AI 提示词库

Overall redraw and optimization of the image without changing the {argument name="original composition" default="original composition"}, {argument name="character pose" default="character pose"}, character design, original lighting effects, and lens relationship. Maintain the {argument name="illustration style" default="2.5D Japanese illustration style"}, but reconstruct the frame for a cleaner, more unified visual output. Perform global readjustment of color and lighting on the existing image so that the overall light and shadow relationship follows a single, stable lighting logic. Use a soft, uniform primary light source, avoiding layered shadows and fractured color blocks caused by multiple light sources. Shadow edges should transition naturally without hard boundaries or blocky filling. Unify material representation, ensuring continuous gradient structures in the color transitions of skin, hair, and clothing, reducing local noise, color block accumulation, and non-continuous shadow areas. Avoid any mosaic, dirty spots, or cloud-like color structures. The overall style should be within the high-quality 2.5D animation illustration range, with a slight sense of volume and soft spatial depth, but without introducing realistic skin textures, pore details, or photographic sharpening effects. Retain a moderate hand-painted feel and smooth color gradations. Reduce local high-frequency detail interference, avoid over-sharpened edges and excessive texture enhancement, and give the picture a clean visual fluidity and unified color rhythm. Keep character outlines and hair structures continuous, smooth, and consistent, avoiding edge jitter, breakage, or granular jaggedness. Lines should resemble stable hand-drawn animation line art rather than boundaries formed by local sharpening or noise stacking. Outline changes should transition smoothly without local 'jumps' or irregular fluctuations. Overall reduce microscopic contrast and local sharpening tendencies, keeping edge transitions soft for visual fusion. Avoid knife-cut hard edges or overly clear separation at the boundaries of skin, hair, and clothing. Highlights and shadows should be connected by flexible gradients rather than sharp-line structures. The overall look should be 'clear but not sharp,' retaining physical recognition while suppressing pixel-level sharpening marks. The final output should be a clear, soft, high-quality 2.5D illustration without regional contamination, featuring a stable lighting system, continuous line structures, and natural gradient transitions.

Premium {argument name="snack type" default="Crunchy Potato Sticks"} Product Commercial Storyboard FORMAT: • Single-page premium storyboard • 3:4 Portrait ratio • Modern snack food campaign • 8 cinematic product-focused scenes • Product remains the main hero • Premium advertising agency presentation HEADER: • Bold contemporary typography • Information cards: - Duration: 20 Seconds - Style: {argument name="commercial style" default="High-End Snack Commercial"} - Product: Crunchy Potato Sticks - Audio: Crisp Crunch + Fast ASMR • Why This Style Works section • {argument name="color palette" default="Golden yellow, cream and black"} aesthetic • Minimal geometric snack-inspired accents STORYBOARD: 1. Premium snack tube standing dramatically on a glossy surface 2. Container opening with crunchy potato sticks visible inside 3. Potato sticks cascading out in slow motion 4. Extreme macro revealing ridges, seasoning and crispy surface 5. Seasoning particles floating around the product 6. Several sticks snapping together with tiny crumbs suspended in air 7. Potato sticks arranged in a dramatic circular composition around the package 8. Final hero product shot with the package, overflowing snack sticks and controlled crumb explosion EVERY PANEL: • Scene number • Duration badge • Camera direction • Visual • Action • Product detail CAMERA: High-speed photography, extreme macro, slow-motion product movement, floating crumb capture, low-angle product shot, dramatic push-in, clean hero framing. STYLE: Ultra-realistic snack advertising, incredibly detailed crispy texture, realistic seasoning particles, sharp packaging, controlled studio lighting, glossy reflections, premium commercial photography, 8K.

