AI Image Prompts · 绘画提示词库
Source & License
This site lists project metadata and a short summary only. We do not republish the author’s original content — follow the source link above for the full text. Verify the LICENSE and NOTICE in the original repository before commercial use.
Show full license text
MIT License Copyright (c) 2026 YouMind-OpenLab Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
LICENSE · 2026-09-26 23:14:14
Introduction
Recommend curated prompts from a 10,000+ real-world image generation prompt library. Works with ANY AI image model — Nano Banana Pro, Nano Banana 2, Seedream 5.0, GPT Image 1.5, Midjourney, DALL-E 3, Flux, Stable Diffusion, and more. Use this skill when users want to: - Find proven image generation prompts (any model) - Get prompt inspiration for portraits, products, social media, posters, etc. - Create illustrations for articles, videos, podcasts, or marketing content - Browse categorized prompt templates with sample images - Translate and understand prompt techniques
Resource Information
License information is collected from official sources and may be outdated or missing. Refer to the LICENSE file in the original repository before commercial use.
Installation
npx skills i YouMind-OpenLab/ai-image-prompts-skill
Features
- 按用途分成 11 类,不是随机堆砌
- 全文已入库,站内直接检索、可复制
- 语义检索(描述需求即可)
- 内容改写模式(贴正文生成定制提示词)
- 数据每日同步两次
- MIT 许可,无需 API Key
Prompt Library · 22.8K
Source: community prompt library (MIT, synced daily)15141 distinct prompts across 11 use cases; one prompt may serve several, so category counts add up to more than the deduplicated total.
725 matches
Ultra-Realistic 3D Inflatable Object Render
game-assetA prompt designed to generate ultra-realistic 3D renders of an object (Valentine themed in this case) with a glossy, inflated plastic material, focusing on smooth contours, high specular highlights, and a designer collectible toy aesthetic against a plain studio background.
Show full prompt
"{argument name="subject" default="SUBJECT"}, ultra realistic 3D render, smooth inflated glossy plastic material, soft pillowy balloon like contours, subtle stitched panel seams and paneling, high specular highlights, clean reflections, soft contact shadows, designer collectible toy aesthetic, cute proportions, premium polished finish, floating centered, plain light gray studio background, soft diffused studio lighting, sharp focus, high resolution, 8k, 1:1, no text, no logo, no watermark.”
Fisheye Product Photo with Forced Perspective
game-assetA prompt designed to create a high-impact, cartoonish yet realistic product photo using a fisheye lens effect, emphasizing warped forced perspective where the character's head is immense in the foreground and limbs taper into blur.
Show full prompt
A high impact fisheye product photo of the character, shown full-body in warped forced perspective. The head is immense in the foreground, limbs taper into blur, cartoonish yet grounded in realistic detail. Pose: The character leans slightly forward, stepping into the viewer’s space. The oversized head pushes closest to the lens, eyes sharp and engaging, while the arms hang loosely at its sides. Shoes recede into deep blur, vanishing into the background. Camera: Fisheye macro, 15–20mm lens, tilted slightly upward to exaggerate the forward lean. Lighting: Strong frontal light reveals fine surface textures, with soft fill shadow along the torso and a faint rim halo separating the figure from the backdrop. Eyes: Eyes are emphasized with dimensional lighting subtle gradients across the surface, crisp specular highlights, and surrounding contrast shadows to add depth and expression without altering their shape. Background: Neutral infinite fade in soft grey/white, with no visible floor or horizon. Material Focus: Extreme emphasis on surface qualities textures, finishes, and subtle reflections are captured with high fidelity, whether matte, soft, or powdery. Gloves appear simple and soft, shoes trail into blur with faint sheen. Emotion: Direct and intimate the figure feels like it’s leaning out of the photo toward you, demanding presence. Negative prompt: No props, no clutter, no flat neutral standing pose
Cinematic Action Video Prompt: Banana Fairy vs. Zombies
game-assetA detailed video generation prompt for Vidu AI, creating a cinematic action scene where a banana fairy girl fights off zombies using only bananas as weapons in a fantastical indoor space filled with banana decorations. The prompt includes specific instructions for character, setting, action, camera work, lighting, and style.
