curtains, foreground eaves, and the frame. The uppermost connection disappears into the sky, while the descending ends are swallowed by blue-gray clouds, leaving the thickness and height of the palace ambiguous. Some parts even show massive inverted courtyards and descending bases, but do not add a second visual protagonist. About 80% of the frame is controlled by gray-blue sky, blue-gray cloud sea, cold air, and deep cyan-gray shadows. Royal materials are extremely refined: vermilion high-gloss lacquer pillars, dark wood beams, white jade platforms, delicate jade screens, dark cyan glazed eaves, and partially gilded bracket edges. Gold is strictly controlled, appearing only on edges. Soft directional light from the right passes through clouds, creating a tiny amount of amber rim light on the eaves and beams. Foreground detail is sharp, mid-ground moderate, and distant massive blocks are simplified into dark gray-blue silhouettes. Cold large field + warm localized accents + luminous ivory mist, monumental inverted celestial palace as artificial terrain, architecture swallowing the sky, solemn stillness, deep atmospheric layering, large depth of field, no ruined architecture, no crowded palace city, no sci-fi spacecraft feeling.

{ "identity": { "reference": "Use the uploaded face photo as the STRICT identity reference.", "priority": "ABSOLUTE", "preserve": [ "exact facial identity", "facial proportions", "eyes", "eyebrows", "nose", "lips", "jawline", "ears", "skin tone", "hairline", "hairstyle", "recognizable appearance" ], "gender": "Automatically adapt naturally for male or female." }, "image": { "title": "Metropolitan Identity", "aspect_ratio": "4:5", "style": [ "Ultra Photorealistic", "Luxury Editorial", "Swiss Graphic Design", "Double Exposure", "Fine Art Portrait", "Minimal Poster Design" ], "composition": { "shot": "Waist-up portrait.", "camera_angle": "Three-quarter profile.", "background": "Clean light gray studio backdrop.", "framing": "Centered minimalist editorial composition." }, "subject": { "expression": { "emotion": "Calm, elegant, confident.", "eyes": "Looking slightly downward into the distance.", "overall": "Professional luxury fashion model." }, "hair": { "style": "Natural premium editorial styling." }, "wardrobe": { "description": "red" } }, "city": "<New York>", "placement": "Modern skyscrapers are visible only inside the silhouette of the subject's clothing and lower body. The face remains clean and fully recognizable.", "blend": "Soft realistic double exposure with architectural transparency." }, "graphic_design": { "style": "Modern Swiss typography layout.", "accent_color": "<ACCENT COLOR>", "elements": [ "large translucent horizontal rectangle crossing the eyes", "minimal geometric blocks", "clean vertical typography", "subtle transparent overlays" ], "text": "<New York>" } }

Perfect for modern design and conceptual projects, this prompt creates serene and balanced minimalist scenes. These visuals use clean compositions, negative space, and subtle textures to evoke calmness and sophistication.