A gigantic abstract entity formed entirely from swirling black mist, its body loosely humanoid but constantly shifting, a large traditional Noh mask emerging from the fog as its face, no solid body, only dense smoke shaping arms and torso, black mist flowing and dissolving into the surrounding air, dark mountainous landscape beneath, soft diffused moonlight cutting through the fog, highly cinematic realism, symbolic and minimal, film still, subdued color palette, 35mm lens look

双岸记忆之翼 龙不选择世纪。 是世纪在选择自身， 穿过由拖延与余晖构成的躯体。 一翼拖曳着图书馆最后一页未被焚毁的纸张， 那是在纸张学会自己的名字之前就已陨落的文明。 另一翼梳理着全息雨幕， 来自那些尚未决定 是屹立不倒还是化为传闻的高塔。 在双翼之间，空气保留着两份历法， 却忘记了哪一份正在运行。 它的鳞片比“鳞”这个词本身更为古老。 它们记得那些尚未点燃的 熔炉的炽热， 以及在某种叙述中 早已熄灭的 星辰的寒冷。 符文如同一座城市尚未获得思考资格的念头， 浮现又沉没。 每一次光脉冲既是遗迹也是原型。 下方，街道湿润，承载着已然发生之事 以及仍在协商抵达之事。 霓虹灯在曾是冰川耐心的水面上书写， 而后来，那只会成为关于干渴的传闻。 地基坐落在森林的静谧之上， 森林仍梦见自己依然挺立。 广告牌宣传着明天， 却已对尚未到来的黄昏感到思乡。 这生物的双眼承载着唯一诚实的天气： 一只眼是记忆的翠绿， 尚不知晓自己已然终结； 另一只眼是疑问的湛蓝， 尚未找到将要问出它的那张嘴。 它们注视我们，如同时间在无人窥视时 注视着自身—— 没有怜悯，没有判决， 只有长久的、礼貌的凝视。 我们生活在它掠过时短暂的气流中， 那奇特而仁慈的间隙， 过去尚未与我们了结， 而未来已开始 怀念我们尚未失去的一切。 除了转动，别无承诺： 那缓慢而明亮的翅膀铰链， 同时属于每一个时刻， 却又不完全属于其中任何一个。 它留下的薄雾既非终结，也非开端。 那是它们在下一个不确定性 张开安静而必要的手掌之前， 短暂相认的地方。

High-impact cinematic depiction of {argument name="character name" default="LEAD_CHARACTER_NAME"}, featuring youthful features with light stubble, messy short brown hair, and an astonished yet amused expression while pointing upward, dressed in a casual light grey t-shirt. Positioned in a crowded theme park plaza featuring a large illuminated fantasy castle beneath a massive hovering metallic UFO, accompanied by {argument name="crowd" default="BACKGROUND_CROWD"}. The scene conveys a surreal sci-fi invasion mood through gloomy overcast natural lighting mixed with glowing UFO rim lights, with dramatic highlights and shadows. Captured using a wide-angle smartphone selfie camera perspective with dynamic motion blur, shallow depth of field, cinematic framing. Ultra-detailed textures, realistic skin tones, professional cinematic color grading, sharp focus, HDR, ultra-high resolution, 4K, masterpiece quality.

标题： 高端草莓芝士蛋糕冰淇淋产品广告项目 格式： • 单页高端项目 • 3:4 竖屏比例 • 电影级冷冻甜点广告宣传片 • 8 个完全不同的产品聚焦场景 • 产品始终作为核心视觉主体 • 高端广告代理商演示风格 页眉： • 粗体编辑排版 • 信息卡片： * 时长：{argument name="duration" default="10 秒"} * 风格：电影级甜点广告 * 产品：{argument name="product" default="草莓芝士蛋糕冰淇淋"} * 音频：勺子刮擦声 + 奶油甜点 ASMR • “为何这种风格有效”部分 • {argument name="color palette" default="柔粉色、草莓红、奶油色和香草色"} 美学 • 极简草莓与甜点主题装饰细节 项目内容： 1. 高端草莓芝士蛋糕冰淇淋桶戏剧性地矗立在冰镇大理石台面上，产品包装正面清晰可见 2. 打开冰淇淋桶，特写镜头展示盖子被掀开，露出下方奶油质感的冰淇淋 3. 冰淇淋勺压入表面，展现出丰富的草莓芝士蛋糕层次和逼真的奶油质感 4. 草莓果肉和芝士蛋糕碎屑嵌入冰淇淋中的极致微距镜头 5. 新鲜草莓酱缓慢流过完美成型的冰淇淋球，形成光泽感十足的红色丝带 6. 冰淇淋球被盛入玻璃甜点杯中，细小的芝士蛋糕碎屑自然散落周围 7. 勺子切开奶油冰淇淋球的特写，露出内部的草莓馅料和柔软的芝士蛋糕块 8. 最终产品定格镜头，原装冰淇淋桶放置在优雅的甜点杯旁，新鲜草莓、碎屑和草莓酱围绕在产品周围 每个面板： • 场景编号 • 时长标签 • 运镜方向 • 视觉画面 • 动作描述 • 产品细节 摄像机： 极致微距美食摄影、平滑的挖勺动作、慢动作草莓酱浇淋、细腻的质感特写、受控的推拉镜头、浅景深、优雅的俯拍构图、电影级最终包装定格。 风格： 超写实高端冷冻甜点广告、浓郁的奶油冰淇淋质感、逼真的草莓果肉、细腻的芝士蛋糕碎屑、光泽感草莓酱、真实的冷凝水珠、清晰的包装细节、柔和的电影级灯光、奢华的美食造型、专业甜点广告、8K 分辨率。

