AI 提示词库

{argument name="subject" default="Avengers: Doomsday"} but into a hyper-realistic glossy 3D CGI caricature masterpiece.

{argument name="style" default="Flat vector illustration"}, side profile of a {argument name="subject" default="Sa'idi Egyptian woman"}, featureless face, wearing a long black tarha (headscarf) decorated with flat silver Tally embroidery patterns, geometric triangles and stars rendered as flat metallic dots. She holds a stylized clay water jar (ballas) on her shoulder. Background: flat warm ochre desert with a distant pharaonic temple silhouette on the horizon. Palette: ochre, black, silver-cream, terracotta, deep red. Fusion of Upper Egypt folk and pharaonic memory.

A graceful {argument name="subject" default="Japanese woman"} in a traditional kimono kneels gracefully by a serene garden pond adorned with vibrant red autumn leaves, moss-covered rocks, and a classic stone lantern. In the first scene, she carefully arranges delicate floral stems in a ceramic vase, while the second captures a peaceful moment as she holds a small teacup near a traditional wooden pavilion and arched bridge, all beautifully rendered in a soft watercolor style with gentle mist and vivid reflections dancing on the water's surface.

Imagine what {argument name="subject" default="software engineers"} would do in {argument name="year" default="2050"} when all code is written by AI. {argument name="style" default="Pixel art"}.

the {argument name="subject part" default="dog's mouth"} is {argument name="state" default="closed"}.

Convert a user's short image request, attached image, or combination of text and images into one finished prompt for GPT Image Gen V2 or Nano Banana Pro. Apply the Redphase Survey visual system defined below. Do not explain the system, show analysis, or add commentary. Return only the finished image prompt in the selected model's format. Use the model named by the user or supplied by the host application. If neither names one, use GPT Image Gen V2. For GPT Image Gen V2, return natural language prose only. For Nano Banana Pro, return one valid JSON object only. When both are explicitly requested, return the GPT prompt first under the line "GPT IMAGE GEN V2", then the JSON under the line "NANO BANANA PRO". Add no other text. Write prompt content in the user's language unless another language is requested. Keep Nano Banana Pro keys in English. Preserve exact in-image text in its required script. CORE LOGIC Every image contains two simultaneous records. The survey record describes what occupies the scene. It uses observed geometry, dry line, hatching, reserved paper, and a restricted set of local colors. The red phase describes what is active through the scene. It uses one red spectral range, deep black, limited bloom, rapid falloff, and selective loss of ordinary color and detail. The survey record answers: What is present? The red phase answers: What is acting, passing, gathering, or changing? Do not average the two records into red-tinted illustration. Give each record a separate job. Before writing the final prompt, silently determine: 1. The literal subject and setting. 2. The stable material structure that belongs to the survey record. 3. One active condition that belongs to the red phase. 4. The source of that condition. 5. The path or relation through which it travels. 6. The receiving subject, object, region, or surface. 7. The visible consequence. 8. Two or three transition sites where the records meet. Choose the active condition from the user's request. It may be physical, social, temporal, or perceptual. Valid causes include heat, movement, wind, electric current, pressure, recent touch, reflected light, sound, gaze, attention, occupancy, growth, chemical change, distance, or another condition that can be shown through visible evidence. If the request names no active condition, use the strongest existing process already implied by the subject or action. Do not invent danger, desire, trauma, memory, menace, or supernatural activity without evidence. Red is a measurement channel, not a mood label. Do not equate red with violence, horror, romance, or urgency unless the user's content establishes that meaning. The active condition must have one source, one route or relation, and one consequence. Do not scatter red accents without a causal structure. CONTENT CONTROL Preserve the user's explicit subject, count, identity requirements, action or state, setting, era, clothing, objects,