Goal: Create a clean 16:9 Japanese corporate proposal cover slide for a construction/civil engineering or industrial automation company, introducing recruitment and PR branding preparation for ZAXIS. Canvas: Wide presentation slide, 1200×675 or 16:9 ratio, bright white background with soft lavender gradients. Use a modern consulting-deck aesthetic: generous whitespace, bold typography, rounded cards, and subtle tech/corporate accents. Layout: Top left has a small rounded purple square badge with white “01”, followed by the company name in black. Below it, place a large bold black Japanese headline “採用・PR” and directly underneath a large bold lavender headline “ブランディング基盤整備”. Add a thin horizontal lavender rule with a small circular dot at the left under the headline. In the left-middle area, place a rounded square outline icon card showing a simple people/community icon in lavender. To the right of that icon, add two lines of black Japanese body copy: “求人募集を始める前に、” and “信頼される公式情報の受け皿を整える”. Near the lower left, place a vivid lime-green rounded pill button containing a small white circle with a black flag icon and the text “第一段階のご提案”. Right side visual: Place one large rounded-rectangle photo card occupying the right half of the slide, with a subtle drop shadow and thin white border. The photo should show a bright futuristic factory floor with orange and white industrial robotic arms assembling mechanical parts, blue-tinted machinery, and a translucent holographic CAD blueprint/interface floating above the work area. The image should feel like advanced manufacturing, automation, precision engineering, and smart factory technology. Background details: Add a large semi-transparent lavender circular gradient shape in the upper-right corner behind the photo, and a soft lavender wave/blob sweeping upward from the lower center toward the right. Add a small dotted halftone pattern in pale lavender at the bottom-left corner. Keep the overall look polished, airy, and presentation-ready. Visual style: Minimal Japanese B2B proposal design, crisp sans-serif typography, high contrast black and purple text, accent color neon lime, rounded corners, subtle shadows, premium consulting slide layout. Text parameters: Use 採用・PR, ブランディング基盤整備, 求人募集を始める前に、\n信頼される公式情報の受け皿を整える, and 第一段階のご提案 exactly as visible Japanese text unless customized. Constraints: Include exactly one section badge, one company name, two headline lines, one divider line with one dot, one people icon card, one two-line body text block, one lime-green CTA pill with one flag icon, one main factory photo card, one upper-right lavender circle, one lower lavender wave/blob, and one dotted bottom-left pattern. Do not add extra logos, watermarks, charts, or additional paragraphs.

Using REFERENCE_0 as the source material, create a dark-mode Japanese evaluation report slide that compares attempts to generate a character sheet while preserving the same rough watercolor/ink illustration style. Goal: Turn the reference into a structured experiment-summary dashboard titled 「M2　ステップ2—ムードボードを見本に GPT でキャラクターシート」. Keep the subject as the small armored child character from the reference, but present the result as an analysis screen rather than a single illustration. Canvas and style: Use a wide 16:9 dark UI layout with charcoal background, thin dividers, rounded cards, small Japanese UI text, orange/green accent labels, and embedded white image panels. Make it look like a screenshot of an internal comparison report. Top section: Create exactly 3 comparison cards across the top. Each card contains a white character-sheet panel showing the same character in multiple views and a small detached head/bust crop. Use these card titles and edge scores: 1. 「経路A：GPT ムードボード経由」 with 「エッジ 57.6」 2. 「経路B：grok ムードボード経由」 with 「エッジ 52.6」 3. 「参考：grok キャラ直接（B-5）」 with 「エッジ 48.8」 Under each card add one short Japanese note line; the third card should mention that it is the softest and inherits the sample’s green cloth. Middle section: Add a comparison table with exactly 4 columns labeled 「条件」, 「エッジ強度」, 「コントラスト」, 「見本の欠陥」. Include exactly 7 data rows: 1. 「grok（目標）」 — 47.2 — 82.4 — 「—」 2. 「1) 共通のみ」 — 53.6 — 90.8 — 「—」 3. 「2) + 筆致（言葉）」 — 52.0 — 88.4 — 「なし」 4. 「3) + grok キャラ見本」 — 「48.8 ← 最良」 — 85.4 — 「緑の布を継承」 5. 「4) + GPT ムードボード」 — 「57.6 ← 最悪」 — 84.5 — 「なし」 6. 「5) + grok ムードボード」 — 52.6 — 87.3 — 「なし」 7. 「素材 image3.png（参考）」 — 58.8 — 69.1 — 「—」 Bottom section: Add several short Japanese conclusion lines beginning with 「結論1」 through 「結論5」, explaining that moodboard routes are not very effective, direct character reference works better, and the practical workaround is to avoid moodboards and use the Grok output with cropping. Finish with a collapsible dark bar labeled 「プロンプト（ステップ2の冒頭：ムードボードの役割指定）」. Constraints: Preserve the character identity and rough painted look from REFERENCE_0 in the embedded panels, but do not recreate the full original reference screenshot. Do not add unrelated characters. Use crisp UI typography and readable Japanese labels. The main title topic may be customized as M2　ステップ2—ムードボードを見本に GPT でキャラクターシート; the three route labels may be customized as 経路A：GPT ムードボード経由 / 経路B：grok ムードボード経由 / 参考：grok キャラ直接（B-5）; the character-sheet subject may be customized as the small armored child knight from the reference; the table metric emphasis may be customized as 48.8 ← 最良, 57.6 ← 最悪.

