← 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 702 条

Four-Panel Manga Prompt: Oil Rush Panic

comic-storyboard

A prompt for Nano Banana Pro (Gemini 3 Pro Image) to generate a four-panel manga titled 'Oil Rush Panic,' mentioning the comedy duo Viking and specifically the member Eiji Kotouge.

展开提示词全文
4-panel manga
Title:
'Oil Rush Panic'

Viking?

Only Eiji Kotouge
Barely...

Others changed characters

Four-Panel Manga Prompt: Judgment of the Ministry of Magic

comic-storyboard

A prompt for Nano Banana Pro (Gemini 3 Pro Image) to generate a four-panel manga titled 'Judgment of the Ministry of Magic,' focusing on the concept of 'cache' (reality) versus 'cache' (footprints) and the weight of cache.

展开提示词全文
4-panel manga
Title: Judgment of the Ministry of Magic

Cache (reality)
and
Cache (footprints)

What is the weight of cache...

Manga Magazine Cover Prompt for Nano Banana Pro

comic-storyboard

A prompt used with Nano Banana Pro to generate a cover for a manga weekly magazine featuring a character named Haruka, walking and eating a chocolate crepe under cherry blossoms, pointing at the reader, and announcing a new series start.

展开提示词全文
While the cherry blossoms are blooming, {argument name="character name" default="Haruka"} is walking around town eating a chocolate crepe.

She is pointing at the reader, as if saying, "New serialization starts this week!!"

3D Diorama of Comic Book Covers and Toys

comic-storyboard

A complex, structured prompt instructing Nano Banana to visualize the DC Universe using a specific structural geometry: a 3D diorama featuring a vertical scenic backdrop (book cover), 3D toys of heroes and villains, and open book pages with comic panels, all rendered photorealistically.

展开提示词全文
2x2 grid, do this for 4 titles in a famous comic universe: <spatial_map>
    [ VERTICAL SCENIC BACKDROP (BOOK COVER) ]
    |      (Floating Label: [{argument name="issue title" default="ISSUE_TITLE"} | {argument name="year" default="YEAR"}])
    |                                            |
    |      [3D TOY: {argument name="hero" default="HERO"}] vs [3D TOY: {argument name="villain" default="VILLAIN"}]   |
    |          \__ [3D PLASTIC FX] __/           |
    |                                            |
 ---+--------------------------------------------+---
   /    [OPEN BOOK PAGES WITH COMIC PANELS]     /
  /                                            /
 /      [ FLOATING 3D TEXT: [SFX_WORD]! ]     /
/____________________________________________/
</spatial_map>
Instruction: Render a photorealistic 3D diorama matching this exact structural geometry.

AI Image Generation Prompt for 4-Panel Manga: Academic Silence

comic-storyboard

A user shares the prompt used with Nano Banana Pro (Gemini 3 Pro Image) to generate a 4-panel manga titled 'Academic Silence,' focusing on university names and academic hierarchy.

展开提示词全文
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

World Baseball Classic Scene Generation

comic-storyboard

A prompt for Nano Banana Pro (Gemini 3 Pro Image) to generate a chaotic scene related to the World Baseball Classic, involving South Korea, the Dominican Republic, and a Gorenja Rugby Ball.

展开提示词全文
World Baseball
South Korea Dominican Republic
South Korea tattered
Like a bee on a crying face
Set up a Gorenja Rugby Ball
Until the kick and boom

Cinematic Portrait in Ruined Gothic Cathedral

comic-storyboard

A dramatic, cinematic portrait prompt for Gemini Nano Banana 2, featuring a subject in an abandoned gothic cathedral at twilight, emphasizing low-angle perspective, intense volumetric god rays, chiaroscuro shadows, and a dark fantasy epic film aesthetic.

展开提示词全文
Cinematic dramatic portrait of [{argument name="subject description" default="your description"}], low camera angle looking up heroically, intense volumetric god rays piercing through shattered stained-glass windows and collapsed roof in an abandoned gothic cathedral at twilight, cool blue-purple volumetric light beams cutting through thick dust, incense-like haze, and floating particles, strong rim lighting and subtle warm backlight halo around the face and hair, sharp chiaroscuro shadows sculpting facial features, towering ruined stone arches, broken pews and debris softly blurred in background, dynamic contemplative pose - head tilted slightly upward toward the divine light, eyes gazing intensely into the beams, subtle wind moving hair and clothing, moody ethereal atmosphere blending dystopian ruin with sacred majesty, film still from dark fantasy epic, volumetric fog everywhere, god rays dominating composition, ultra-realistic skin texture with pores and subtle highlights, anamorphic lens flares, cinematic color grade with deep teal shadows and faint golden rim, shot on 35mm Arri Alexa, directed by Ridley Scott or Guillermo del Toro, atmospheric depth, masterpiece.

Cinematic Superhero Rooftop Prompt

comic-storyboard

A detailed prompt for generating a cinematic film still of a superhero (in a red and blue spider suit) sitting on a skyscraper rooftop, focusing on specific camera angles, lighting (overcast/diffused daylight), texture realism (torn suit fabric), and cool color grading.

