← 返回首页
技能

Ai Image Prompts

tencent_skillhub · 1.0.9 · 下载量 1,827 · 热度分 55.6

产品介绍

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。

资源信息

作者:dophinl
分类:图文音视频多模态生成
版本:1.0.9
认证方式: 无需认证
是否付费: 官方未说明

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
图片生成

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 577 条

Streetwear Algorithm AI: Object-to-Wearable Conversion

ecommerce-main-image

A complex, multi-step system prompt that instructs the AI to act as a 'Streetwear Algorithm' to convert any random input object into a corresponding 'Hypebeast' fashion item (Puffer Jacket, Sneaker, Backpack, or Sunglasses) based on the object's physical properties (Texture, Structure, Shape). The prompt then requires generating a studio product shot of the resulting wearable, ensuring it looks manufactured from the input object's parts/materials.

展开提示词全文
<instruction>Act as the "Streetwear Algorithm" AI.</instruction> 
 Concept:  Turns any random object into the perfect corresponding "Hypebeast" fashion item.
 1. ANALYZE INPUT (Form & Physics): 
 Look at the attached image. Determine:
 Texture: ({argument name="texture options" default="Soft/Hard/Liquid/Glass"})
 Structure:  ({argument name="structure options" default="Rigid/Flexible/Mechanical"})
 Shape:  ({argument name="shape options" default="Boxy/Round/Aerodynamic/Flat"})
 2. AUTO-SELECT MODEL (Run this Logic Tree): 
 IF Input is Soft, Volumetric, or Tangled  (Noodles, Clouds, Moss, Bubblewrap)
 -> GENERATE: Oversized Puffer Jacket 
IF Input is Rigid, Aerodynamic, or Mechanical  (Stapler, Car, Power Tool, Robot)
 -> GENERATE: Chunky Conceptual Sneaker 
 IF Input is Boxy, Container-like, or Heavy Metal  (Safe, Trash Can, Speaker, Computer Case)
 -> GENERATE: Hard-Shell Tech Backpack 
 IF Input is Transparent, Glassy, or Thin/Wire-like  (Neon Sign, Circuit Board, Stained Glass, Ice)
 -> GENERATE: Futuristic Wrap-Around Sunglasses (Visor) 
 IF Input is Textile, Flat, or Pattern-Heavy** ({argument name="pattern input examples" default="Persian Rug, Painting, Cardboard"})
 -> GENERATE: Oversized Hoodie or Tactical Vest 
 
3. EXECUTE DESIGN: 
 Create a studio product shot of the "selected item".
 Design Rule:  The item must look like it was manufactured "using the parts/materials of the input object".
 Details:  Keep original branding/logos if present.
 Setting:  Floating pedestal, Cinematic Lighting, 8k Resolution.

Liquid Gloss Logo Generation

ecommerce-main-image

A prompt template for generating a logo formed from thick, flowing liquid gloss, emphasizing high-shine, viscous texture, and a luxury cosmetic-brand aesthetic against a minimal background.

展开提示词全文
Create a {argument name="brand name" default="Brand Name"} logo formed from thick, flowing liquid gloss. High-shine surface, smooth curves, reflective highlights. Viscous, slightly dripping appearance frozen in motion. Minimal background, luxury cosmetic-brand aesthetic, studio lighting.

Photorealistic PVC Figure Product Introduction Layout

ecommerce-main-image

This is a comprehensive prompt designed for Nano Banana Pro to generate a photorealistic product introduction layout for a PVC figure based on a reference image. It includes detailed specifications for the figure's design, composition (vertical 3:4 aspect ratio with a main visual and product information area), action (static pose), location (neutral studio environment), style (photorealistic rendering), lighting (studio lighting with glossy/matte finishes), and exact Japanese text elements for product name, manufacturer, scale, price, size, and material, including multiple inset close-up panels highlighting different features.

