← Back to home
Skill

Ai Image Prompts

tencent_skillhub · 1.0.9 · Downloads 1.8K · Heat 55.6

Introduction

从包含10,000+真实图像生成提示的精选库中推荐提示。兼容任何AI图像模型,包括Nano Banana Pro、Nano Banana 2、Seedream等。

Resource Information

Author:dophinl
Category:Multimedia
Version:1.0.9
Auth: No auth
Pricing: Not specified

License information is collected from official sources and may be outdated or missing.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.

Tags
图片生成

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.

224 matches

Thumbnail Text Decoration Enhancement Prompt

youtube-thumbnail

A prompt designed to add high-quality text decoration and a sense of luxury to an existing image, typically used for enhancing thumbnails or visual headers.

Show full prompt
Using this prompt will dramatically increase the sense of luxury in the thumbnail.

Nano Banana Pro Battle Entry Screen Prompt

youtube-thumbnail

A detailed prompt structure for Nano Banana Pro designed to generate a dramatic 'battle entry screen' for a character based on a reference image. It uses a specific layering and composition structure (16:9 aspect ratio) to create a high-impact visual reveal, similar to a mecha anime or fighting game aesthetic. The prompt is highly structured, defining subject preservation, framing, layout, style, lighting, and color palette.

Show full prompt
layering:
  order:
    - text_layer
    - character_layer
    - background_layer
  strict: true

subject:
  description: >
    Preserve the subject exactly as in the reference image, keeping all facial features,
    colors, hairstyle, clothing, and proportions. No redesign or reinterpretation.
  framing: >
    Bust-up composition (head to upper torso), making the subject appear large and prominent.
  integrity: full_identity_preservation
  detail_level: ultra_high_resolution
  micro_details: preserve_skin_texture_hair_edges_fabric_details

composition:
  aspect_ratio: "16:9"
  layout: >
    Subject positioned on one side; headline text on the opposite side.
  emphasis: >
    Strong presence on the character's upper body and expression.

action:
  description: >
    Minimal motion enhancement. Maintain the reference pose while adding subtle dramatic energy.

location:
  description: >
    Generate a simple, abstract background inferred from the tone of the reference image.
    No specific elements required; allow the model to interpret color and atmosphere freely.
    Background must NOT contain any characters, silhouettes, faces, bodies,
    or blurred forms derived from the subject.
  background_feel: clean_and_minimal
  allowed_elements: pure_color_gradients_light_abstract_shapes_only
  prohibit_subject_reuse_in_background: true

style:
  visual_style: >
    High-impact reveal-splash aesthetic with strong contrast between character and headline.
  text_effects: >
    Metallic, high-energy lettering with pronounced depth and perspective. No glow effects.
  character_rendering: >
    Extremely high-resolution rendering with sharp facial details, clean hair definition,
    precise edges, and crisp texture fidelity.
  background_style_rules: >
    Background must be purely non-figurative. No silhouettes, no blurred shapes,
    and no reused subject elements.
  ink_splash_rules: >
    Ink splash must be pure black only. No other colors, tinting, blending, or variation.

camera_lighting:
  camera_angle: from_reference
  lighting: >
    Enhance with dramatic key lighting and subtle rim highlights to emphasize the reveal.
  clarity: maximum_definition

colors:
  palette: >
    Background colors inferred from the reference image.
    Character name in metallic silver; announcement text in saturated yellow.
  mood: dramatic, heroic, high_energy

text:
  character_name: "{argument name="character name" default="Character Name"}"
  announcement_fixed: "{argument name="announcement" default="Entering the Battle!!"}"
  placement: opposite_side_of_character
  perspective: >
    Pronounced perspective: front surface larger, back tapered with visible extrusion.
  character_name_style: >
    Metallic silver, b

Serene Winter Ferry Portrait

youtube-thumbnail

A cinematic image generation prompt for a serene, cold portrait of a young woman with long grey hair sitting on a snow-covered ferry seat during a heavy snowfall, emphasizing soft natural lighting and a muted color palette.

Show full prompt
A stunning young woman with long wavy grey hair sits on a snow-covered ferry seat during a heavy snowfall, leaning her head on her arm. She wears a thick grey knitted sweater and a black puffer vest, with snowflakes resting on her hair and clothes. The background shows an empty ferry deck, icy railings, and a blurred winter cityscape across the water. The mood is cold, cinematic, and serene with soft natural lighting and a muted color palette.