Show full prompt
Use the woman in the attached image as the main character. The setting is a fantastical indoor space surrounded by bananas, identical to the attached image. Zombies attack one after another in a space lined with banana trees and giant banana objects. The banana fairy girl quickly grabs bananas around her and throws them in rapid succession. The only weapon is bananas. The bananas fly straight while spinning at high speed, hitting the zombies' heads. The moment of impact is in slow motion. Strong shockwave effects, light trails, dynamic motion blur. The hit zombies are blown away, collapsing while shaking the surrounding banana decorations. The girl jumps onto a giant banana and throws bananas accurately while spinning. Emphasize the impact with a low-angle hero shot. Dynamic tracking camera following the flying bananas. Camera Direction: ・High-speed dolly-in towards the girl ・Intense handheld-style shake when zombies approach ・Slow motion upon banana impact ・Smooth tracking shot following the flying bananas ・Tense low angle Lighting: Strong contrast Rim light emphasizing contours Contrast with the fantastical gold interior light Style: Anime Cinematic Action High definition Smooth animation Strong motion blur Dynamic composition Depth of field Atmosphere: Tension Thrilling Heroic No text No subtitles
Cinematic Portrait of Sydney Sweeney as Wanda Maximoff
game-assetA highly detailed JSON prompt for generating a cinematic, realistic portrait of actress Sydney Sweeney as Wanda Maximoff (Scarlet Witch). The prompt specifies a dramatic visual tone, deep crimson costume details (satin dress, angular crown), and focuses heavily on facial expression (determined, conflicted) and realistic skin texture.
Show full prompt
{
"style": {
"genre": "cinematic realistic portrait",
"visual_tone": "dramatic, intense, slightly surreal realism",
"color_palette": "deep reds, warm skin tones, muted background colors",
"overall_mood": "controlled chaos, emotional tension, restrained power"
},
"character": {
"subject_reference": "Sydney Sweeney",
"gender_presentation": "female",
"age_range": "mid 20s to early 30s",
"expression": "lips pressed together, slightly pursed, holding back emotion",
"emotion": "determined, conflicted, restrained anger",
"gaze": {
"direction": "directly into camera",
"intensity": "wide-eyed, intense, unblinking",
"feeling": "alert, emotionally charged, on edge"
}
},
"face": {
"face_shape": "soft oval with defined cheekbones",
"skin": {
"texture": "natural skin texture, subtle pores visible",
"finish": "soft matte with slight natural sheen"
},
"eyes": {
"size": "large",
"shape": "round with sharp focus",
"eye_color": "light green to hazel",
"eyelashes": "defined but not exaggerated"
},
"eyebrows": {
"shape": "natural, slightly arched",
"thickness": "medium, realistic"
},
"lips": {
"shape": "medium fullness",
"expression_detail": "pressed inward, tension visible",
"lip_color": "natural pink-red"
}
},
"makeup": {
"foundation": "light coverage, natural skin visible",
"eyes": {
"eyeshadow": "neutral warm tones, very subtle",
"eyeliner": "thin, close to lash line",
"mascara": "natural definition"
},
"cheeks": "very subtle blush",
"lips": "soft red or rose tint, satin finish"
},
"hair": {
"color": "warm light brown to dark blonde",
"length": "shoulder length",
"style": "loose, softly waved",
"parting": "slightly off-center",
"movement": "natural, slight motion as if caught mid-step"
},
"costume": {
"outfit_type": "superhero-inspired elegant costume",
"primary_color": "deep crimson red",
"fabric": "smooth satin-like material with soft sheen",
"dress": {
"cut": "sleeveless, fitted bodice",
"neckline": "structured with ribbon tie at collarbone",
"fit": "form-fitting but elegant"
},
"gloves": {
"length": "elbow-length",
"material": "matching satin fabric",
"color": "same deep red"
}
},
"accessories": {
"headpiece": {
"type": "stylized crown / headband",
"color": "dark red",
"shape": "sharp angular points rising upward",
"material": "matte or semi-gloss hard surface"
},
"jewelry": "none visible"
},
"pose": {
"body_position": "standing upright",
"hands": "hands close together near chest level",
"gesture": "subtle tension in fingers"
},
"framing": {
"shot_type": "medium close-up portrait",
"crop": "head to mid-torso",
Cinematic Concept Art of Sadie Sink as Jean Grey
game-assetA structured JSON prompt for generating a cinematic 3D render or concept art of actress Sadie Sink portraying the Marvel character Jean Grey, detailing three distinct costume variants, specific physical features, and technical rendering parameters like 4K resolution and Unreal Engine 5 style.