AI Image Generation 4-Panel Manga Title Absolute Zero Ignore Super ignore A cannon of disinterest explodes If the goal was to cause a flame war Serves you right! I guess? lol

A girl in a fully closed modern white spacesuit with a tinted visor stands on the surface of Mars. Red-orange rocky desert, dusty pinkish sky. Opposite her stands a tall humanoid {argument name="alien description" default="alien with pale pink semi-transparent skin, an elongated skull, large black almond-shaped eyes, and a graceful slender body"} in a tight-fitting bio-organic suit of silvery-pearl color. Realistic detailed anatomy. The girl is handing him an {argument name="human item" default="iPhone in a pink case"}. The alien is handing her a {argument name="alien item" default="crystal vial glowing from within, containing a shimmering blue substance"}. The moment of exchange. Both are looking at each other with cautious curiosity. In the background: a landing module and a strange organic alien ship, a setting Martian sun, long shadows, dust in the air, cinematic epic lighting, 8k, photorealism.

Use the provided reference image for the {argument name="subject" default="young man"}'s face and appearance. Transform the scene so that he is smiling confidently while holding {argument name="cartoon character" default="Jerry, the playful cartoon mouse"}, gently in his arms. Behind him on the left, {argument name="companion character" default="Tom the cartoon cat"} gives a big thumbs-up with an excited expression. The background should be a cozy interior with cinematic lighting, a glowing chandelier, and soft ambient light. Blend live-action and 3D animation in a hyper-rea,

A full-body stylized 3D cartoon girl with {argument name="eye color" default="large sparkling emerald-green"} eyes, soft rosy cheeks, a gentle smile, and an innocent expression. She has {argument name="hair style" default="messy platinum-blonde hair tied in a voluminous bun"} with loose strands framing her face. She is sitting on a small stone stool with both hands supporting her cheeks in a cute, dreamy pose. She wears an {argument name="outfit" default="off-shoulder light gray top, loose white harem pants with a rustic fabric belt"}, and is barefoot with delicate ankle wraps. A small blue butterfly rests near her feet, with a white flower on the ground. Soft studio background, cinematic lighting, highly detailed stylized 3D rendering, painterly textures, smooth skin shading, whimsical fantasy atmosphere, premium animated feature film quality.

This girl is my original character, "{argument name="character_name_en" default="Anna"}." She is the female warrior holding a sword in the second image. Please create a 4-panel manga with an RPG game-style world. The panels are arranged vertically. The dialogue should be in Japanese.

4-panel manga title: Academic Silence (Shijima) University of Tokyo, Kyoto University 6 major universities MARCH NG Nittokoma-sen I don't even know the names of the universities anymore... lol

