prompts.

A stunning mermaid bursts upward from the ocean at high speed, water exploding around her in slow motion. As she rises into the air, the camera begins an orbiting cinematic move around her. Her shimmering scales glow in the sunlight while her body twists gracefully. Mid-air, she transforms seamlessly into a same-size dragonfly — wings unfolding, iridescent and hyper-detailed. The transformation is fluid and dramatic, Hollywood-style. The camera completes its orbit as the newly formed dragonfly catches the light, then darts away into the sky with elegant speed. Ultra-realistic, breathtaking, highly detailed, cinematic lighting, dramatic atmosphere.

真人实拍风格，一位美丽的留着黑色波浪长发的少女，穿着粉色露脐装和瑜伽裤要求性感，皮肤白皙，正随着Future House风格的DJ舞曲俏皮地舞动，舞蹈动作包含俏皮的摆胯、手臂波浪步和定点pose，且与音乐节拍完全吻合；镜头会跟着音乐节拍前后推拉运镜，背景是在卧室里面，顶部有柔和的聚光灯打下来照亮少女，整体光影柔和、氛围感强，画面比例为9:16。排除：模糊，低清，噪点，水印，文字，logo，扭曲，变形，五官崩坏，动作僵硬，画面抖动，比例失调。

真人实拍风格，一位美丽的留着黑色波浪长发的少女，穿着白色露脐装和jk短裙，要求性感，皮肤白皙，正随着《胜利之舞》DJ舞曲俏皮地舞动，舞蹈动作包含俏皮的摆胯、手臂波浪步和定点pose，且与音乐节拍完全吻合；镜头会跟着音乐节拍前后推拉运镜，背景是卧室里面，顶部有柔和的聚光灯打下来照亮少女，整体光影柔和、氛围感强，画面比例为9:16。排除：模糊，低清，噪点，水印，文字，logo，扭曲，变形，五官崩坏，动作僵硬，画面抖动，比例失调。模型 2.0，比例 9:16，时长 10s。

超高清纯欲风美女变装短视频，电影级柔焦柔光，清透磨皮质感，肤色白皙粉嫩，画面干净温柔，细节细腻。 场景是温馨卧室，浅色系温柔背景，暖黄柔和光影，氛围感拉满。前期是慵懒居家造型，宽松软糯上衣，自然素颜感淡妆，头发松散温柔，表情干净无辜，动作松弛慵懒，低头浅笑、轻撩头发，纯欲感拉满。 随着音乐卡点完成丝滑变装，转场自然柔和，光线变得更温柔朦胧。变装后是精致纯欲女神造型，清透妆容，眼妆淡粉细闪，唇色水润嫩粉，肤色白皙，发型温柔卷曲，碎发精致，身穿修身温柔小吊带，搭配精致细巧项链、耳饰，气质温柔又撩人。 人物姿态优雅松弛，眼神干净又带点小魅惑，动作轻柔舒缓，氛围感十足。镜头以近景特写为主，运镜平稳温柔，突出前后气质对比。 整体色调暖粉温柔，低饱和高级感，动态自然流畅，无崩坏、无扭曲，画质细腻高清，节奏舒缓卡点，纯欲天花板，温柔又撩人，完美呈现高级纯欲变装效果。

## 📸 Examples Here are some scientific figures generated using FigForge: ### Sample Input - Neural Architecture  ### [LiveSearchBench](https://arxiv.org/abs/2511.01409)  ### [ReSo](https://arxiv.org/abs/2503.02390)  ### [VIKI-R](https://arxiv.org/abs/2506.09049)  > All figures are generated with clean conference-style design, featuring flat aesthetics, consistent line weights, and professional color palettes. --- ## 📖 How It Works ### Step 1: MODULE LIST Generation (GPT-5) The GPT-5 model analyzes your scientific text and creates a structured MODULE LIST that breaks down your architecture into: 1. **Input(s)**: Data sources and preprocessing 2. **Preprocessing/Encoding/Embedding**: Feature extraction layers 3. **Core Architecture/Stages/Blocks**: Main model components in sequence 4. **Special Mechanisms**: Attention, memory, routing, etc. 5. **Output Head**: Final prediction layers ### Step 2: Figure Generation (Gemini-2.5-flash-image) Using the MODULE LIST as a guide, nano banana generates a clean, professional figure following these design principles: - ✅ Flat, clean conference style (no gradients, shadows) - ✅ Consistent thin line weights - ✅ Professional pastel color palette - ✅ Rounded rectangles for module blocks - ✅ Clear arrows indicating data flow - ✅ Concise labels (no long sentences) - ✅ Pure white background with clean spacing ## 📁 Project Structure

for MODULE LIST │ └── step2_figure_generation.txt # Prompt for figure generation ├── examples/ │ └── sample_input.txt # Sample scientific text ├── outputs/ # Generated figures (auto-created) └── README.md # This file