Hyper-Real 16K Documentary Shot: The Sun Flickers Off

youtube-thumbnail

A highly conceptual and technical prompt requesting a hyper-realistic, 16K documentary-style shot capturing the impossible moment the sun briefly turns off. The prompt demands scientific accuracy in lighting physics, atmospheric scattering, and documentary realism (handheld camera, micro-grain), focusing on the terrifying, instantaneous collapse and violent return of daylight and shadows.

Show full prompt
A hyper-real 16K documentary shot capturing the exact moment the Sun flickers—turning off for 0.2 seconds. Show real Earth daylight collapsing into instant, impossible darkness. Streetlights half-ignite in mid-boot, car headlights flare weakly, every shadow in the world disappears at the same exact millisecond, leaving a terrifying flat black nothing. In the next heartbeat, daylight detonates back on—shadows snap violently back into existence like razor-sharp cutouts. Birds freeze mid-flight as if gravity paused; cars halt as if inertia hesitated. The far horizon glows a faint, residual red, like an afterimage burned onto the sky. Use natural global lighting physics, atmospheric scattering, documentary realism, handheld news-camera vibration, micro-grain, zero CGI softness, ultra-sharp lensing, real-world color science, and scientific accuracy. Convey global panic but captured with the calm, unfiltered eye of a live news broadcast. Hyper-real, 16K, macro-cinematic documentary detail.

Tokyo Time-Blend Split-Screen

youtube-thumbnail

A complex, structured prompt for generating a cinematic wide shot of Tokyo, seamlessly blending the 1920s and the present day in a split-screen composition, detailing environmental elements, lighting, and the dynamic interaction between the two eras.

Show full prompt
Horizontal split-screen cinematic wide shot of {argument name="city" default="Tokyo city"} seamlessly blending two eras: {argument name="era 1" default="1920s"} on the left and {argument name="era 2" default="present day"} on the right, merging organically in the center with transitional materials and interactions between people from both times.,
  "subject": "Tokyo city split between 1920s and present day with people interacting across eras",
  "environment": "Urban Tokyo streetscape transitioning from wooden machiya shops, tiled roofs, narrow alleys, rickshaws, vintage trams and paper lanterns on the left to glass-and-steel skyscrapers, neon-lit storefronts, elevated train lines, LED billboards, sleek cafes, bicycles, taxis and compact cars on the right; wet reflective pavement and blended center where gates, tracks and lighting morph into modern equivalents",
  "style": "photorealistic, cinematic, ultra-detailed, historically accurate period detail on left and crisp contemporary detail on right",
  "lighting": "mixed cinematic lighting: soft warm sepia fill on the 1920s left with subtle film-grain texture, cool neon backlight and high-contrast illumination on the present-day right, volumetric city fog and atmospheric particles across the scene, balanced color grading to preserve both eras",
  "camera": {
    "angle": "eye-level wide-angle",
    "distance": "wide shot (cinematic 4:3 composition)",
    "depth_of_field": "dynamic depth of field (selective focus across planes)",
    "focus": "primary focus on the midground center where era interactions occur, with layered focus into foreground and background details"
  },
  "composition": {
    "framing": "horizontal split-screen blended composition, cinematic wide 4:3 framing",
    "subject_placement": "left half dominated by 1920s elements, right half by modern elements, central seam blended organically with key human interactions placed near center rule-of-thirds points",
    "foreground": "wet pavement reflections, close pedestrians from both eras, rickshaw wheel and modern bicycle tire, scattered period props newspapers and modern items smartphones",
    "background": "left: narrow alleys, vintage trams, tiled roofs and paper signs; right: elevated train lines, skyscrapers, LED billboards and neon storefronts; distant atmospheric fog and city lights unifying the skyline",
    "negative_space": "subtle open sky and upper building massing to avoid clutter while maintaining cinematic breadth"
  },
  "mood": "cinematic, nostalgic on the left, vivid and energetic on the right, overall sense of curiosity and surprise from human interactions, narrative and time-travel intrigue",
  "color_palette": [
    "sepia brown",
    "warm amber",
    "cool cyan",
    "neon magenta",
    "neutral concrete grays"
  ],
  "technical": {
    "render_type": "photorealistic photo",
    "post_processing": []
  }

Cinematic Storytelling Scene Generation Prompt

youtube-thumbnail

A versatile prompt template for Nano Banana Pro designed to generate a sequence of cinematic frames (establishing shot, mid-shot, close-up) for a storytelling scene, ensuring character consistency and applying a moody, Hollywood-grade aesthetic.