展开提示词全文
A superhero in a {argument name="suit color" default="red and blue"} spider suit sits on the edge of a skyscraper rooftop, viewed from the back, in a close-up medium shot, with the hero in the center of the frame. He is leaning slightly forward, one hand resting on his knee. On his back is a large red spider logo. At the top of the suit, there is a small torn hole on the right shoulder, revealing the texture of the fabric. The background is a densely built-up metropolis with beige and gray high-rises in the New York style, the windows repeating in a geometric pattern. The city is slightly blurred (slight depth of field), the hero is in sharp focus. The weather is overcast, soft diffused daylight, a cloudy sky without bright sun. The cinematic style of a 2010s superhero film, realistic texture of the suit fabric (micromesh, volumetric pattern), physically correct lighting, natural shadows, light atmospheric haze between the buildings. Color grading: cool bluish-teal tone, muted colors, reduced saturation, soft contrast, slight gray-green tint in the shadows, slightly darkened and not acidic red suit. Overall tone is dramatic, calm, slightly melancholic. Camera: 70-100mm lens, eye level, waist-length framing, cinematic framing. No action, static scene. High detail, realism, film still, ultra realistic, soft natural lighting, shallow depth of field, subtle film grain.

Photorealistic Superhero on Skyscraper Rooftop

comic-storyboard

A highly detailed prompt for generating a photorealistic image of a superhero (Spider-Man suit) sitting on a skyscraper rooftop, with the mask removed to reveal the face of Caleb McLaughlin (using reference image 2). It specifies the pose, suit details (torn shoulder), background (industrial rooftop), overcast lighting, and a cool bluish-teal cinematic color correction.

展开提示词全文
A superhero in a red and blue spider suit sits on the edge of a skyscraper rooftop, frontal view, camera at eye level, waist-length medium shot. The pose is identical to the first shot: the body is slightly leaning forward, one hand resting on the knee, a static, cinematic composition.

The mask is removed and held in the hand or lowered downwards. Face: Use the face from the uploaded reference photo 2 (identity reference 2), as photorealistic as possible, without stylization, with natural skin texture and natural shadows. Expression: calm, pensive, slightly tired, with a dramatic mood.

The suit is completely identical to the first shot: 
red and blue, micro-mesh texture, 3D webbing, realistic fabric, and retains a small torn detail on the shoulder. Physically correct fit.

The background corresponds to the camera's 180-degree rotation around the hero: 
another part of the roof features industrial elements—ventilation units, metal structures, and technical equipment. Beyond the roof are lower city buildings, an urban landscape, concrete facades, and a subtle industrial atmosphere. The city is slightly blurred (shallow depth of field). Atmospheric haze.

The lighting is identical to the first shot: 
an overcast day, soft, diffused light, and a cloudy sky, without harsh shadows. The light falls as naturally as in the first scene.

The color correction is exactly the same as in photo 1: 
a cool bluish-teal tone, muted colors, reduced saturation, soft contrast, a slight gray-green tint in the shadows, and a darkened, not bright, red suit. A dramatic, calm, cinematic tone.

Camera: 
70–100mm lens, cinematic framing, shallow depth of field, hero in focus, background slightly blurred. Ultra-realistic, film still, subtle film grain, physically accurate lighting, cinematic superhero movie 2010s style.

Maximum realism, the feeling of a real film frame, without CGI stylization.

Mechanical Wind-Up Diorama Concept

comic-storyboard

A complex, structured prompt template for Nano Banana Pro designed to visualize a mechanical wind-up diorama where a superhero bursts free from comic panels, specifying base setup, core action, variants, narrative elements, and technical details like steampunk aesthetic and octane render.

展开提示词全文
2x2 grid, do this for 4 lesser known scifi movies <instructions>
 
concept: a mechanical wind-up diorama where a superhero activates and bursts free from confining comic panels in perpetual motion.
input variables:
[insert superhero character]
[insert signature action or power]
[insert mechanical elements]
[insert surrounding props]
workflow:
base setup: intricate brass and wood kinetic sculpture on a desk, with visible gears, springs, and [insert mechanical elements] implying automated movement.
core action/composition: [insert superhero character] posed mid-[insert signature action or power], breaking out of hinged comic panel frames that flip open like doors.
variants and details: multiple pose variants connected via hidden mechanisms, with scattered tools hinting at handmade construction.
narrative/interactive elements: implied perpetual motion as the figure "escapes" repeatedly, with engraved onomatopoeia plates.
god-level details: steampunk-inspired macro photography angle, warm tungsten lighting casting gear shadows, polished metal reflections and aged paper textures, hyper-realistic kinetic blur, 8k octane render with caustic light play.
</instructions>

Operation Epic Furry Illustration Prompt

comic-storyboard

A prompt given to Gemini Nano Banana to generate an illustration capturing the essence of a geopolitical conflict, specifically the war between Iran and the United States & Israel, framed under the fictional name 'Operation Epic Furry.'

展开提示词全文
Draw an illustration that captures the essence of the war between Iran and the United States & Israel as if the name of the operation was 'Operation Epic Furry.'

Lego Racer on Exploding Volcano Slope

comic-storyboard

A prompt for Nano Banana Pro and Kling 3.0 to generate an image of a Lego racer minifigure driving down a jungle volcano slope, with a lava explosion lighting the scene and the minifigure showing cartoonish panic, stylized as an action blockbuster LEGO scene.

展开提示词全文
Lego racer minifigure driving at full speed down a jungle volcano slope, lava explosion behind lighting the kart in orange glow, hot rocks raining down, cartoonish panic face and comic urgency, stylized as action blockbuster LEGO scene.

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 2026-10-21

内容有更新(细节请以官方文档为准)

内容更新 2026-10-20

内容有更新(细节请以官方文档为准)

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。

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。

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。

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。

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。

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。