← 返回首页
技能

AI Image Prompts · 绘画提示词库

github · 1.0.4 · 下载量 1,092 · 热度分 90.0

出处与授权

作者: YouMind-OpenLab
许可证: MIT 许可证 (原文)
版权声明: Copyright (c) 2026 YouMind-OpenLab

本站只收录项目元数据与简介,不对作者正文做转载;完整内容请到上方出处查看。商用前请自行核对原仓库的 LICENSE 与 NOTICE。

展开许可证全文
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

产品介绍

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

资源信息

作者:YouMind-OpenLab
分类:图文音视频多模态生成
语言:en
版本:1.0.4
认证方式: 无需认证
许可证:MIT
是否付费: 官方标注免费
发布方:YouMind

许可证信息摘自官方源,可能过时或缺失。商用前请以原仓库的 LICENSE 文件为准。

标签
绘画提示词文生图营销素材

安装方式

npx skills i YouMind-OpenLab/ai-image-prompts-skill

功能说明

  • 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
适用场景
找某类图片的现成提示词给文章/视频/播客配插画做海报 / 封面 / 社媒图翻译并理解提示词写法

提示词库 · 2.3万

数据源:社区提示词库(MIT 许可,每日同步)

共 15141 条提示词,按 11 个用途归类;同一条可归入多个用途,因此分类条数相加会大于去重后的总数。

命中 224 条

Cinematic Venom Symbiote Action Scene

youtube-thumbnail

A prompt for generating an ultra-realistic cinematic movie scene featuring a young man (with identity locked to a reference photo) in a dynamic action pose, with a Venom-like symbiote bursting from his back. The prompt specifies the setting (dark industrial warehouse), lighting (sharp light beams, warm highlights), and camera angle (low and tilted) for a Hollywood action-poster vibe.

展开提示词全文
16K ultra-realistic cinematic movie scene of the same young man (keep his real facial features and proportions exactly as in the reference photo, no changes to his face). He is in a dynamic action pose, slightly crouched forward as if ready to leap or attack, one knee bent, one hand pressing firmly on the ground, the other reaching out with intense motion. From his back, a Venom-like symbiote bursts outward, forming huge menacing tendrils and a snarling monstrous face rising above his shoulder. The dark industrial warehouse setting is filled with mist and drifting smoke, with sharp light beams slicing through the fog. Camera angle is low and slightly tilted for dramatic tension and high-energy impact. Add warm highlights, motion blur, lens flares, and subtle film-grain for a Hollywood action-poster vibe. Generate the face using the uploaded reference image with accurate likeness.

Cinematic 3D Action Advertisement Prompt

youtube-thumbnail

A prompt for generating a cinematic, action-packed 3D advertisement. It emphasizes dramatic studio lighting, dynamic particle effects, high-impact slow-motion energy, and ultra-hyperrealistic rendering, with instructions to seamlessly integrate the brand logo and a modern slogan.

展开提示词全文
Cinematic 3D action-packed advertisement for {argument name="product brand" default="[INSERT PRODUCT/BRAND HERE]"}, captured in an intense mid-motion moment with dramatic studio lighting, dynamic particle effects, and high-impact slow-motion energy. Ultra-hyperrealistic rendering, razor-sharp details, glossy commercial finish, atmospheric depth, and powerful contrast. Viral-ready composition with the brand logo seamlessly integrated into the scene and a sleek, modern slogan positioned cleanly beneath. High-end blockbuster commercial aesthetic

Kling 3.0 video using Nano Banana Pro images

youtube-thumbnail

A video created using Kling 3.0 Omni, which sequences four images generated by Nano Banana Pro based on a shared prompt (developed by GPT-5). The prompt focuses on a cinematic, four-cut sequence depicting a woman's charm through casual conversation, using ASMR and binaural audio.

展开提示词全文
A cinematic video composed of 4 cuts, depicting a woman's charm through casual conversation with friends. Seamlessly connect a medium shot of a smile, ASMR sounds of hands, a close-up of a whisper, and a lingering pull-back shot. Combine film-like texture and binaural audio to enhance immersion from both visual and auditory perspectives.

Detailed Japanese Prompt for Variety Show Style Image

youtube-thumbnail

A highly detailed, multi-section prompt written entirely in Japanese for generating an image in the style of a TV variety show, focusing on a funny cat mishap. The prompt specifies text content, color schemes (using hex codes), fonts (variety show style, friendly rounded text), layout, subject (cat failing a jump, making a funny face), and overall style (comedy, decisive moment, smartphone photo look).

展开提示词全文
Please output 4 times using the following prompt:

Text
Main: {argument name="main text" default="Miracle at Broadcast Accident Level"}
Sub: {argument name="sub text" default="Funny Cat Mishap Collection"} (bottom banner)
Other: Why did that happen w (speech bubble), Decisive Moment (top right stamp)

Color Scheme
Background: General living room (with signs of life, no specific brand furniture).
Text: Pop colors like #FFA500 (orange) and #FFFF00 (yellow).
Accent: Colors for comic-style concentration lines.

Font
Main: Fun, extra-bold font typical of variety shows.
Sub: Friendly, rounded text.