A cinematic candlelit interior scene of a {argument name="subject" default="male figure"} seated at a small wooden table across from two other blurred figures, seen in the framing gap between their silhouetted heads and shoulders in the foreground. He sits directly facing the camera, gaze locked intensely on something or someone off-frame with a calm, cold, and calculating expression — the look of someone in the middle of a tense negotiation, secretive meeting, or confrontation. His hair is short, dark, neatly styled. He wears a {argument name="clothing" default="dark olive-charcoal crew-neck sweater"}, casual but composed. His arms rest on the wooden tabletop in front of him. A single lit candle sits in a small holder on the table between the figures — the flickering warm flame the primary light source of the entire scene, with a thin wisp of smoke curling upward from it. A folded paper or document is visible on the table near the candle, suggesting a covert exchange or deal being discussed. The two figures in the foreground on either side are cropped and out of focus — only the backs and sides of their heads and shoulders visible in silhouette, framing the central figure’s face between them. The background reveals a dimly lit bar, tavern, or restaurant interior — blurred wall-mounted picture frames, soft warm and cool bokeh window lights, and other blurred patrons visible in the deep background. The lighting is entirely warm and low — dominated by the candle’s flickering amber-gold glow illuminating the central figure’s face from below-front, casting soft warm highlights on his forehead, nose, cheekbones, and beard, while everything else recedes into rich brown-black shadow. Color palette: warm candlelight amber-gold, deep brown-black shadow, muted background bokeh. Photography style: cinematic film noir interior scene, candlelit dramatic lighting, {argument name="lens" default="50mm lens"}, shallow depth of field with foreground figures softly blurred, photorealistic, 8K. Use the provided face reference image for all facial details — bone structure, eyes, brow, nose, jawline, beard, and skin tone should be taken entirely from the reference photo. Apply only the calm calculating direct gaze and warm candlelit lighting on top of the reference face.

8-panel manga Title {argument name="title" default="The Agony of the Baccarat Venue"} Culprit "{argument name="dialogue" default="Go ahead and try it if you can"}" Afterward "{argument name="development" default="Whoa! Don't all rush in at once"}"

Charming stylized 3D animated {argument name="character" default="mother character"} in the {argument name="setting" default="kitchen"} with her baby in one arm while she talks on her {argument name="object" default="mobile phone"} with the other hand, the baby is trying to grab the mobile phone, large expressive brown eyes, soft rounded facial features, slightly exaggerated proportions, long dark frizzy morning hair hair, warm rosy cheeks, warm golden morning light, cinematic soft lighting, shallow depth of field, subsurface scattering skin, ultra-detailed 3D animated film render, Pixar-inspired style, 85mm lens, high resolution, highly detailed, soft bokeh background, volumetric light, hyperreal human, ultra realistic face, doll face, realistic 3d human portrait

Cinematic scene, heavy silhouetting, {argument name="lighting" default="daybreak light"}, {argument name="atmosphere" default="dark, foggy"}

Create a low-angle neo-noir sci-fi still, chest-up portrait of a person with a {argument name="gaze" default="slight upward gaze"}. Place the subject in the lower-right third, leaving vast negative space with about 80 percent empty sky above. The subject wears a {argument name="coat" default="matte shearling-collar heavy coat"} over a dark knit sweater in a desaturated olive and charcoal palette. The background is a {argument name="sky" default="blank, overcast sky"} with subtle haze. Use soft ambient daylight with negative fill on camera-left. Use image for face reference.

Transform the {argument name="input" default="uploaded portrait"} into a {argument name="style" default="whimsical handcrafted paper diorama illustration"} with a soft cute aesthetic. Reimagine the people using {argument name="elements" default="simplified layered paper shapes, rounded forms, and minimal facial details"} suitable for paper craft art. Preserve the affectionate pose and emotional warmth of the original image while giving everything a charming handmade appearance.

Ultra-detailed {argument name="style" default="neo-editorial cinematic portrait illustration"} rendered in a hybrid semi-realistic painterly-engineering style, combining refined anatomical realism with expressive sketch abstraction and futuristic conceptual drafting. The artwork fuses elegant graphite fashion illustration, architectural blueprint aesthetics, and richly layered digital painting into a visually immersive mixed-media composition. Human anatomy is highly accurate and sculptural, featuring luminous skin gradients, tactile pores, subtle wrinkles, refined facial contours, and soft glossy highlights. Facial rendering is extremely expressive and emotionally cinematic, with sharp eyes, delicately shaded brows, natural lips, and dramatic warm rim lighting blending golden-orange glow with cool bluish atmospheric reflections. Smooth polished rendering transitions coexist with rough textured brushwork, etched shading, and visible painterly grain. Fine ultra-thin micro-line drafting overlays merge seamlessly with loose hand-drawn sketch strokes, cross-hatching, technical contour lines, and schematic construction marks. Portions of the composition intentionally retain unfinished artistic energy: faded edges, dissolving contours, partially erased strokes, and fragmented geometric segmentation inspired by cubist blueprint structures. Certain body and fabric areas transform into layered painterly abstraction with watercolor diffusion, acrylic textures, expressive oil-like sweeps, and deconstructed mixed-media detailing. Lighting is highly cinematic and directional, creating deep volumetric shadows and glowing highlights with dramatic editorial contrast. The composition balances softness and structure—warm sepia shadows, sculpted highlights, and atmospheric diffusion enhance emotional depth while maintaining precision-engineered visual control. Background features a fusion of textured parchment surfaces and futuristic technical collage aesthetics: faint diagrams, circles, architectural annotations, handwritten marks, layered scratches, drafting elements, grain textures, faded paper imperfections, and atmospheric painterly haze. The environment feels both vintage and futuristic, combining analog artistic charm with advanced conceptual design language. Color palette blends earthy sophistication with vibrant cinematic accents: {argument name="colors" default="warm beige, creamy ivory, ochre, burnt orange, golden yellow, sepia brown, muted charcoal, deep bluish-green, cobalt blue, and subtle black-red highlights"}. Tonal harmony emphasizes warm desaturated realism contrasted by luminous energetic accent colors. Overall aesthetic is intimate, melancholic, elegant, high-fashion, cinematic, and emotionally powerful—an artistic fusion of vintage sketchbook illustration, futuristic engineering draftsmanship, cubist fragmentation, expressive oil painting, and contemporary editorial digital realism. No text, no logo and no watermark. R 9:16.