Construct a modern graphic visual around any theme object by intersecting giant Chinese thematic characters, geometric color blocks, and single-line organic outlines. Maintain high brightness and a pure background overall, leaving more than two-thirds of the space white, with an off-center cluster bearing the main weight: a saturated main theme color rectangle establishes a stable plane, a deeper low-brightness cool-toned circle overlaps from below and extends beyond the boundary, and the core Chinese theme character advances vertically with oversized heavy black geometric strokes across the white base, rectangle, and circle, forming strong scale jumps through incomplete occlusion and cropping; the left side features a compact digital column of dates and bilingual sub-headers, while the right side balances the weight with two levels of vertical short sentences, and small micro-descriptions, short dashes, and linear symbols are scattered at the edges, creating a three-level reading rhythm from the main character to side notes and then to details. Abstract the dynamic relationship derived from the theme into a continuous organic outline, using clear near-black single lines to outline a broad swirling form that enters the color blocks from large areas of white space and crosses the edges, with curves intersecting multiple times but maintaining a discernible single path; the interior of the outline mostly retains the base color, with only local transparent thematic main color fades added at the tips and extensions, creating tension between the light lines and the heavy giant characters. Colors are adaptively selected by theme: the largest area is borne by a clean high-brightness neutral base, the main theme color remains bright and highly saturated for rectangles and local transparent gradients, the circle uses a deep cold supporting color pulled away from the main color's hue, blending into the main color at the end to form soft diffusion, and text and outlines use near-black, with a very small amount of light-colored text pressed against the main color block; fix unequal areas, light/dark levels, and saturation differences, without fixing specific color names. All elements present a flat surface like matte digital printing, with sharp outline edges, gradients only slowly feathered in two parts, and no three-dimensional shadows, paper grain, or global lighting effects. The font maintains a low-contrast, thick, near-geometric sans-serif skeleton, with main characters allowed to have components elongated, displaced, cut, or changing background across color blocks; numbers use a rounded futuristic structure, and small text is enclosed by narrow bodies and loose letter-spacing; brand names, advertising copy, and specific dates are all replaced by new theme information. Finally, preserve extreme white space, geometric gathering below center, vertical and horizontal text pulling against each other, and the moment organic lines cross rigid graphics, avoiding filling parts evenly or simplifying outlines into isolated decorations.