Layout
Text Placement:
Large main title at the top. Add movement to the text (italics, curve).
Place 'w' or speech bubbles near the cat's face.
Decoration/Figures:
Main subject is a cat (or dog).
Moments like failing a jump and being in a strange posture, funny faces, or flipping things over.
Motion blur to emphasize movement (subject blur) and comic sound effect effects (Don!, Biyooon, etc.).
Show the messy room without hiding it.

Style
Pet video, mishap, comedy, decisive moment, sense of dynamism, friendly, smartphone photo style

Historical Social Media Feed Generation

youtube-thumbnail

A prompt testing Nano Banana 2's ability to research historical events from a specific date (December 9, 1971), find images of the key figures, and generate a vertical format image simulating an X (Twitter) feed where those figures post in Spanish on that day.

展开提示词全文
I was born on December 9, 1971. Investigate what happened in the world that day, search for images of the protagonists of those news stories, and create an image of an X (Twitter) feed as if those characters had posted that day. Vertical format, all writing in Spanish.

Cinematic 4-Cut Video Prompt of Woman Painting

youtube-thumbnail

A cinematic video prompt, generated by GPT-5 and used for Kling 3.0 Omni, describing a four-cut sequence of a cute woman painting a wall. The prompt specifies camera shots (extreme close-up, medium, artistic cut, wide shot), soft pastel tones, and shallow depth of field to create a warm atmosphere.

展开提示词全文
A cinematic video composed of four cuts, themed around a cute woman painting a wall. Smoothly transitioning to an extreme close-up of scooping paint onto a brush, a medium shot of her happily painting the wall, an artistic cut capturing dripping paint, and a wide shot of her smiling contentedly in front of the finished wall. The overall look uses soft pastel tones and shallow depth of field to create a unified, warm atmosphere.

Interstellar-Inspired Cinematic Film Stills Prompt

youtube-thumbnail

A simple prompt requesting the generation of cinematic, ultra-realistic film stills inspired by the movie Interstellar, emphasizing consistent color grading and outputting an aesthetic storyboard.

展开提示词全文
Create cinematic, ultra-realistic film stills inspired by Interstellar (Christopher Nolan vibe).
Look like real movie screenshots with consistent color grading, 
Output as an aesthetic storyboard:  visually cohesive, no text

Spy Thriller Comedy Storyboard Prompt

youtube-thumbnail

A narrative prompt designed to generate a multi-panel storyboard for a comedic spy thriller scene involving tactical agents staking out a suburban dad who is just looking for milk in his fridge.

展开提示词全文
Early morning. Total silence.

A hyper-serious spy thriller vibe.

A sleek black SUV idles outside a quiet suburban house. Two “tactical” men in earpieces whisper intensely.

“Target is inside.”
“Movement in the kitchen.”
“Stand by.”
Inside the house—
A dad in sweatpants stands in front of the fridge, staring into it like it contains classified secrets.

Cut back outside.

“Subject has opened the cold storage unit.”
“Thermal spike detected.”

Inside, he pulls out milk. Shakes it. Sniffs it. Makes a face.

Outside, one agent squints at a tablet.
“He’s analyzing chemical integrity.”
The front door creaks open.

The dad steps onto the porch in mismatched slippers to grab the newspaper.

Both agents snap alert.
“Target exiting the compound.”
“Visual confirmed. He looks… unstable.”
The dad waves awkwardly at them.

They freeze.

He squints. “You guys DoorDash?”

Beat.
They look at each other.
Inside the SUV, a third “commander” barks through the radio:
“Abort. Wrong house. Again.”

The agents slowly lower their sunglasses.

The dad shrugs, walks back inside.

Cut to kitchen.

He sits at the table with his cereal. Turns on the TV.

BREAKING NEWS: “International spy ring apprehended after staking out random suburban dad for third straight day.”

He crunches loudly.
Milk drips down his chin.

He sighs.

“Finally.”

Epic Fantasy Warriors: Fire and Ice Cinematic Poster

youtube-thumbnail

A prompt for generating a dramatic, cinematic poster-style portrait of two epic fantasy warriors standing back-to-back in a frozen storm, symbolizing the clash of fire and ice with specific details on their appearance, attire, and the contrasting lighting and atmosphere.

展开提示词全文
A dramatic cinematic poster-style portrait of two epic fantasy warriors standing back-to-back in a frozen storm. On the left, a battle-worn male warrior with wet, curly dark hair, head bowed in quiet resolve, gripping a medieval sword planted into the ice. Frost and snow cling to his fur-lined cloak and shoulders. On the right, a powerful female warrior in profile, pale skin glowing with fiery orange light, her body partially engulfed in flames thatcontrast against the icy blue atmosphere. Snow particles swirl through the air, blending fire and ice in a symbolic clash. Ultra-detailed faces, emotional intensity, volumetric fog, cinematic lighting, cold blue tones mixed with warm fire highlights, shallow depth of field, epic fantasy movie poster, hyper-realistic, 8K resolution, dramatic composition, sharp focus, high contrast, photorealistic textures.