Use Nano Banana Pro — cinematic photograph of the Charles Bridge in Prague at golden hour, must be architecturally accurate

"I need a 10s product launch video for my SaaS app, dark mode UI, minimal aesthetic" → SaaS Launch skill generates a complete prompt with camera, lighting, sound, timing "Create a 6s before-after for my client's brand redesign" → Before-After skill builds the contrast arc: cold→warm lighting, gray→saturated color, chaos→order "Give me a viral TikTok hook for a faceless finance channel" → Viral Hook + Faceless Channel skills combine for a scroll-stopping 2-second opener "Turn this podcast clip

Turn this podcast clip into a cinematic video for Reels

doesn't work Most people write 2-line prompts and get generic output. Seedance 2.0 is a precision engine — it responds to specific camera language, lighting terminology, and timing breakdowns. Vague input = vague output. These skills solve that by encoding three things into every prompt: 1. **The physics of attention** — the first 2 seconds determine if 50% of viewers bounce or 90% stay. Every prompt opens with a hook engineered for retention. 2. **Cinematic vocabulary** — Seedance 2.0 understands

3. **Sound-visual sync** — every audio element (bass hit, whoosh, ASMR texture, silence) is timestamped to a visual beat. Sound is 50% of the hook. The result: prompts that produce studio-quality video instead of generic AI slop. --- ## Install ```bash # Clone into your Claude skills folder cd ~/.claude/skills git clone https://github.com/rediumvex/ai-video-generator-claude.git ai-video-generator ``` Or if you use gstack: ```bash cd ~/.claude/skills/gstack git clone https://github.com/rediumvex/ai-video-generator-claude.git ai-video-generator ``` Restart Claude Code. Each skill inside `skills/` is now available. You can also use the interactive installer: ```bash cd ai-video-generator python install.py # Pick which skills to install python install.py --all # Install everything python install.py --list # Preview available skills ``` --- ## Usage ```

→ SaaS Launch skill generates a complete prompt with camera, lighting, sound, timing

【风格】极致第一人称女友视角(Ultimate Girlfriend POV)，手持自拍VLOG，竖屏(9:16)，胶片感滤镜(Film Grain)，自然光(夕阳黄金时刻)，镜头有自然的晃动和调整。 【时长】15秒 【主角】台湾女生，长发微卷，穿着温柔的针织开衫，妆容清透，说话语气软糯、带点撒娇和抱怨的口音(台湾腔)。 【场景】台北淡水河岸边，背景是金色的夕阳、波光粼粼的河面和远处的情人桥。 [00:00-00:05] 镜头1：赶路与抱怨(The Rushed Intro)。 画面：镜头晃动得比较厉害，因为她在快走。她一边走一边回头看镜头（看你），眉头微皱，假装生气。 动作：一只手举着手机，另一只手还要压住被风吹乱的头发。 【台词/口型】（软糯抱怨音）："欸你走快一点啦！太阳快要下山了捏！都是你拖拖拉拉的，等一下拍不到好看的照片我要生气喔！" [00:00-00:10] 镜头2：绝美景色分享(Sharing the View)。 画面：她停下脚步，把镜头从自己脸上移开，转向河对岸绝美的金色夕阳（Golden Hour）。 动作：镜头展示了2秒景色，然后马上又转回对准她的脸。她的脸被夕阳照得金灿灿的，露出超级开心的笑容，眼睛眯成一条缝。 【台词/口型】（惊叹音）："哇～你快看！有没有超美！这个光线真的绝了，随便拍都好看！" 提示词1-【名称】 [00:10-00:15] 镜头3：互动与结尾(The Interaction)。 画面：她手里突然多了一支淡水特有的超高霜冰淇淋。 动作：她先把霜冰淇淋举到镜头前（喂你），然后自己快速舔了一口，鼻尖不小心沾到一点冰淇淋。她对着镜头傻笑。 【台词/口型】（调皮音）："呐，第一口先给你吃。快点，要融化了啦！嘻嘻，今天的约会算你及格！"

