← 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 2105 条

Ultra Photorealistic Selfie Prompt for Sydney Sweeney

profile-avatar

A highly detailed, structured JSON-style prompt for Nano Banana Pro designed to generate an ultra-photorealistic, dreamy close-up selfie of Sydney Sweeney, specifying camera settings, lighting, expression, and appearance details.

展开提示词全文
{
  "prompt_metadata": {
    "version": "4.0",
    "style": "ultra photorealistic dreamy close-up selfie",
    "aspect_ratio": "9:16",
    "resolution": "8k RAW"
  },

  "reference_rules": {
    "face_source": "use uploaded reference image as exact identity",
    "face_integrity": "no changes to facial structure, eyes, lips, proportions",
    "hair_color_rule": "hair color must match reference exactly"
  },

  "subject": {
    "pose": {
      "type": "extreme close-up selfie",
      "orientation": "camera rotated slightly clockwise",
      "head_position": "tilted sideways, resting on right hand",
      "face_angle": "3/4 angle, slightly downward toward camera",
      "distance": "very close to lens, filling most of frame"
    },

    "expression": {
      "type": "soft closed-lip pout",
      "eyes": "relaxed, slightly heavy eyelids",
      "emotion": "calm, intimate, dreamy"
    },

    "appearance": {
      "skin": {
        "texture": "natural pores but softened by light diffusion",
        "tone": "warm flushed cheeks and nose",
        "finish": "glowy with slight highlight bloom"
      },

      "hair": {
        "style": "loose, slightly messy",
        "placement": "falling across forehead, partially covering one eye",
        "strands": "fine strands visible across face",
        "color": "same as reference"
      }
    },

    "hand_pose": {
      "position": "right hand supporting chin",
      "fingers": "loosely curled, soft natural positioning",
      "contact": "cheek gently pressed into palm"
    }
  },

  "environment": {
    "setting": "indoor bedroom",

    "background": {
      "bed_frame": "light warm wood headboard behind subject",
      "wall": "cream or beige wall",
      "decor": "small framed artwork slightly visible above"
    }
  },

  "lighting": {
    "primary_source": {
      "type": "warm indoor ambient light",
      "position": "above and slightly behind camera",
      "color_temperature": "3200K warm yellow",
      "intensity": "medium soft"
    },

    "secondary_effect": {
      "type": "diffusion bloom",
      "effect": "light wraps around face, softens edges",
      "highlight_behavior": "slight overexposure on forehead, cheeks, nose"
    },

    "shadow_behavior": {
      "style": "very soft, almost no harsh shadows",
      "contrast": "low contrast, hazy look"
    }
  },

  "camera": {
    "device": "smartphone front camera",
    "lens": "26mm equivalent",
    "aperture_simulation": "f/2.0",
    "focus": "sharp on eyes, soft falloff on edges",
    "angle": "handheld, slightly tilted frame",
    "distance": "arm-length but zoomed/cropped for tight face framing"
  },

  "color_grading": {
    "tone": "warm golden",
    "contrast": "low",
    "saturation": "slightly muted skin tones",
    "effect": "dreamy haze, slight glow filter"
  },

  "render_constraints": {
    "must_have": [
      "soft glow effect",
      "

3D Chibi Character Transformation Prompt for Nano Banana 2

profile-avatar

A detailed prompt for Nano Banana 2 to transform a subject from a reference image into a cute 3D anime chibi character. The prompt emphasizes preserving the subject's likeness, race, and skin tone while applying classic chibi proportions, rendering it as a high-quality, glossy 3D model with vibrant colors and a premium collectible aesthetic.

展开提示词全文
Transform the subjects into a cute anime chibi characters rendered as a clean 3D model, while preserving the subject's original race, ethnicity, skin tone, and recognizable facial likeness, including hairstyle, hair color, and clothing colors. The character must remain clearly the same person with the same skin tone and ethnic appearance as the original input, translated into a chibi form. The character should have classic chibi proportions: a very large head, small torso, simplified cute facial features, and large expressive anime-style eyes. Translate the subject's facial features into a chibi style while keeping their recognizable likeness. Frame the character from the torso up, with the camera slightly closer and cropped in, so the upper body and head fill most of the frame. The character should be looking directly into the camera, with an energetic and expressive anime-style pose. Use a slightly dynamic camera angle, such as a subtle tilt or off-axis perspective, similar to modern anime promotional artwork. The camera can be slightly lower than the character's face, creating a playful upward Tr angle that makes the character feel lively and engaging. Render the character as a high-quality glossy 3D anime-style model, similar to a polished Blender render or premium collectible figure. The materials should be smooth, reflective, and toy-like, with strong highlights, soft reflections, and shiny surfaces. The overall look should clearly resemble a 3D rendered character, not a flat 2D illustration. Use bright anime-style shading with vibrant but polished colors, giving the character a premium anime aesthetic. Place the character in front of a colorful gradient background. Behind the character, include a small number of large, translucent decorative shapes such as bubbles, stars, or rounded forms. These shapes should appear semi-transparent, glossy, and subtle, similar to clear acrylic or glass, softly floating in the background without drawing attention away from the character. The background elements should remain light, minimal, and softly visible, acting as stylish accents rather than prominent objects. The final image should look like a dynamic, glossy 3D chibi character render, with polished materials, vibrant colors, and a premium collectible aesthetic,while clearly maintaining the subject's original race, skin tone, and recognizable identity.