Show full prompt
{
"subject": {
"actor_likeness": "Sadie Sink",
"character": "Jean Grey (Marvel Comics)",
"physical_features": {
"hair": "Long, voluminous, wavy ginger-red hair, natural texture",
"eyes": "Pale blue/grey",
"expression": "Stoic, determined, heroic neutral",
"build": "Athletic, slender"
}
},
"costume_variants": {
"variant_1_phoenix": {
"base_color": "Metallic emerald green",
"accents": "Gold chrome gloves, boots, and waist sash",
"insignia": "Large white Phoenix force emblem on chest",
"style": "Sleek spandex-metallic hybrid"
},
"variant_2_classic_xmen": {
"base_color": "Dark forest green leather",
"accents": "Yellow 'X' symbols on belt and boot straps",
"armor": "Metallic green sculpted breastplate and abdominal plating",
"headwear": "V-shaped green metallic tiara/circlet"
},
"variant_3_tactical_armor": {
"base_color": "Deep crimson/burgundy textured fabric",
"armor_plating": "Iridescent cobalt blue metallic plating on chest, shoulders, forearms, and shins",
"insignia": "Red and black 'X' logo on belt",
"headwear": "Integrated blue metallic combat headset/tiara"
}
},
"artistic_style": {
"type": "Cinematic 3D render / Concept art",
"lighting": "Studio rim lighting, volumetric glow, subtle golden aura or smoke",
"background": "Solid studio backdrop (Slate blue or Charcoal) with atmospheric haze",
"framing": "Full-body shot, centered composition, eye-level camera angle",
"texture_quality": "High-fidelity fabric weave, metallic reflections, realistic skin pores"
},
"technical_parameters": {
"aspect_ratio": "2:3",
"resolution": "4K",
"rendering_engine": "Unreal Engine 5 / Octane Render style"
}
}
Whimsical 3D Clay Sculpture of Stacked Children Prompt
game-assetA detailed prompt for generating a highly detailed 3D cartoon sculpture of three whimsical children stacked on top of each other on a rustic house, emphasizing exaggerated features, expressive poses, and realistic clay texture.
Show full prompt
A highly detailed 3D cartoon sculpture of three whimsical children stacked on top of each other on a small rustic house. The children have exaggerated features: large round eyes, big ears, small noses, and rosy cheeks. The top child has wild, messy hair, wearing a blue checkered dress, looking excited with mouth open showing teeth. The middle child has short hair, a brown sweater, gray pants, a gray cap, looking surprised. The bottom child has wavy hair, wearing a muted patterned dress, looking worried and nervous. Each child has oversized shoes and expressive poses. The tiny house at the bottom is textured with stone walls, wooden windows, and a blue door, slightly crooked. Realistic shading and textures, ultra-detailed, cinematic lighting, depth of field, hyper-realistic clay sculpture style. Optional style tags: hyper-detailed, cinematic lighting, whimsical, clay sculpture, 3D cartoon, expressive characters, ultra-realistic textures, soft shadows
Caricature Bobblehead Figurine Prompt
game-assetA detailed prompt for generating a highly detailed caricature figurine in a bobblehead style, emphasizing exaggerated features, intense expression, and professional product photography aesthetics.
Show full prompt
Highly detailed caricature figurine of {argument name="subject" default="[SUBJECT]"} as a cute but intense bobblehead-style collectible statue, exaggerated oversized head with hyper-realistic facial features, deep wrinkles, fierce intense expression, slightly open mouth showing teeth, sitting relaxed in a vintage armchair, wearing {argument name="outfit" default="[OUTFIT]"}, dramatic colorful studio lighting, {argument name="background" default="[BACKGROUND]"}, modern collectible toy photography style, sharp focus, cinematic depth of field, extremely intricate sculpting details, realistic skin texture, professional product photography, 8k quality
Maze Generation Failure in Nano Banana
game-assetThe user attempted to generate a maze using Nano Banana, concluding that the model's underlying mechanism prevents it from successfully creating functional mazes.
Show full prompt
I tried generating a maze, but I see that this kind of thing isn't possible due to the mechanism...
Character Generation from Google Slides Shapes
game-assetThis tweet expresses surprise at the high quality of a cute character generated by Nano Banana Pro using simple shapes from Google Slides as input, implying an image-to-image or style transfer process.