Double Exposure Portrait of an AI Engineer

youtube-thumbnail

A prompt for generating a hyper-realistic, double exposure photograph combining the silhouette of an AI engineer's profile with a symbolic environment of glowing neural networks and futuristic server architecture, suitable for a stock photo aesthetic.

展开提示词全文
A double exposure photograph depicting the silhouette profile and headshot portrait of an AI engineer, with a symbolic environment constructed from glowing neural networks, floating code streams, and futuristic server architecture, in the style of double exposure photography, sun flare, white background, hyper-realistic, detailed rendering, stock photo, high-resolution photography.

Cinematic Futuristic Portrait Prompt

youtube-thumbnail

A prompt for generating a cinematic, high-contrast futuristic portrait of a person in streetwear, seated on a neon-green glowing Spotify stool, set in a dark digital studio with holographic elements.

展开提示词全文
Cinematic futuristic portrait of a person in casual streetwear and headphones, seated on a neon-green glowing stool with Spotify emblem. Dark digital studio with holographic sound waves, floating notes, and playlist UI panels. Dramatic neon rim lighting, volumetric fog, ultra-realistic, high contrast, shallow depth of field, 35mm lens, 8K, immersive musical vibe --ar 16:9.

Cyber Projector Numbers Beauty Portrait Prompt (JSON format)

youtube-thumbnail

A highly detailed JSON-formatted Nano Banana Pro prompt for generating an ultra-high resolution, cinematic beauty portrait of a woman with green numeric projections mapped across her face and body, emphasizing photorealism, low-key lighting, and correct projection physics.

展开提示词全文
{
  "generation_request": {
    "meta_data": {
      "tool": "NanoBanana Pro",
      "task_type": "cyber_projector_numbers_beauty_portrait",
      "version": "v1.0_GREEN_NUMBERS_PROJECTION_BEAUTY",
      "priority": "highest",
      "language": "en"
    },
    "input": {
      "mode": "text_to_image",
      "notes": "Create a photoreal cinematic beauty portrait of an adult woman with green numeric projections mapped across her face, neck, shoulders, and chest. The projection must look physically correct (wraps around facial contours), with soft falloff and slight blur at edges. Dark background with faint out-of-focus green digits. Moody, low-key lighting. No text, no watermark."
    },
    "output_settings": {
      "aspect_ratio": "4:5",
      "orientation": "portrait",
      "resolution_target": "ultra_high_res",
      "num_images": 1,
      "render_style": "cinematic_techno_editorial",
      "sharpness": "filmic_crisp",
      "grain": "subtle_35mm",
      "color_grading": "green_on_black_low_key"
    },
    "scene": {
      "environment": "dark studio, faint bokeh of green digits in the background",
      "lighting": "primary light is the projector (green digits), minimal soft fill from camera-left, deep shadows",
      "mood": "mysterious, cyber, intimate"
    },
    "subject": {
      "person": "adult woman",
      "pose": "3/4 portrait, head slightly tilted, eyes looking sideways past camera",
      "expression": "calm, intense, enigmatic",
      "hair": "long dark hair, sleek and natural",
      "makeup": "clean editorial skin, subtle shimmer on eyelids, soft liner, natural glossy lips",
      "skin_texture": "natural pores, realistic highlights"
    },
    "camera": {
      "camera_body": "ARRI Alexa Mini",
      "lens": "85mm prime",
      "aperture": "f/2.0",
      "focus": "sharp on eyes and lips, projection still readable, background bokeh"
    },
    "projection_details": {
      "content": "random green digits (0-9) and simple symbols",
      "behavior": "wraps on cheekbones, nose bridge, jawline, collarbones",
      "quality": "slight projector softness, realistic spill light on skin"
    },
    "negative_prompt": "text, watermark, logo, letters, neon signs, cartoon, anime, CGI look, plastic skin, over-smoothing, distorted face, extra limbs, extra fingers, deformed hands, incorrect projection physics, flat overlay, low resolution, heavy HDR"
  }
}

提示词模型无关,可直接用于 Midjourney / DALL·E / Flux / Stable Diffusion 等任意文生图模型。

依赖说明

- 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

适用平台

claude-codecursorcodexgemini-cliopenclaw

更新记录

官方源监视中
内容更新 2026-10-19

官方源有内容更新(首次纳入监控)

信息边界说明

本页信息摘自官方文档,仅作技术信息呈现,不代表本站的判断或推荐。价格、配额、可用性以官方最新说明为准。你离开本站后的任何行为(申请 Key、付费、部署、商用),本站不做任何承诺与担保。

同类推荐

视频短视频图片音乐音频配音口播数字人全能创作

官方标注收费

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

官方标注收费

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。

罗斯蒙德可视化图形生成

官方未说明

【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。

架构图一键生成

官方标注收费需申请 Key

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。

架构图一键生成

官方标注收费需申请 Key

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。

海报设计工作室

免费增值需申请 Key

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。

什么是技能

技能(Skill)是封装好的单项能力,比如文字识别、文档摘要、关键词提取。开发者把调用逻辑和提示词打包,使用者像用插件一样加载即可,无需了解底层模型。启枢技能 收录大量技能,可直接试跑或编入方案。