Build a modern flat graphic around any chosen subject object, with a huge Chinese theme character, geometric color blocks, and a single-line organic contour interlacing through one another. Keep an overall bright, clean background field, letting approximately two-thirds or more of the space remain empty, and concentrate the main weight in an off-center cluster: a saturated rectangular block in the theme's primary color first establishes a stable plane; a deeper, low-lightness, cool-toned circle overlaps from beneath it and extends beyond the boundary; the core Chinese theme character advances vertically in oversized, heavy, geometric black strokes across the white background, rectangle, and circle, with partial occlusion and cropping creating a strong scale jump; on the left, the date and a bilingual subtitle form a compact numeric column, while on the right, short vertical sentences set in two stroke-weight tiers balance the weight; a few miniature captions, short dashes, and linear symbols scatter along the edges, creating a three-level reading rhythm from main character to side notes to details. Abstract the dynamic relationship derived from the theme into one continuous organic contour, drawn as a broad sweeping loop with a clear, near-black single line that enters from the large empty field, passes through the color blocks, and crosses the edge; the curve intersects itself several times while remaining legible as a single path; keep most of its interior as the underlying background color, adding only localized, transparent, fading fills in the theme's primary color at the tips and extended sections, so the light line creates tension with the heavy giant character. Adapt the color scheme to the theme: let a clean, high-lightness neutral background carry the largest area; keep the theme's primary color vivid and highly saturated, using it for the rectangle and the localized transparent gradient; use a deep cool support color for the circle, separated in hue from the primary color and blending softly into the primary color at its end; use near-black for text and the contour, with a very small amount of light text reversed onto the primary color block; fix unequal areas, lightness hierarchy, and saturation differences, but do not fix specific color names. Render all elements as flat surfaces like matte digital printing, with sharp contour edges and gradients only as two localized, slowly feathered transitions; do not add dimensional shadows, paper grain, or global lighting effects. Keep typography in a low-contrast, broad, thick, near-geometric black sans-serif skeleton; allow the main character's parts to be elongated, offset, cut out, and switched between color-block backgrounds; give numbers a rounded futuristic structure and compress small text with narrow letterforms and loose tracking. Replace brand names, advertising copy, and concrete dates with new theme information. Preserve in the final result the extreme white

Goal: Create a dark futuristic widescreen tech poster introducing 今日分享：5 个 GitHub 爆涨星技能, showing five fast-growing GitHub AI tools as large interface cards above a modern developer workstation. Canvas: 16:9 horizontal social media banner, cinematic dark office background, cool blue lighting, high contrast, sharp UI mockup style. Use a blurred glass-and-metal workspace at night, with a desk across the bottom third. Layout: Place a large centered Chinese headline at the top; make the number “5” oversized and bright electric blue while the rest of the headline is bold white. Under it, add a thin decorative line and small subtitle text reading 真实工作流场景一图看懂. Arrange exactly 5 numbered feature sections in a row across the middle, each with a circular number badge, tool name, short Chinese description, and a large screenshot-style UI card beneath it. The five sections are: 1) GPT-Image2-Skill, description “图像生成与创意设计”; 2) smart-ralph, description “智能开发规划与执行”; 3) OpenAI Skills, description “可复用的 AI 技能目录”; 4) open-saas, description “开箱即用的 SaaS 基础设施”; 5) awesome-copilot, description “Copilot 配置与提示词库”. Subject details: Section 1 card is a dark prompt-library interface titled “GPT-Image2-Skill,” with a left sidebar of categories and a grid of poster/UI/illustration thumbnails. Section 2 card is a dark coding workflow interface titled “smart-ralph,” showing a task checklist, code editor, checkpoints, and an approval dialog. Section 3 card is a clean light “Skills Directory” interface with category navigation and stacked skill cards such as Code Reviewer, API Documenter, SQL Assistant, and Data Analyst. Section 4 card is an open-saas dashboard mockup with a landing page headline “Build Faster. Launch Sooner.”, login panel, dashboard metrics, payments table, and deploy panel. Section 5 card is a dark awesome-copilot marketplace/library interface with tabs and exactly 6 visible tool rows: Python Best Practices, API Development, Test Generator, Code Refactor, Security Scanner, and Docs Generator. Desk foreground: Include exactly 4 main desk objects: an open laptop on the left showing a mountain wallpaper, two wide desktop monitors in the center showing code and analytics dashboards, a white keyboard with mouse, and a small potted plant near the right monitor. Keep the desk minimal, premium, and realistic. Visual style: Premium GitHub/AI tooling aesthetic, dark navy and black palette, electric blue accents, crisp white typography, glassmorphism UI panels, realistic reflections, subtle depth of field, cinematic lighting, polished startup presentation design. Constraints: Keep all five tool names readable. Do not add extra tools or extra numbered sections. Preserve the Chinese headline and short Chinese descriptions. Avoid watermarks, logos unrelated to the five tools, clutter, distorted text, or cartoon styling.