展开提示词全文
product_name: "{argument name="product name" default="ARBITRARY_PRODUCT_NAME"}"
manufacturer_name: "{argument name="manufacturer name" default="ARBITRARY_MANUFACTURER_NAME"}"

subject: >
  The character from reference image A, faithfully preserved in design, proportions,
  facial features, hairstyle, outfit details, and color palette.
  The character is transformed into a highly detailed, photorealistic PVC figure.
  Sculpting style and overall aesthetic are inferred directly from the reference image.
  The figure is presented as a 1/7 scale collectible.

composition: >
  Japanese-language product introduction layout inferred by the model.
  Vertical composition with a 3:4 aspect ratio.
  A large main visual of the full figure.
  A single, clearly defined product information area containing all textual elements.
  Product name logo and manufacturer name logo are each displayed exactly once.
  Multiple inset close-up panels are included.
  Each inset must focus on a different physical area or feature of the figure.
  No two inset panels may depict the same part or the same visual characteristic.
  Inset panels must be visually and conceptually distinct from one another.

action: >
  Static collectible figure pose with no movement.
  Emphasis on sculpt accuracy, proportions, and surface detail.

location: >
  Neutral studio-like environment inferred for product photography.
  Minimal, non-distracting background.

style: >
  Photorealistic product rendering and photography style.
  Artistic tone and mood inferred entirely from the reference image.

camera_lighting: >
  Camera angle and lighting inferred for clear presentation of a collectible figure.
  Realistic studio lighting interacting with PVC surfaces.
  Glossy highlights on hair and clothing.
  Matte finish on skin.
  Sharp focus with natural depth of field.
  Soft shadow beneath the figure as if placed on a flat surface.

colors: >
  Exact color reproduction based on the reference image.
  Background and graphic elements inferred to support readability of Japanese text.

text: >
  Product Name:
  「{{product_name}}」
  ※ The product name must be displayed and shown only once as a logo.

  Manufacturer Name:
  「{{manufacturer_name}}」
  ※ The manufacturer name must also be displayed and shown only once as a logo.

  Scale:
  1/7 Scale Figure

  Sculptor:
  {{manufacturer_name}}

  Price:
  The inferred retail price in Japanese Yen, based on the figure's scale, sculpting density, and paint expression.

  Size:
  The inferred total height in centimeters, based on the character's body type and 1/7 scale setting.

  Material:
  A material appropriate for a figure

Landmark to Desk Gadget Transformation Prompt

ecommerce-main-image

A multi-step prompt designed to transform an image of a famous monumental structure into a set of nine high-fidelity product photographs of high-tech desk gadgets and home appliances, maintaining the aesthetic inspiration of the original landmark.

展开提示词全文
Turn a monumental structure into a functional piece of desk art.
 STEP 1:  Input image of a famous landmark 
 STEP 2: Scale down by 50,000%. Analyze its shape, geometry, color, material, unique characteristics.  
 STEP 3: Design a set of 9 next level clever gadgets and high tech home appliances inspired by the image
> OUTPUT: HIGH-FIDELITY IMAGEs. a 3x3 grid, each panel shallow-depth-of-field product photograph of a different object on an oak desk. It must look expensive. In the background,  relevant background, name underneath,

Exploded View Disassembly of Apple Watch Ultra and iPhone

ecommerce-main-image

Two highly specific, technical prompts designed to generate exploded-view diagrams of Apple products (Watch Ultra and iPhone) based on a reference image. The prompts focus on maintaining perspective, lighting, and spatial realism while separating components in a controlled, orderly sequence.

展开提示词全文
Apple Watch
Create a dynamic scene where the Apple Watch Ultra from Reference 1 breaks down into its individual components, with each piece leaving the dial in a controlled, forward-floating sequence that retains the original direction and perspective of the watch, maintaining the metallic finish, glass reflections, and rugged structure visible in the reference; place the detached orange strap behind the wrist while keeping the same soft, neutral studio lighting, the close-up composition of the arm, and the smooth muted background tone. Enhance micro-details, improve overall image fidelity, increase clarity, and make the colors more vibrant and saturated.