【风格】抖音直播带货风，极速语速(Rap-like Speed)，情绪极其亢奋。 【时长】15秒 【主角】激情带货主播（网红美女），面前摆着手机平板，背景是杂乱但真实的高端直播间。 [00:00-00:05] 镜头1：痛点爆破(The Hook)。 画面：分屏显示。左边是模糊扭曲的PPT式AI视频（打上大大的红叉），右边是电影级质感的Seedance生成视频。 动作：主播猛拍桌子，把脸凑近镜头，表情痛心疾首。 【语速极快/喊麦感】："别再用那些全网都在推的垃圾模型了！生成的视频脸都歪了还能看吗？家人们，今天我把Seedance 2.0的价格给你们打下来了！" [00:05-00:10] 镜头2：实操炸场(The Demo)。 画面：特写手机屏幕。主播输入："赛博朋克雨夜，少女回眸"。按下生成键，进度条瞬间跑满（0.5秒）。 特效：屏幕上瞬间炸出（粒子光影弹出）一段4K、60帧、毛发毕现的绝美视频。 动作：主播指着屏幕疯狂敲击，口沫横飞。 【语速极快】："看这个速度！看这个光影！不用显卡，不用排队！这就是Seedance 2.0！好莱坞导演都在用！" [00:10-00:15] 镜头3：限时逼单(The Call to Action)。 画面：主播身后出现巨大的倒计时（只剩3秒），屏幕下方出现疯狂闪烁的"免费试用"小黄车图标。 动作：主播双手合十拜托，然后疯狂指着屏幕下方。 【语速极快】："全网首发！仅限前100名免费试用！链接就在左下角！手慢无！给我冲！3、2、1，上链接！"

【风格】国产神豪爽剧（Viral CEO Drama），竖屏构图（Portrait Mode），高饱和度滤镜，极致的面部特写，情绪大开大合。 【时长】15秒 【角色】被羞辱的新郎（穿着廉价西装，眼神隐忍不发）VS 刻薄丈母娘（满身珠宝，一脸嫌弃）。 [00:00-00:05] 镜头1：极致羞辱（Humiliation）。 画面：豪华婚礼现场。丈母娘当着众人的面，将一纸"退婚书"狠狠摔在男主胸口，周围宾客哄堂大笑。 动作：丈母娘手指戳着男主的额头。 【对白口型指导】"没车没房还想娶我女儿？拿着这一百块滚出去！" [00:05-00:10] 镜头2：风云突变（The Turn）。 画面：豪华婚礼现场。丈母娘当着众人的面，将一纸"退婚书"狠狠摔在男主胸口，周围宾客哄堂大笑。 动作：男主突然冷笑一声，撕碎退婚书。此时，巨大的螺旋桨声（音效感）盖过全场，狂风吹乱了丈母娘的发型。男主整理了一下衣领，气场瞬间变得霸气侧漏。 【对白口型指导】"这婚，可是你们要退的。" [00:10-00:15] 镜头3：神豪降临（The Reveal）。 画面：大门被撞开，两排黑衣保镖冲进来，单膝跪地铺上红地毯。一位老管家颤抖着捧着一件黄袍（或至尊黑卡）跑到男主面前深深鞠躬。丈母娘吓得瘫坐在地上，瞳孔地震。 动作：老管家鞠躬。 【对白口型指导】老管家喊："恭迎龙王（少爷）归位！家族资产已解冻！"

【风格】伪纪录片（Vlog Style），超写实主义，固定机位实拍感，自然光，带有一点点悬疑喜剧色彩。 【时长】15秒 【主角】一个普通的年轻人美女，在自家卫生间洗漱台前。 [00:00-00:06] 镜头1：日常铺垫（Normalcy）。 场景：普通的卫生间大镜子前。 动作：主角正在对着镜子刷牙，满嘴泡沫。她一边刷牙一边对着镜子做各种搞怪的鬼脸（挤眉弄眼）。 关键细节：此时镜子里的倒影完全正常，动作同步。 [00:06-00:11] 镜头2：BUG出现（The Glitch）。 动作：主角刷完牙，低头吐掉泡沫，然后转身准备离开卫生间。 高能时刻（核心爆点）：就在主角真身已经转身离开镜子画面范围的时候，镜子里的那个"倒影"竟然没有动！那个"倒影"依然保持着刷牙的姿势，甚至还坏笑着冲着镜头挑了一下眉毛，停留了整整2秒钟，才突然惊慌失措地"快进"追上本体的动作消失。 导演备注：要做出极其真实的"网络延迟"感，倒影有独立意识的感觉。 [00:11-00:15] 镜头3：喜剧回马枪（The Punchline）。 动作：已经走到门口的主角似乎感觉到了不对劲，猛地回头看向镜子。 结果：镜子此时已经完全恢复正常，空空荡荡，只照出对面的墙壁。主角一脸懵逼地挠头，对着镜头露出怀疑人生的表情。画面在主角的懵逼脸中定格（喜剧效果）。