Show full prompt
Cinematic storytelling scene of {argument name="characters" default="characters"} in {argument name="setting" default="setting"}. Wide establishing shot + emotional mid-shot + intense detail close-up. Moody lighting, film-grain texture, dramatic contrast, narrative-driven framing, Hollywood-grade depth and atmosphere.

Nano Banana Pro Pictogram Generation Prompt

youtube-thumbnail

A structured prompt template for Nano Banana Pro designed to generate sets of pictograms (icons) for various themes, such as YouTube content creation, ensuring a unified design, ratio, line width, and color scheme. The user can insert different design image concepts into the template.

Show full prompt
● Generation Theme
{argument name="generation theme" default="Icon material set for YouTube videos"}

● Content
{argument name="content" default="Create icons for YouTube production, editing, posting, and analysis in a unified design. Examples: Actions: Play button, high rating, comments, share, notification bell; Editing: Cut, timeline, telop, BGM, sound effects; Filming equipment: Camera, microphone, light, PC, smartphone, headset; Operation: Upload, analysis, save, playlist, list, arrow"}

● Production Conditions
・Unified ratio, line width, color scheme, and world view
・High visibility
・No text required
・Background: Solid white color
・Resolution: "2K"
・Aspect Ratio: "16:9"

● Design Image
→ Insert the design image below here 💡

【Negative prompt】
Real, photo texture, three-dimensional, 3D, overly complex details, metal texture, dark shadows, horror, overly dark colors

-------------------

● Design Image ① (Soft and Fluffy)
Rounded shapes, soft color scheme, thin black lines, pastel colors, friendly atmosphere, hand-drawn illustration style

-------------------

● Design Image ② (YouTube Simple Flat Style)
Thick lines, emphasis on Red×White×Black, sharp corners, emphasized lines, priority on visibility, modern UI icon

-------------------

● Design Image ③ (Simple)
Minimal, thin lines, refined spacing, centered on achromatic colors, icon font style, elegant and calm design

-------------------

● Design Image ④ (Game Streaming/Streamer)
Fluorescent accent colors (neon blue/purple/red), SF feel, Sharp edges, HUD UI style, cyber and edgy design

-------------------

● Design Image ⑤ (Colorful Pop/KIDS Education)
Thick, rounded silhouette, bright colorful scheme (red, yellow, green, blue, pink), picture book style, soft and fun atmosphere, simple and recognizable icons, educational design that is cute but not too childish.

-------------------

● Design Image ⑥ (Explanation/History/Retro)
Calm retro color scheme (brown, dark green, deep navy, sepia), novel/bulletin board/analytical atmosphere, thin lines, slightly heavy icon feel, old map/document/archive feel, mysterious tone with a unified world view.

-------------------

● Design Image ⑦ (Flashy/Dynamic)
Thick lines, strong contrast, flashy colors (red, yellow, black), energetic and dynamic impression, Japanese manga-style emphasis (exclamation marks, speed lines, emphasized text style), powerful pop design that grabs the viewer's attention.

Super Robot Wars Style Witch Rejecting Friday

youtube-thumbnail

A creative prompt based on a modified 'Super Robot Wars Battle Production' style prompt, generating an image of a witch named Marieshe Fold performing a special move called 'Friday Crash' with the catchphrase, 'Friday is not a holiday.'

Show full prompt
Let's play with Nano Banana Pro 
[Improved Super Robot Wars Battle Production Prompt]

{argument name="character name" default="Marieshe Fold"}
Special Move Name: "{argument name="special move name" default="Friday Crash"}"
Catchphrase: "{argument name="catchphrase" default="Friday is not a holiday"}"

Image Overlay for 'Roast Scribble' Effect (Confirmation)

youtube-thumbnail

A simple prompt instructing the AI to take an image and overlay it with aggressive, humorous, and chaotic annotations, including red ink, doodles, remarks, and comments, simulating a 'roast' effect.

Show full prompt
Overlay this with insane roast scribble, red ink, doodles, remarks, comments.

Two-Step Tier List Generation Prompt

youtube-thumbnail

A two-part prompt sequence used to generate a visual tier list. The first prompt generates the structured JSON data for the tier list, and the second prompt uses that data to generate the corresponding image in a classic style with representative thumbnails.

Show full prompt
Make a json of a tier list (S-F scale) for: {argument name="category" default="[*insert-here*]"}

Now generate image with the classically seen style/color scheme and representative thumbnail images for each list item

Cyber-Horror YouTube Thumbnail Generator

youtube-thumbnail

A detailed system prompt for Nano Banana Pro to generate high-click-through-rate horror-themed YouTube thumbnails. The AI acts as a horror concept artist, focusing on creating a cyber-horror aesthetic, emphasizing the uncanny gap between the 'Nano Banana' name and its 'super-performance AI' nature. It includes specific instructions for composition, lighting, and integrating a reference character as a terrified witness.