Black-and-White Fine Art Studio Portrait Prompt

profile-avatar

A detailed prompt for generating a black-and-white fine art studio portrait using Nano Banana 2. The prompt specifies a seated subject (based on an uploaded person), relaxed pose, slightly messy hair, natural skin tones, a clean white shirt, a calm expression, cinematic angle, minimalist grey backdrop, soft diffused cinematic lighting, and ultra-realistic, high-contrast textures in 8K resolution.

展开提示词全文
Black-and-white fine art studio portrait of the uploaded person, seated with both hands resting naturally in front, relaxed and composed. Slightly messy hair softly framing the face, natural smooth skin tones, wearing a clean white shirt. Calm, thoughtful expression while looking directly at the camera. Shot from a subtle cinematic angle (not strict eye level). Minimalist studio setup with a plain grey backdrop, soft diffused cinematic lighting creating gentle shadows and depth. Elegant, sophisticated mood, ultra-realistic textures, sharp facial details, professional photography quality, high contrast, smooth tonal gradation, 8K resolution, no noise, no pixel loss.

T-Shirt : Black-and-white fine art studio portrait of the uploaded person, seated with both hands resting naturally in front, relaxed and composed. Slightly messy hair softly framing the face, natural smooth skin tones, wearing a clean white oversized Tshirt. Calm, thoughtful expression while looking directly at the camera. Shot from a subtle cinematic angle (not strict eye level). Minimalist studio setup with a plain grey backdrop, soft diffused cinematic lighting creating gentle shadows and depth. Elegant, sophisticated mood, ultra-realistic textures, sharp facial details, professional photography quality, high contrast, smooth tonal gradation, 8K resolution, no noise, no pixel loss.

3D Cartoon Caricature Portrait

profile-avatar

A prompt for generating a highly detailed, semi-realistic 3D cartoon caricature portrait of an excited bearded man wearing a maroon Nike hoodie and black oval sunglasses, emphasizing bold outlines, glossy eyes, and detailed hair texture.

展开提示词全文
A highly detailed cartoon caricature portrait of an excited bearded man with thick brown hair and large black oval sunglasses, with natural smile, brown eyes, and a joyful expression. He has a trimmed brown beard and mustacheb and wears a {argument name="clothing item" default="nike maroon nike hoodie"} over a white shirt. Around his neck is a silver dog tag necklace engraved with the initials "{argument name="initials" default="X"}". The style is vibrant, semi-realistic caricature illustration with bold outlines, glossy eyes, detailed hair and beard texture, and dramatic facial exaggeration. Clean light gray background, centered portrait, studio lighting, ultra sharp, digital illustration, 4k, professional cartoon character design.

Character Description for Rock Music Fan

profile-avatar

A detailed character description of a female Japanese rock fan who prefers listening from the back with a beer, enjoys music, smoking, and motorcycles, and wants to be called 'cool' rather than 'cute.' This description was used as a prompt for both Nano Banana Pro and Nano Banana 2.

展开提示词全文
A woman who likes Japanese rock music. The type who quietly listens at the back of the venue with a beer in hand. Hobbies include music, smoking, and motorcycles. She wants to be called 'cool' rather than 'cute.' She is surprisingly sweet in front of the person she likes.

Professional ID Photo Generation Prompt for Nano Banana 2

profile-avatar

A prompt designed to generate professional, high-quality ID photos suitable for platforms like LinkedIn, using the Nano Banana 2 model. It specifies a smart, professional style, soft natural lighting, a subtle gray-to-white gradient background, and focuses on clear facial features, good skin texture, and a relaxed, confident expression. The output is intended to be a refined professional portrait.