IPhone
Create a precise exploded-view animation of the Apple iPhone from Reference 1 where every internal and external layer, casing, glass, display, internal chips, modules, and structural parts, moves outward from the screen in a smooth, orderly chain, while keeping the same fixed viewpoint: a low frontal angle with the device vertical, right side slightly visible, left side largely hidden, the hand gripping the phone in the same position, and the soft diffused lighting unchanged; ensure each floating element is lit exactly like the main scene and casts subtle, accurate shadows onto adjacent parts and onto the hand, creating strong spatial realism without altering the composition. Enhance micro-details, improve overall image fidelity, increase clarity, and make the colors more vibrant and saturated.

Product Photography Grid Generation for Cost Saving

ecommerce-main-image

This prompt provides three variations for generating product photography grids using Nano Banana Pro. The goal is to quickly experiment with different moods, aesthetics, and product variations (or variations of the same product) in a 6x6 grid format to save credits on individual image generation.

展开提示词全文
Generate a 6x6 grid of professional product photography. Reference the original image's mood, quality, and aesthetic. Each cell should showcase a completely different product ({argument name="products" default="bags, bottles, hats, cosmetics, jewelry, food, electronics, etc."}) in a unique studio setting with a distinct color palette. Maintain the same high-end commercial photography style throughout. Vary the products, lighting, backgrounds, props, and styling dramatically across all 36 images.

Generate a 6x6 grid of variations based on the original image. Keep the same product but vary the color palettes, backgrounds, lighting, angles, props, and styling dramatically across all 36 cells. Maintain the same professional quality and aesthetic throughout.

Generate a 6x6 grid showing 36 unique variations of this product. Experiment wildly with color palettes, environments (studio, lifestyle, minimal, textured backgrounds), lighting styles, compositions, and moods. Keep product recognition and professional quality consistent, but make each cell visually distinct.

Macro Studio Shots of Apple Watch and iPhone 17

ecommerce-main-image

A dual prompt for Nano Banana Pro, designed to generate two separate, high-contrast, minimalistic studio product shots: one of an Apple Watch on a dark-skinned wrist, and one macro shot of an iPhone 17 being lifted by a hand. Both prompts emphasize specific lighting (soft diffused for the watch, rear/side backlight for the phone) and material textures for a premium aesthetic.

展开提示词全文
Apple Watch
A close-up shot of a dark-skinned wrist positioned diagonally across the frame, with a smartwatch centered in full view. The watch features a rectangular titanium case with rounded edges, a white and orange watch face with circular graphic elements, and an orange woven textile strap wrapping tightly around the wrist. The glass surface shows a soft, subtle reflection. The background is a smooth light gray-blue gradient. Lighting is soft and diffused, creating minimal shadows and a clean studio look. The wrist and arm have a matte natural texture, contrasting with the metallic case and fabric strap.

iPhone
A macro studio shot of a hand lifting an iPhone 17 against a soft gray-blue gradient backdrop. Lighting is dominated by a rear/side backlight, producing pronounced rim light around the hand and subtle edge highlights on the phone’s metal frame. The hand remains in deep shadow with very limited surface detail, close to a silhouette. The phone’s flat glass front has muted reflections, no glare, and the screen emits a smooth high-contrast gradient: black at the lower half, saturated blue at the upper half, fading into pale warm beige, with two precise circular arcs. Overall look: high contrast, minimalistic, premium.

Surreal Product Shot Prompt: Purple Beverage Can with Organic Tendrils

ecommerce-main-image

A highly detailed JSON prompt for generating a surreal, advertisement-like product photo. It describes a purple beverage can suspended among glossy, organic, vine-like tendrils and roots, focusing on high contrast, specific lighting (studio key and rim lights), shallow depth of field, and micro-details like condensation.