A cinematic film roll sequence capturing fragments of the character’s exact likeness life as {argument name="scene context" default="extreme scene's fearless stunt actor"}. Each frame reveals a different moment, or phase, evolving naturally without rigid narrative structure. Maintain exact proportions, eye shape, spacing, stylization, silhouette, and material identity from the reference. Do not humanize. Do not alter anatomy. Identity must remain perfectly consistent across all frames. Do not modify facial structure. Do not add a mouth or nose if the uploaded character does not have them. The film strip flows through the composition with subtle perspective depth, frames partially cropped, slightly misaligned, or overlapping to avoid perfect symmetry. Some frames feel sharp and present, others fade into blur or grain, creating temporal depth. Lighting varies per frame but remains cinematic and directional, forming a cohesive emotional rhythm rather than uniform exposure. Introduce natural film imperfections: grain, light leaks, slight motion blur, exposure shifts, and analog texture. Color grade is controlled and intentional, supporting the emotional tone of the chosen life theme without oversaturation. The scene must feel like a discovered memory archive rather than a clean layout. No flat composition, no centered symmetry, no full-frame sharpness. ar 16:9

A 1950s American diner scene, centered on a cheerful waitress in vintage uniform. She has softly curled dark hair styled in classic mid-century fashion, warm expressive eyes, and a natural smile with subtle freckles across her cheeks. She wears a {argument name="dress color" default="light blue"} short-sleeve dress with white trim, a fitted bodice, and a crisp white apron tied at the waist, complete with a name tag reading “{argument name="name" default="Betty"}.” A small white diner cap rests neatly on her head.

Currency issuance rights, my favorite (lol) ~Melody of infinite balance~ The “Truth” of NOBU, President of the Space Bank “The illusion of ‘trillions’ crushed by a tank” “The Hermit’s Smile: Japan’s Exemption Plan” 4-panel manga

Output image is 2x2 four-panel split Video Style: {argument name="video style" default="2000s Japanese Anime OVA style"} Background: Late-night city streets. Gaps between deserted roads and skyscrapers. Black liquid splattered from a giant monster adheres to the asphalt and walls. Dust and sparks are flying. Character 1: High school girl with long black hair. Wearing a sailor uniform. A Japanese sword is sheathed and hanging at her waist. She looks around cautiously, adopting a low stance. Character 2: High school girl with blonde bob hair. Wearing a blazer uniform. Standing behind her companion, holding one hand ready to use her ability at any moment. Her expression is tense. Enemy Presence: Black liquid gathers and transforms into a humanoid entity with a fibrous, thread-like texture. Its face is vague, its limbs are unnaturally long, and it crawls across the ground and walls with unnatural movements. Cut 1: Close-up of the ground. Filled with black liquid. Flying dust and sparks Cut 2: Medium shot. The humanoid anomaly, rising from the black liquid, awkwardly begins to stand up. Thread-like shadows droop, dragging on the ground. Cut 3: Low-angle wide shot. Multiple anomalies appear one after another from the roads and walls, encircling the girls in a circle. The cityscape emphasizes the feeling of being surrounded. Cut 4: Shot over the girls' backs. In front of them is a horde of black humanoids closing in. The black-haired girl rests her hand on the sheath, and the blonde-haired girl prepares to activate her ability. The composition emphasizes the tension right before the battle.

the Villain fairy performs the “you think because you say nothing You’re strong?”

Draw a four-panel manga in which a fairy comes down from the heavens to Earth. Because the first being it meets is a calico cat, it thinks this planet belongs to cats and talks to the cat.

TITLE: Premium {argument name="product" default="Frozen Pizza"} Product Commercial Storyboard FORMAT: • Single-page premium storyboard • 3:4 Portrait ratio • Cinematic frozen food advertising campaign • 8 completely different product-focused scenes • Product remains the main visual hero • Premium advertising agency presentation HEADER: • Bold editorial typography • Information cards: - Duration: {argument name="duration" default="20 Seconds"} - Style: Cinematic Food Commercial - Product: Frozen Pizza - Audio: Oven Crackle + Cheese ASMR • Why This Style Works section • Warm cream, tomato red and golden yellow aesthetic • Minimal pizza-themed decorative details STORYBOARD: 1. Frozen pizza box standing upright on a clean kitchen counter, front packaging perfectly visible 2. Pizza being removed from its packaging, showing detailed toppings and cheese 3. Pizza sliding onto a hot oven tray, dramatic close-up of the crust 4. Oven door closing as the pizza begins baking, warm orange oven glow 5. Extreme macro of cheese bubbling and melting across the toppings 6. Freshly baked pizza coming out of the oven with visible steam 7. Pizza cutter slicing through the crispy crust, melted cheese stretching between slices 8. Final hero product shot with the original pizza box beside the finished pizza EVERY PANEL: • Scene number • Duration badge • Camera direction • Visual • Action • Product detail CAMERA: Overhead food photography, extreme macro, slow-motion cheese melt, low-angle crust shot, controlled push-in, steam close-up, cinematic final packshot. STYLE: Ultra-realistic frozen food commercial, bubbling melted cheese, crispy golden crust, detailed toppings, realistic steam, premium packaging, warm cinematic lighting, shallow depth of field, professional food advertising, 8K.