A highly refined contemporary mixed-media portrait style combining an ethereal high-key fashion aesthetic with intricate concentric contour-line artwork inspired by natural growth rings and fingerprint-like topographic patterns. The human subject should retain an elegant, soft, luminous, almost dreamlike presence while the entire visual structure is organically constructed through hundreds of delicate continuous concentric and flowing contour lines. Replace conventional flat shading and ordinary outlines with dense nested curved lines that follow the anatomy and three-dimensional structure of the human form. The lines should naturally conform to the contours of the face, forehead, cheeks, nose, lips, jaw, neck, shoulders, clothing, and hair, creating the illusion that the entire subject is sculpted from flowing rings and fine engraved contours. The line patterns should behave like organic topographic growth rings rather than geometric circles. Lines continuously expand, contract, bend, merge, separate, and change direction according to the underlying form. Around the eyes, nose, lips, and facial planes, the contours become tightly concentrated and delicately curved; across larger surfaces they gradually spread into wider flowing rings. Hair should be constructed from extremely fine flowing contour strands, with hundreds of elongated curved lines sweeping outward and merging into larger concentric patterns. The hair should feel weightless and fluid, almost dissolving into the surrounding background while maintaining an intricate line-based structure. Preserve the reference's soft luminous skin, delicate facial modeling, subtle blush, translucent highlights, pale neutral tones, and graceful editorial beauty. The human form should remain recognizable and aesthetically refined beneath the contour-line construction rather than becoming an abstract pattern. Introduce selective high-density black and charcoal contour areas to create depth and visual emphasis. Lines become progressively darker and closer together in recessed areas and shadow regions, while highlights contain sparse, extremely fine pale-gray lines and larger areas of negative space. This creates tonal values entirely through line density and spacing, rather than conventional painted shading. Allow portions of the contour lines to extend beyond the silhouette, forming expansive organic rings and flowing arcs around the subject. These external lines should gradually fade, break apart, become thinner, and dissolve into the background, creating an atmospheric transition between the human figure and the surrounding space. Use a predominantly luminous ivory, white, soft gray, pale beige, and subtle natural skin-tone palette, with deep charcoal-black lines providing controlled contrast. Maintain a clean, airy, almost monochromatic elegance while preserving subtle full-color skin undertones. The surface should have a fine tactile paper or archival-art texture,

Goal: Create a cinematic editorial social media banner for an AI model review of Qwen Image 3.0 Pro, combining a luxury lounge fashion photograph with bold tech-review typography. Canvas: Wide landscape 16:9 banner, about 1200×675 px, optimized for X/Twitter link preview. Use a warm golden color grade with high contrast and shallow cinematic lighting. Main image: A fashionable adult woman lounges diagonally across a mustard-yellow velvet sofa in an art deco interior. She has loose wavy brown hair, red lipstick, and a confident relaxed expression, wearing a deep royal-blue velvet blazer dress with gold buttons, bare legs, and dark brown knee-high boots. Her left side is toward the viewer; she holds a champagne flute in one hand near the left edge while reclining with legs extended toward the lower right. The room has patterned gold wallpaper, dark wood wall paneling, a green carpet, and a glowing vintage floor lamp near the upper center-right. Include a small side table on the right with stacked art/design books and a draped burgundy fabric, creating a 1920s Parisian hotel lounge mood. Text layout: Place large review text on the right half, leaving the woman mostly unobstructed on the left. Use exactly 3 main text elements: top label “Model Review” in bold teal-green sans serif, main title “Qwen Image” in very large bold white sans serif, and version line “3.0 Pro” in extra-large bold teal-green sans serif. Add a small footer credit near the bottom left-center reading “302.AI | Benchmark Labs” in thin white sans serif. Logo area: At the bottom right, show exactly 3 discrete logo/comparison elements: a rounded white square containing a purple circular clover-like AI icon on the left, a white “×” multiplication sign in the middle, and a rounded white square containing a purple geometric star/cube-like logo on the right. Visual style: Luxury editorial photography blended with clean SaaS benchmark branding; crisp typography, warm amber highlights, teal accent text, deep green floor, mustard sofa, royal-blue outfit. Make the composition look like a polished review thumbnail for Qwen Image 3.0 Pro versus GPT-Image-2. Constraints: Keep all visible text legible and spelled exactly as specified. Do not add extra captions, watermarks, interface chrome, or additional logos. Preserve the wide banner layout and the three-logo comparison cluster.