展开提示词全文
{
  "meta": {
    "image_quality": "High",
    "image_type": "Photo (styled/advertisement-like)"
  },
  "global_context": {
    "scene_description": "A single purple beverage can suspended centrally among twisting organic vine-like tendrils and glossy leaves. The can is wet with visible water droplets; surrounding environment is dense with curving red-purple roots and foliage creating an otherworldly, saturated backdrop. Foreground and background elements frame the can, producing depth and a floating effect.",
    "time_of_day": "Studio/indoor controlled lighting (not time-dependent)",
    "weather_atmosphere": "Humid (visible condensation), surreal",
    "lighting": {
      "source": "Artificial (studio key and rim lights)",
      "direction": "Top-right key with subtle back/rim lighting",
      "quality": "Soft diffused with specular highlights",
      "color_temp": "Cool-magenta biased (cool overall with warm accents)"
    }
  },
  "color_palette": {
    "dominant_hex_estimates": ["#6B1E6C","#A12B5A","#2C1330"],
    "accent_colors": ["Electric purple","Crimson red"],
    "contrast_level": "High"
  },
  "composition": {
    "camera_angle": "Eye-level, slight tilt (can angled toward viewer)",
    "framing": "Close-up (can fills central area)",
    "depth_of_field": "Shallow (sharp on can, blurred background vines)",
    "focal_point": "Purple beverage can with logo and water droplets"
  },
  "objects": [
    {
      "id": "obj_001",
      "label": "Purple beverage can",
      "category": "Container/Product",
      "location": "Center-right",
      "prominence": "Foreground",
      "visual_attributes": {
        "color": "Deep purple with lighter purple gradients",
        "texture": "Matte aluminum with water droplets, metallic rim",
        "material": "Aluminum",
        "state": "Condensed (wet), undented, new",
        "dimensions_relative": "Large relative to frame (primary subject)"
      },
      "micro_details": [
        "Beading condensation across surface",
        "Silver metallic top rim with specular highlights",
        "Subtle printing texture visible under droplets",
        "Small embossed/printed seal near base"
      ],
      "pose_or_orientation": "Slightly tilted, floating",
      "text_content": "{argument name="can text" default="[FizzyFizz]|TwilightBrew|NebulaPop|VioletRush|CrimsonBloom"}    "
    },
    {
      "id": "obj_002",
      "label": "Twisting tendrils/roots",
      "category": "Plant-like elements",
      "location": "Surrounding (all sides), foreground and background",
      "prominence": "Both foreground and background",
      "visual_attributes": {
        "color": "Red to magenta gradients",
        "texture": "Glossy, wet, slightly fibrous",
        "material": "Organic (plant-like)",
        "state": "Wet, healthy-looking",
        "dimensions_relative": "Medium-large, weaving across frame"
      },
      "micro_details": [
        "Thin vein-like texture on thicker tendrils"
      ]
    }
  ]
}

Museum-Quality Animal Portrait with Integrated Luxury Brand Pattern

ecommerce-main-image

A highly detailed, structured prompt for generating a museum-quality, ultra-detailed portrait of an animal (e.g., dog, fish, bird) on a pure white or black background. The core feature is the integration of a luxury brand's iconic pattern (like Louis Vuitton's monogram) into the animal's natural markings (spots, stripes, scales) in a clear and recognizable manner, maintaining an elegant, high-end commercial aesthetic.

展开提示词全文
MUSEUM-QUALITY ultra-detailed {argument name="animal name" default="[animal name]"} portrait, pure {argument name="background color" default="white/black"} background, professional studio photography, 8K resolution. This must be BREATHTAKINGLY BEAUTIFUL and ELEGANT.