{argument name="format" default="4-panel manga"} {argument name="theme" default="vision test using symbols"} Eye exam using {argument name="brand name" default="Chanel"}'s Azura Azura!

{argument name="scene" default="In the middle of a scramble with older women for discounted side dishes at a supermarket"}

8-panel manga A retro game called 'The World' {argument name="dialogue" default="Yikes! This is a total ultra-super-special crap game!"} I pressed the reset button.

{argument name="style" default="Ultra-realistic fine art portrait illustration"} with accurate human anatomy and natural facial proportions. Realistic cranial structure, balanced forehead dimensions, anatomically correct neck length, natural shoulder width, and authentic facial construction. Facial features rendered with lifelike proportions, including naturally positioned eyes, realistic nose structure, balanced cheekbones, accurate jawline definition, and subtle facial asymmetry found in real human faces. Highly detailed mixed-media illustration combining refined pen-and-ink craftsmanship, observational drawing techniques, delicate watercolor rendering, and realistic painterly textures. Dense yet controlled cross-hatching, contour shading, and subtle linework enhance skin texture, pores, wrinkles, facial hair, and natural anatomical details without exaggeration. Hair rendered with thousands of individual strands, realistic volume, organic flow, and natural movement. Editorial portrait realism blended with contemporary fine-art illustration. Strong emphasis on believable anatomy, authentic facial expression, natural posture, and accurate light interaction across skin and hair. Rich observational detail, realistic depth, subtle imperfections, and sophisticated rendering quality. {argument name="palette" default="Warm earth-tone palette"}, aged sketchbook paper texture, restrained watercolor washes, delicate ink splatters, unfinished artistic edges, museum-quality portrait illustration, highly refined linework, and premium sketchbook presentation. Professional fine-art portraiture with realistic proportions, natural character presence, and lifelike visual authenticity. No TYPOGRAPHY, no initial, no brand and no watermark.

Cutaway diagram of a {argument name="object" default="mechanical device"}, detailed with interior component, on a white background. High-resolution, with a focused technique and text callouts providing information about each part. Interior details, hyper-realistic appearance.

turn this into an {argument name="type" default="infographic"}

Whimsical {argument name="setting" default="miniature world"} depiction of a {argument name="subject" default="floating ritual island chain"} in a {argument name="style" default="mythpunk"} style

Character sheet, multiple angles of a sci-fi tactical soldier in full body {argument name="armor color" default="matte-black"} armor. He wears a faceless helmet with a thin, dim {argument name="visor color" default="white"} horizontal glowing visor line. Gritty, brutalist design, heavy Kevlar and carbon fiber textures. Clean studio lighting, white background, ultra realist, front view, side view. --ar 16:9

Vercel {argument name="subject" default="container ship"} 8-bit

Secrets of the {argument name="location" default="Sea"}

A ultra-detailed, hyper-realistic 3D rendering of {argument name="subject" default="human waist"}, presented against a {argument name="background" default="smooth, clean white-to-light-grey gradient background"}. The style must mimic that of an advanced, glowing medical scan (like an fMRI combined with a CT). The entire structure is composed of luminous, semi-transparent green tissue (fascia, muscles, nerves), with skeletal elements and major neural pathways rendered in a {argument name="glow color" default="brilliant, glowing electric green light"}. The texture is that of a complex, interwoven network of lit fibers and nodes. The aesthetic is clean, technological, and medicinal, with the green structures appearing to emit light. The composition must be high-focus and minimalist, with a single, clear subject and no text.