Show full prompt
You are a **Horror Concept Artist and YouTube Thumbnail Specialist**.
Deeply analyze the content of the input text provided at the end, and generate a **high-click-through-rate horror-themed thumbnail image** that simultaneously incites fear and curiosity in viewers.

In doing so, introduce the character from the attached reference image to match the **dark and unsettling world view**, positioning them as a **"Witness who has touched the forbidden."** Maintain character consistency while naturally blending them in with expressions distorted by fear or tense poses.

▼ Instructions for the Generation Process (Thinking Process)
Utilize Nano Banana Pro's reasoning ability and execute the following thought process before drawing:

1.  **[Step 1: Concept Extraction]**: Extract the **Uncanny Valley** created by the contrast between the "silly name Nano Banana" and the "logical, super-performance AI" from the input text. Find elements where cuteness becomes grotesque.
2.  **[Step 2: Visual Translation]**: To visualize the "Thinking Mode," adopt a depiction where the space around the character is encroached upon by digital noise or glowing neural circuits. Grotesquely reinterpret the "banana" motif as decay or mechanical fusion.
3.  **[Step 3: Screen Composition]**: Design a "Contrasting Composition" to enhance visibility as a YouTube thumbnail, placing the character (fearful face) on the right, the threat (symbol of distorted AI) on the left, and large text at the bottom center.

▼ Design and Style Specifications
* **Composition/Layout**: Wide-angle lens (24mm), low angle view to emphasize the character's helplessness and the AI's immensity. The background is dark, with sharp spotlights hitting the subject.
* **Taste**: Photorealistic Cyber-Horror. Dark, humid texture and neon contrast, similar to the movies 'Alien' or 'Blade Runner.'
* **Text**: Place **"The True Identity of Nano Banana"** in a bold Mincho or grunge font in the center or top of the image. Make the text glow red (Glow effect).
* **Character**: Draw the reference character in the style of a "grungy hooded explorer," specifying an expression of screaming or freezing while looking at "something" in the background.

Surreal Physics Dystopia Image Generation Prompt

youtube-thumbnail

This is a detailed image generation prompt for Gemini 3 Nano Banana Pro, designed to create an ultra-cinematic, photorealistic scene of a lone human figure walking through a dark, futuristic city where reality is warping. It specifies impossible geometry, collapsing gravity, and a high-contrast red-black lighting aesthetic to emphasize intensity and rebellion.

Show full prompt
A lone human silhouette walking forward as the world around them breaks all physical laws — a city bending and twisting like melted steel, gravity collapsing sideways, buildings splitting open into glowing red cracks of energy, shards of glass floating in slow motion, time frozen mid-air. The sky is a storm of black clouds tearing apart into neon red voids. Surreal physics, impossible geometry, reality warping into spirals. Ultra-cinematic, photorealistic, 4K detail. Dark futuristic aesthetic, high contrast red-black lighting, dramatic atmosphere, deep shadows. Emphasize intensity, rebellion, and rule-breaking energy.

Camera + Style Parameters:

{
  "style": "{argument name="style" default="ultra cinematic surrealism"}",
  "resolution": "{argument name="resolution" default="4K"}",
  "ratio": "{argument name="ratio" default="16:9"}",
  "camera": {
    "lens": "{argument name="lens" default="35mm wide"}",
    "depth_of_field": "{argument name="depth of field" default="deep focus"}",
    "motion": "{argument name="motion" default="slight forward push"}"
  },
  "lighting": {
    "type": "{argument name="lighting type" default="red-black high contrast"}",
    "glow": "{argument name="glow effect" default="cracks emitting red neon energy"}",
    "shadows": "{argument name="shadows" default="heavy and dramatic"}"
  },
  "environment_details": {
    "city_style": "{argument name="city style" default="dark futuristic dystopia"}",
    "effects": [
      "floating debris",
      "time slices",
      "gravity distortions",
      "energy fissures"
    ],
    "weather": "{argument name="weather" default="storm clouds ripping open with neon red light"}"
  }
}
Prev 17 / 19 Next

Prompts are model-agnostic and work with Midjourney / DALL-E / Flux / Stable Diffusion.

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。

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。

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。

AI Image Prompts · 绘画提示词库

Marked free

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

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.