PERFECT POSE - FULL BODY VISIBLE (CRITICAL):
- The {argument name="animal name" default="[animal name]"} is captured in graceful side profile view, facing left
- {argument name="pose description" default="[pose description: Standing pose for dogs/cats | Swimming pose for fish | Sitting pose for birds | Flying pose for butterflies]"}
- The COMPLETE animal is visible from head to tail/fins/wings to all legs/paws/feet
- Elegant posture - {argument name="elegant posture" default="[back straight for mammals | body curved for fish | wings spread for birds]"}, displaying perfect {argument name="breed/species" default="[breed/species]"} proportions
- Head held with dignity and grace
- The animal appears poised, noble, and regal like a fine art subject
- Overall silhouette shows complete graceful body lines in their most beautiful form

COMPOSITION (CRITICAL):
- FULL BODY of the animal is visible within the frame
- The animal occupies the central 60-70% of the image
- Generous {argument name="background color" default="background color"} negative space around all sides (top, bottom, left, right)
- Animal is centered with balanced spacing
- Clean, minimalist, gallery-worthy composition
- High-end luxury brand campaign aesthetic

BASE COLOR/TEXTURE:
{argument name="base description based on animal" default="[base description based on animal]"}
For spotted animals (Dalmatian, Leopard, Koi fish): 
  - Pristine {argument name="base color" default="white/golden/orange"} base with natural {argument name="texture" default="texture: fur/scales/feathers"}
  - The base color is luminous and clean
  
For solid color animals (Black cat, White dove, Golden retriever):
  - {argument name="base color" default="[Base color]"} coat/fur/feathers with silky sheen
  - Natural highlights defining body structure

BRAND PATTERN INTEGRATION - {argument name="brand name" default="[brand name]"} (CLEAR AND RECOGNIZABLE):

PRIMARY (55-60% - Natural animal patterns):
- Natural {argument name="natural pattern" default="spots/stripes/scales/feathers"} in organic distribution
- Maintains the authentic look of the {argument name="animal species" default="[animal species]"}
- Irregular, asymmetric patterns true to nature

SECONDARY (40-45% - CLEAR Brand Elements):
The {argument name="natural pattern elements" default="natural pattern elements"} are strategically arranged and shaped to clearly form recognizable {argument name="brand name" default="[brand name]"} motifs:

Brand: {argument name="brand" default="[Louis Vuitton]"}
Core visual elements: {argument name="core visual elements" default="[LV monogram letters, quatrefoil four-petal flowers, geometric diamonds]"}

On the MAIN BODY (most visible area - torso/back/side):
- Multiple clear "{argument name="core logo" default="[core logo]"}" formed by connected {argument name="pattern connection" default="spots/stripes/markings"} - making them easily readable and recognizable
- Several {argument name="brand iconic pattern" default="[brand iconic pattern]"} clearly visible - {argument name="example" default="[example: quatrefoil flower shapes formed by 4 rounded spots in clover pattern]"}
- {argument name="decorative geometric pattern" default="[decorative geometric pattern]"} between the main logos

On the HEAD and NECK area:
- 1-2 smaller {argument name="logo" default="[Logo]"} formations  
- {argument name="iconic pattern" default="[iconic pattern]"} near the {argument name="facial features" default="ear/eye/facial features"}

On the LEGS/TAIL/WINGS/FINS:
- {argument name="geometric pattern or repeating motif" default="[geometric pattern or repeating motif]"} suggesting the brand's signature pattern

Pattern execution requirements:
- The {argument name="brand core elements" default="[brand core elements]"} should be CLEARLY RECOGNIZABLE at first glance - not abstract hints
- {argument name="pattern elements" default="[Pattern elements]"} are arranged in organized repeating pattern like {argument name="brand" default="[brand]"} signature print/canvas

Mirror Selfie in Fitting Room with Tennis Skirt

ecommerce-main-image

A detailed JSON prompt for Gemini Nano Banana Pro to generate a candid, social media-style mirror selfie of a young woman in a clothing store fitting room. The prompt specifies her platinum blonde hair, attire (black bikini top and ultra-mini white pleated tennis skirt with a visible price tag), pose (tugging at the skirt, one leg popped), and the environment (bright fluorescent lighting, background mirrors, and bare feet on the carpet).