Create a 9:16 vertical photorealistic office portrait using the second image as the exact face reference. Keep her facial identity, hair, and features unchanged. Place her seated at the same modern office desk with laptop, documents, books, floor-to-ceiling windows, city view, and wall frame from the first image. Keep the same outfit and professional mood. Soft natural daylight, premium corporate aesthetic, realistic details, 4K.

Goal: Create a realistic dark-mode social media post screenshot about a leaked AI image model, styled like an X/Twitter post. Canvas: Square 1024×1024 image, black background, with a single tweet occupying the full frame. Use authentic X/Twitter dark mode spacing, white text, blue hashtag/link color, and slightly rounded embedded media corners. Tweet header: Top left circular profile photo of a light-skinned middle-aged man in a white shirt. Display name is White in bold white text with a blue verified checkmark, and handle @White1637402 in gray beneath or beside it. Top right has two small UI icons: a stylized X/Twitter action icon and a three-dot menu. Tweet text: Use the exact visible post copy: “A new GPT image model just appeared on Arena called mona-lisa-1. It's extremely better than Image 2. 💪💪” followed by a blank line and the blue hashtag #ChatGPT. Keep line breaks and spacing like a real tweet. Embedded media layout: Below the tweet text, show exactly 2 side-by-side image cards in one row, separated by a thin black gutter, both with rounded outer corners. Left media card: A glossy 3D illustration of a cute white-and-black robot mascot in a classroom or chalkboard setting, waving with its right hand and holding a paper sign. The robot has a rounded black face screen with two glowing cyan smiling eyes, white helmet-like head shell, white body panels, black joints, and a friendly pose. Behind it is a chalkboard covered with handwritten educational text and doodles. Large chalk text reads “Hi there! I’m ChatGPT” on the left, and “That’s me!” with an arrow pointing toward an OpenAI swirl logo on the upper right. The robot’s paper sign reads “I was created by OpenAI” with smaller text underneath. Add realistic 3D lighting, slight depth of field, and toy-like materials. Right media card: A dramatic cinematic poster-like fantasy/business scene showing exactly 3 human figures: two giant suited men looming in the stormy cloudy sky above, and one small caped figure seen from behind in the foreground. The two giant men resemble tech executives, wearing dark business suits and looking down sternly; one has glasses and gray hair, the other has curly hair. The foreground figure wears a bright royal-blue cape with a white logo and the word “claeposeok” or similarly misspelled text on the back, standing on rocky ground and holding a sword at their side. The sky is dark, smoky, and epic with strong rim lighting, blue-gray clouds, and heroic scale. Visible text count: Include exactly 7 main visible text groups: 1 display name “White”, 2 handle “@White1637402”, 3 tweet body sentence, 4 hashtag “#ChatGPT”, 5 chalkboard phrase “Hi there! I’m ChatGPT”, 6 chalkboard phrase “That’s me!”, 7 paper sign “I was created by OpenAI”. Minor chalk scribbles may be present but should be small and not dominant. Visual style: Photorealistic screenshot compositing with crisp UI text, realistic embedded image thumbnails, high contrast, and no extra watermark. Keep the overall look like a viral tech leak tweet screenshot discussing mona-lisa-1.