展开提示词全文
Smart professional ID photo, overall style referencing American ID photos, with the subject size being moderate. Use a subtle gray-to-white gradient photographic background, soft and natural lighting, highlighting realistic skin tone and layering. The image should be clear and high-quality, with the face in focus, transparent and good skin texture, and a comfortable, normal head-to-shoulder ratio. The overall temperament is modern and elegant, with a relaxed, naturally confident expression and bright, expressive eyes. Medium shot portrait, moderate subject size, centered subject, low contrast, presenting the refined quality of professional portrait photography. Suitable for business and professional headshots. Aspect ratio 3:4

3D CGI Big-Head Cartoon of Harry Potter

profile-avatar

A prompt for generating a high-quality stylized 3D CGI big-head cartoon render of Harry Potter in a playful Pixar-style caricature, emphasizing exaggerated features, smooth polished materials, cinematic lighting, and a minimalist bright magical red background.

展开提示词全文
High-quality stylized 3D CGI big-head cartoon render of Harry Potter, square profile picture composition, oversized head and small shoulders in a playful Pixar-style caricature. Head and facial features exaggerated while keeping him instantly recognizable: messy dark hair, round glasses, lightning-shaped scar on the forehead, curious and brave expression. Portrait framing focused on the large head with a small portion of the shoulders visible, wearing Hogwarts robe collar and tie. Smooth polished materials, glossy animated film shading with refined Pixar-style character rendering. Soft cinematic lighting creating gentle highlights on the glasses. Minimalist bright magical red background inspired by Hogwarts colors, clean composition optimized for social media profile pictures, ultra-detailed, no text, no watermark.

Transform Photo into Modern Anime Vector Illustration

profile-avatar

A prompt requiring a reference photo to transform a person's facial identity into a modern anime vector illustration, changing the pose to a dramatic side profile with closed eyes and adding black over-ear headphones, set against a minimalist bold red background with high contrast shadows.

展开提示词全文
Transform the person from the reference photo into a modern anime vector illustration while preserving the exact facial identity.
Change the pose to a dramatic side profile with the head tilted upward, eyes closed, relaxed emotional expression as if listening to music.
Add black over-ear headphones.
Stylized messy anime hair, smooth skin shading, sharp jawline, clean cel shading.
Minimalist bold red background, dramatic cinematic lighting, high contrast shadows.Transform the person from the reference photo into a modern anime vector illustration while preserving the exact facial identity.
Change the pose to a dramatic side profile with the head tilted upward, eyes closed, relaxed emotional expression as if listening to music.
Add black over-ear headphones.
Stylized messy anime hair, smooth skin shading, sharp jawline, clean cel shading.
Minimalist bold red background, dramatic cinematic lighting, high contrast shadows.

Highly Realistic Smartphone Selfie of a Woman

profile-avatar

A detailed, multi-section prompt for Nano Banana 2 to generate a highly realistic smartphone selfie of a Caucasian woman, specifying composition, lighting, physical features, clothing, and background details for a natural, high-clarity image.

展开提示词全文
A highly realistic smartphone selfie photo of a Caucasian woman in her mid-20s, captured in a vertical 4:5 aspect ratio. The camera is held at a slightly high angle with an arm-extended perspective, using a wide-angle smartphone lens (around 24–28mm). The framing shows her from head to mid-torso, with sharp focus on her face and a slightly softer background.

She has fair-light skin with warm-neutral undertones, smooth natural texture, and a subtle glow from soft daylight. Her cheeks have a gentle pink flush. Her face is oval-shaped with a soft jawline, medium forehead, slightly full cheeks, and a rounded chin.

Her eyes are light green with hints of hazel, almond-shaped, with long defined lashes and mascara. Her eyebrows are light brown, medium thickness, softly arched, and well-groomed. She has a straight nose with a slightly rounded tip, proportional in size. Her lips are full (especially the lower lip), natural pink in color, with a slightly glossy hydrated look. She is smiling naturally with visible straight white teeth, giving a friendly, confident, relaxed expression, with slightly squinting eyes due to the smile.

Her hair is shoulder-length, dark blonde with lighter blonde highlights (balayage), styled in soft waves with a center part, naturally framing her face.

She has a slim to average build, visible from head to mid-torso, with a relaxed posture slightly angled toward the camera.

She is wearing a fitted white short-sleeve t-shirt made of soft cotton, slightly stretchy, tightly fitted. The shirt has bold red block serif text in all caps on the chest reading: “YES THEY ARE REAL”. She also wears a thin gold chain necklace with a small subtle pendant.

The lighting is natural daylight coming from the front-left, soft and diffused, creating even illumination on her face with gentle shadows and natural highlights.

The background is an indoor home setting, likely a kitchen/living area. Visible elements include a light-colored kitchen counter, white modern bar stools, a dark refrigerator (black or stainless steel) with small items or magnets attached, muted green/gray walls, light brown wooden door trim, and medium warm brown wooden flooring. The background has moderate depth and is slightly blurred.

The overall color palette includes white, red, natural skin tones, green-gray walls, wood browns, and dark appliance tones. The image style is highly realistic with minimal editing, high clarity, low noise, and natural contrast.