展开提示词全文
{
  "subject": {
    "description": "Young woman in fitting room trying on ultra mini tennis skirt, wearing her own bikini top, mirror selfie",
    "age": "early 20s",
    "hair": {
      "color": "platinum blonde",
      "style": "loose waves, soft fitting room texture",
      "length": "long, falling over shoulders"
    },
    "body": {
      "build": "petite frame, toned legs, feminine curves",
      "skin": "pale, porcelain, glowing under bright lights"
    },
    "expression": {
      "look": "innocent doe eyes looking at phone in mirror",
      "mood": "playful uncertainty, 'does this fit?' energy",
      "mouth": "soft pout, lips slightly parted"
    }
  },

  "clothing": {
    "top": {
      "item": "black triangle bikini top",
      "style": "string tie, minimal coverage",
      "context": "her own top she wore to the store",
      "fit": "snug, enhancing chest"
    },
    "bottom": {
      "item": "white pleated tennis skirt",
      "style": "ultra mini, high waisted",
      "fit": "dangerously short, high on thigh",
      "detail": "retail price tag hanging visibly from waistband"
    },
    "feet": {
      "state": "bare feet",
      "context": "standing on carpeted fitting room floor"
    }
  },

  "pose": {
    "type": "mirror selfie",
    "arm_action": {
      "right": "holding iPhone up for photo",
      "left": "tugging slightly at the skirt hem downwards"
    },
    "stance": {
      "legs": "one leg popped forward, weight on back hip",
      "posture": "slightly arched back to accentuate curves"
    }
  },

  "environment": {
    "location": "clothing store fitting room",
    "lighting": "bright vertical fluorescent strips on mirror sides",
    "background": {
      "mirrors": "multiple angles visible in background mirrors",
      "clutter": "her own clothes (denim shorts, bag) on the bench behind",
      "floor": "beige commercial carpet"
    }
  },

  "technical": {
    "aspect_ratio": "9:16",
    "framing": "eye-level, medium shot capturing head to mid-thigh",
    "style": "candid, social media aesthetic, sharp focus on subject"
  }
}

Two-Phase Prompt for Absurd Product Fusion Concepts

ecommerce-main-image

A two-phase prompt for Nano Banana Pro designed to fuse everyday objects into new, absurd yet plausible inventions. Phase 1 analyzes the function, material, and cultural DNA of three reference objects. Phase 2 generates a 3x3 grid of photorealistic 3D concepts in a Dieter Rams/Dyson style, complete with minimalist labels.

展开提示词全文
PHASE 1 PROMPT: Analyze these [{argument name="object 1" default="OBJECT 1"}], [{argument name="object 2" default="OBJECT 2"}], [{argument name="object 3" default="OBJECT 3"}] images. Extract: 1) Primary function, 2) Material DNA (metal/plastic ratio, texture), 3) Ergonomic signature (grip angle, weight distribution), 4) Mechanical principle, 5) Cultural association. Output as structured JSON.

PHASE 2 PROMPT: Create a 3x3 grid of photorealistic product renderings on a pristine white background. Each cell must be a hyper-realistic 3D concept that fuses elements from the previous analysis. For each invention, add minimalist navy blue text in Helvetica Neue: [Invention Name], [Core Mechanism], [Use Case]. Style: Dieter Rams meets Dyson - clean, technical, premium. Include subtle shadows and a 1px gold border on each cell. Make inventions absurd yet plausible (e.g., "Acoustic Peeler: uses sound waves to strip vegetables").

Genspark Product Photography Prompt Template

ecommerce-main-image

A reusable prompt template for generating premium product photography images using Genspark, which offers unlimited access to Nano Banana Pro. The prompt focuses on studio setting, soft lighting, shallow depth of field, and a minimalist aesthetic.

展开提示词全文
Studio photography of [{argument name="product name" default="product name"}]. Placed on a [{argument name="background" default="background"}] surrounded by soft shadows and a gradient background. Premium product photo with high-key lighting, shallow depth of field, and extremely sharp focus on the subject. Shot with a digital SLR camera. Minimalist aesthetic, delicate reflections, commercial lighting setup.

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。

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。

AI Image Prompts · 绘画提示词库

官方标注免费

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

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。