Show full prompt
Such a cute character was created from Google Slides shapes!!
Nano Banana's Understanding of Fallout Terminology
game-assetThis tweet notes that Nano Banana (likely Pro) has a high understanding of specific proper nouns and terminology from the Fallout universe, such as 'Nuka-Launcher' and 'Ghoul,' suggesting these terms were used in a prompt.
Show full prompt
Proper nouns like Nuka-Launcher and Ghoul are understood.
Billie Eilish Fortnite-Style 3D Character Render
game-assetA structured JSON prompt for generating a high-fidelity 3D character render of Billie Eilish in a Fortnite/Unreal Engine aesthetic. The prompt specifies her iconic neon green and black outfit (oversized t-shirt and shorts with graffiti print), space buns, dynamic mid-air pose, and dramatic volumetric neon green stage lighting, aiming for a stylized, action-oriented look.
Show full prompt
{
"subject": {
"name": "Billie Eilish",
"style": "Fortnite stylized 3D character model",
"pose": "Dynamic mid-air jump, forward-leaning, action-oriented",
"features": {
"hair": "Two high space buns, black hair with vibrant neon green roots/highlights",
"eyes": "Light blue/grey, calm and focused expression",
"skin": "Smooth, stylized porcelain complexion"
}
},
"outfit": {
"top": "Oversized neon lime green short-sleeved t-shirt",
"bottom": "Matching neon lime green oversized shorts",
"pattern": "All-over black graffiti print featuring 'Billie Eilish' text and 'Blohsh' stick-figure logos",
"accessories": [
"Stacked chunky silver chrome chain necklaces",
"Silver Blohsh pendant"
]
},
"footwear": {
"shoes": "Monochromatic neon green high-top sneakers (Air Jordan 1 style)",
"socks": "Mid-calf neon green ribbed socks"
},
"environment": {
"setting": "Futuristic concert stage / Festival mainstage",
"lighting": "Dramatic neon green volumetric lighting, laser beams, glowing floor panels",
"atmosphere": "Vibrant green haze and digital energy effects",
"background_elements": "Silhouetted guitarist on a raised platform in the distance"
},
"technical_specs": {
"art_style": "High-fidelity 3D render, Unreal Engine 5 aesthetic",
"color_palette": ["#39FF14 (Neon Green)", "#000000 (Black)", "#C0C0C0 (Silver)"],
"composition": "Low angle, vertical aspect ratio, dynamic motion blur"
}
}
1/7 Scale Collectible Figure Product Shot
game-assetA prompt for generating a high-resolution commercial product photograph of a detailed 1/7 scale collectible figure, emphasizing realistic textures, specific base details, and a background featuring a computer monitor displaying the figure's 3D wireframe mesh and collector's box.
Show full prompt
A high-resolution commercial product photograph of an incredibly detailed {argument name="scale" default="1/7"} scale collectible figure. The figure features intricate clothing textures and realistic skin, standing on a pristine, transparent circular acrylic base on a modern artist's desk surface. Eye-level close-up. In the soft-focus background, a computer monitor displays the complex 3D wireframe mesh of the same figure, next to a premium matte collector's box with vibrant 2D character art. Lighting is a mix of professional soft studio light and the subtle blue glow from the monitor screen creating rim lighting. Highly detailed, photorealistic, shallow depth of field.
Prompts are model-agnostic and work with Midjourney / DALL-E / Flux / Stable Diffusion.
Dependencies
- Node.js 18+ - 无需任何 API Key(数据全部公开) - 提示词已完整同步到本地与数据库,站内检索不依赖第三方
Supported platforms
Changelog
Watching official sources内容有更新(细节请以官方文档为准)
内容有更新(细节请以官方文档为准)
官方源有内容更新(首次纳入监控)
Information boundary
Information on this page is collected from official documents for technical presentation only and does not imply our endorsement. Pricing, quotas and availability are subject to the latest official statements. We make no commitment regarding any action you take after leaving this site.