Professional Photo Restoration and Upscaling

profile-avatar

A prompt for nano-banana-pro to restore an old photo into a professional, high-quality portrait using advanced upscaling algorithms, ensuring natural results and retention of exact facial features.

展开提示词全文
Restore this old photo into professional portrait of DLSR - quality colour and detail, using an advanced upscaling algorithm comparable to the results from canon EOS R6 II. Ensure the restored the image looks natural, retains exact facial features, has great clarity......

Raw iPhone Mirror Selfie Realism

profile-avatar

A detailed JSON prompt structure for Nano Banana 2 focusing on generating an ultra-photorealistic, dark influencer-style mirror selfie with specific camera settings, character details (tattoos, body shape), and scene imperfections to achieve a raw, handheld realism aesthetic.

展开提示词全文
{
  "meta": {
    "camera": "iPhone 17 Pro Max",
    "lens": "24mm wide",
    "aspect_ratio": "4:3",
    "style": "raw iphone mirror snap, zero polish, dark influencer realism",
    "quality": "ultra photorealistic",
    "color_science": "authentic iphone HDR, deep contrast, natural skin texture"
  },

  "character_lock": {
    "face": "pale complexion, alternative beauty, confident playful expression",
    "body": "extreme hourglass silhouette, very narrow waist, significant bust volume, pronounced hips and glute shape",
    "hair": "dark hair slicked back, slightly wet look",
    "tattoos": "large neo-traditional chest piece across collarbones, symmetrical mandala abdomen tattoo with rabbit motif, full sleeve blackwork tattoos"
  },

  "scene": {
    "location": "dim apartment bathroom",
    "time": "1:14 AM",
    "lighting_temperature": "warm overhead bulb 3000K + subtle phone screen glow",
    "atmosphere": "quiet, after-going-out energy"
  },

  "composition": {
    "angle": "mirror selfie, slightly low and tilted",
    "framing": "mid-thigh to head, imperfect crop",
    "imperfections": [
      "mirror water spots",
      "counter clutter slightly visible",
      "minor motion blur from hand movement",
      "uneven lighting falloff"
    ]
  },

  "subject": {
    "pose": "standing relaxed with hip popped, one shoulder forward",
    "expression": "playful confident smirk, slight squint, tongue barely peeking",
    "outfit": {
      "top": "tight black cropped tank, thin cotton, no bra, natural fabric tension",
      "bottom": "low-rise black lounge shorts, hugging hips"
    },
    "details": "tattoos fully visible and undistorted, visible oblique definition, natural skin texture with subtle sheen"
  },

  "photography": {
    "depth_of_field": "natural smartphone depth, background slightly soft",
    "grain": "light low-light grain retained",
    "handheld_realism": true,
    "not_overprocessed": true
  }
}

Cinematic Vector Fashion Portrait

profile-avatar

A highly detailed, structured prompt for Nano Banana Pro to generate a cinematic fashion-editorial portrait in a clean, flat vector illustration style, using a reference photo for identity and specific color blocking for lighting.

展开提示词全文
Create A Cinematic Fashion-editorial Portrait Of A Person Using Their Photo As Identity Reference, Rendered In A Clean Flat Vector Illustration Style With Smooth Geometric Shapes, Crisp Outlines, And Simplified Facial Geometry, Framing Centered Head-and-shoulders Composition With Symmetrical Alignment, Subject Facing Forward With Relaxed Posture And A Subtle Friendly Smile Styling Minimal Modern Casual With A Simple Dark Crew-neck Shirt And Thin Round Eyeglasses, Paired With Small Wireless Earbuds As Understated Accessories, Background A Flat Warm Beige Studio Tone (#ebdbc6) With No Gradients Or Textures To Maintain A Graphic Design Aesthetic, Lighting Implied Through Stylized Flat Color Blocking Rather Than Realistic Shading, With A Gentle Warm Highlight Tone (-4800k) On One Side Of The Face And A Slightly Darker Tonal Block On The Opposite Side To Create Minimal Dimensional Contrast, Reflections Extremely Subtle On The Glasses Lenses, Lens Simulation Conceptually Equivalent To A Neutral Portrait Crop (50-70mm) Translated Into A Flat Graphic Design Perspective With No Depth Blur; Texture Perfectly Smooth Matte Vector Surfaces With Solid Color Fills And Clean Edges, Color Palette Warm Skin Tones, Muted Peach And Coral Accents, Charcoal Black Hair And Clothing, And Soft Beige Background, Overall Mood Friendly, Approachable, Modern Tech-avatar Aesthetic Suitable For Product Design Or Ui Identity Illustration Editorial, Modern, High-detail, Cinematic Composition, Ultra-realistic Textures, Professional Fashion Photography Style
上一页 77 / 176 下一页

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。