I want to know the flow of GE's financial business from entry to withdrawal” → “Summarize it for a 4-panel manga” → 4-panel output instruction

A horizontal split-screen cinematic shot of {argument name="Scene" default="Times Square, New York"}, seamlessly blending two different eras: {argument name="Era A" default="1880s"} on the left and {argument name="Era B" default="present day"} on the right (default: about 100 years ago vs. present day). On the left side ({Era_A}): show era-appropriate architecture, interior or environment design, materials, vehicles, and props that clearly belong to that historical period. People wear authentic clothing from {Era_A}, including hairstyles, accessories, and typical items in their hands (such as books, umbrellas, instruments, letters, newspapers, etc.). The overall mood feels nostalgic and historically accurate. On the right side ({Era_B}): show the same {Scene} in the modern era, with updated architecture or renovated structures, contemporary materials (glass, steel, LED screens, modern furniture), modern vehicles or equipment, and current technology (smartphones, laptops, cameras, etc.). People wear contemporary fashion that matches today’s style in this setting. In the center: the two eras merge and overlap organically, without a hard dividing line. Elements from {Era_A} and {Era_B} visually interact: people from different times look at each other, walk through each other’s space, or seem surprised by the other era’s technology and objects. Architecture and environment smoothly morph from old to new (for example, stone gates turning into modern campus gates, classical concert hall décor fading into a futuristic stage, old street shops transforming into neon-lit storefronts). Make sure the scene is not just a simple left/right comparison but a dynamic time-travel interaction where buildings, clothing, props, and human gestures clearly emphasize the contrast and fusion between the two eras. Photorealistic, 8k resolution, cinematic lighting, wide angle, highly detailed textures, rich sense of time-travel storytelling. Aspect Ratio: 4:3

photorealistic top-down close-up view of a professional animation light table in a dimly lit studio, bright soft white glow coming from underneath through the glass surface, a crisp white sheet of animation paper centered on the light table, detailed pencil sketch of **{argument name="subject" default="a character portrait"}** drawn in clean confident animator lines on the paper, visible rough construction lines, light guidelines, subtle eraser marks and graphite texture, soft rim lighting highlighting the paper edges and pencil strokes, realistic light diffusion and subtle bloom from the table, warm wooden table frame, scattered animation pegs and a few loose pencils nearby, focused cinematic atmosphere, ultra detailed, high-end animation production still

{ "subject": "3x3 grid storyboard collage of a young woman sitting beside a large calm bear in a natural mountain forest setting", "clothing": "Traditional patterned clothing with beads and earth-tone fabrics, textured material with handcrafted details", "hair": "Long dark braided hair, neat braids resting over shoulders", "face": "Serious, calm, focused expression with natural skin texture, subtle emotion and intensity", "accessories": "Handcrafted bead details, natural fabric elements, no modern items", "environment": "Outdoor mountain landscape with trees, misty forest background, soft depth of field, atmospheric wilderness", "lighting": "Warm sunlight rays, gentle shadows, golden hour tones with soft natural highlights across face and fur", "camera": "Full body, medium shots, dramatic close-ups, over-the-shoulder, cinematic low angle, atmospheric wide establishing shots", "style": "Highly realistic, cinematic, film still aesthetic, warm tones, natural imperfections, detailed textures on fur, skin, and fabric" }

A multi-cut image of a Japanese woman doing yoga in a stylish room. Divide the frame into 5x5 sections and create 25 scenes. Display sequential numbers from 1 in the upper left of each scene. 25 scenes are depicted in the image above. The upper left is the scene number. Extract scene number {argument name="extracted scene number" default="7"} and reproduce it in high resolution.