Similar skills
视频短视频图片音乐音频配音口播数字人全能创作
Marked paid通过同一个AI创作工具完成AI图片、视频、音乐与语音创作,查找公开社交数据,编辑视觉成果,并集中查看和管理生成资产。Beatra 全能创作支持AI图片生成(文生图)、AI视频生成(图生视频)和AI视频编辑,通过AI音乐生成把想法做成歌曲或纯音乐,也可用AI语音生成(文字转语音)、多语言旁白和AI语音克隆完成跨媒介内容;还能查找抖音数据、小红书数据,以及 YouTube、TikTok、Instagram 与 X 上的公开帖子、笔记、评论、账号和趋势,把社交媒体数据用于竞品调研和创作。它是一款面向AI内容创作与多模态内容创作的AI创作套件,用起来就是一个把图片、视频、音乐与语音收在同一处的AI创作助手,适用于社交内容、产品展示、营销素材、课程、播客、短片和跨媒介项目;还能在不同媒介间复用原始素材、已有成果和查到的公开资料,查看制作进度、返回文件、实际用量和积分记录。
AI数字人口播与虚拟主播视频
Marked paid用一张人像和一段短脚本或语音,制作有明确讲述内容的数字人口播视频。可使用现成录音,也可以先选择音色,把脚本生成旁白,让画面中的人物随声音完成稳定、自然的口播。适合 AI 主播、虚拟主播视频、数字人讲解、让照片说话、照片说话视频、产品讲解视频、数字人培训视频、公告和社交口播,并在交付后检查人物一致性、声音清晰度、口型同步和动作稳定性。
架构图一键生成
Marked paidAPI key required使用 ContextWeave 生成或修改架构图、流程图、思维导图和复杂信息图。适用于需要从代码、文件或自然语言中提取结构与关系并实际产出 CW、SVG、HTML、可编辑 PPTX 或原生 VSDX 的请求;未特别说明时,PPTX 与 Visio 使用原生形状和连接器导出。不适用于统计图表、手绘插画或像素级排版。
罗斯蒙德可视化图形生成
Not specified【罗斯蒙德可视化图形生成】南京罗斯蒙德基于已交付项目沉淀的可视化图形生成场景:根据业务描述或系统需求,自动生成可编辑的架构图、流程图、ER 图、时序图、状态图与 C4 图。采用「统一图模型」一次建模、多格式输出:结构化文本(Mermaid / PlantUML,可嵌入 Markdown、进 Git 版本管理)、可编辑图形文件(drawio / Excalidraw,可拖拽修改节点与样式)、静态图(SVG 自渲染,PNG 给转换命令),并输出图形说明(图的含义 + 关键节点解释)。内置样式系统统一配色与命名,内置布局引擎自动排布,校验脚本拦截孤立节点、悬挂连线与语义缺漏。触发词:画架构图、画流程图、画 ER 图、画时序图、画状态图、C4 图、系统架构图、部署架构图、数据流程图、实体关系图、生成 drawio、生成 Excalidraw、导出 mermaid、导出 PlantUML、导出 SVG、导出 PNG、可编辑的图、图转可编辑、架构图可编辑、流程图二次修改、画图、出图、画个图。
海报设计工作室
FreemiumAPI key required将一段活动描述、一张产品照片或品牌参考图,转化为抓人眼球的活动海报、宣传 banner 或社交媒体配图。这款 AI海报设计与制作工具生成图片丰富的海报,搭配清晰视觉层级、按品类匹配的风格和预留文字区域,覆盖活动、音乐节、电影、产品发布和促销场景。你可以把它当作 AI宣传单制作工具来制作促销宣传单,当作活动海报设计工具,当作电影海报生成器,当作音乐节海报设计工具,当作产品发布会海报的 banner设计 AI,或当作平面设计海报工作室。从活动主题出发,用品牌风格参考图变换产品或场景照片,或在已认可草稿上做定向精修,完成可印刷或可分享的成品。
架构图一键生成
Marked paidAPI key required使用 ContextWeave 生成或修改架构图、流程图、思维导图和复杂信息图。适用于需要从代码、文件或自然语言中提取结构与关系并实际产出 CW、SVG、HTML、可编辑 PPTX 或原生 VSDX 的请求;未特别说明时,PPTX 与 Visio 使用原生形状和连接器导出。不适用于统计图表、手绘插画或像素级排版。
What is a Skill
A Skill is a packaged single capability, such as text recognition, document summarization, or keyword extraction. Developers wrap the calling logic and prompts into a plugin-like unit, so users can load it without understanding the underlying model. StackPilot indexes a large number of skills that can be tried instantly or assembled into a plan.