[00:00-00:10] 镜头2：绝美景色分享(Sharing the View)。 画面：她停下脚步，把镜头从自己脸上移开，转向河对岸绝美的金色夕阳（Golden Hour）。 动作：镜头展示了2秒景色，然后马上又转回对准她的脸。她的脸被夕阳照得金灿灿的，露出超级开心的笑容，眼睛眯成一条缝。 【台词/口型】（惊叹音）：

生成每张幻灯片图片 3. **版本管理**：每次生成结果作为新版本保存，支持切换与回溯 ### 主要 API - 演示文稿：

[NAME] = Your name [SUBJECT] = The person shown in the uploaded reference image Make a symbolic portrait reinterpretation for artistic editorial use, using a medium-close framing at eye level with a calm, centered composition, set within an abstract environment derived from the etymological and cultural meaning of [NAME], with [SUBJECT] placed as the sole figure in the midground, preserving the facial structure, facial proportions, expression, and overall likeness of [SUBJECT] exactly as in the reference image, allowing no alteration, stylization, or reinterpretation of the face under any circumstance, avoiding any reuse, reference, or derivation of clothing, accessories, or styling from the reference image, and instead designing a new outfit that supports the mood and symbolism of [NAME] without echoing the original wardrobe, translating the meaning of [NAME] into visual elements such as light behavior, color palette, atmosphere, natural forces, or metaphoric forms surrounding [SUBJECT], reflecting the name’s meaning through mood and symbolism rather than literal text, letters, or icons, using lighting direction, intensity, and color temperature to reinforce the emotional essence of [NAME], integrating symbolic elements so they feel physically and spatially connected to [SUBJECT] rather than decorative, keeping the background restrained and uncluttered with no readable text, and ensuring strong visual impact, coherence, and stylistic integrity throughout the composition.

Make a photorealistic studio head portrait of the person from the provided reference photo, used strictly as identity anchor, captured at eye level with natural facial perspective, framed as a clean studio photograph with only the head and upper shoulders visible, set against a professional studio background with soft neutral tones, showing the subject centered and facing forward with a calm direct gaze, with a single hand gently closing a metal zipper over the lips, ensuring the lips and zipper merge naturally and convincingly without distortion, rendered with realistic skin texture and fine detail, lit with soft controlled studio lighting to create an elegant, cinematic, and emotionally restrained image.

></a> ## 1. Portraits & Identity ### 1.1. Name Meaning Portrait Turn Your Name’s Meaning Into an Iconic Portrait **Prompt:** ```text [NAME] = Your name [SUBJECT] = The person shown in the uploaded reference image Make a symbolic portrait reinterpretation for artistic editorial use, using a medium-close framing at eye level with a calm, centered composition, set within an abstract environment derived from the etymological and cultural meaning of [NAME], with [SUBJECT] placed as the sole figure in the midground, preserving the facial structure, facial proportions, expression, and overall likeness of [SUBJECT] exactly as in the reference image, allowing no alteration, stylization, or reinterpretation of the face under any circumstance, avoiding any reuse, reference, or derivation of clothing, accessories, or styling from the reference image, and instead designing a new outfit that supports the mood and symbolism of [NAME] without echoing the original wardrobe, translating the meaning of [NAME] into visual elements such as light behavior, color palette, atmosphere, natural forces, or metaphoric forms surrounding [SUBJECT], reflecting the name’s meaning through mood and symbolism rather than literal text, letters, or icons, using lighting direction, intensity, and color temperature to reinforce the emotional essence of [NAME], integrating symbolic elements so they feel physically and spatially connected to [SUBJECT] rather than decorative, keeping the background restrained and uncluttered with no readable text, and ensuring strong visual impact, coherence, and stylistic integrity throughout the composition. ``` #### Example Outputs <img width=

/> ### 1.3. Zipped Lips Portrait **Prompt:** ```text Make a photorealistic studio head portrait of the person from the provided reference photo, used strictly as identity anchor, captured at eye level with natural facial perspective, framed as a clean studio photograph with only the head and upper shoulders visible, set against a professional studio background with soft neutral tones, showing the subject centered and facing forward with a calm direct gaze, with a single hand gently closing a metal zipper over the lips, ensuring the lips and zipper merge naturally and convincingly without distortion, rendered with realistic skin texture and fine detail, lit with soft controlled studio lighting to create an elegant, cinematic, and emotionally restrained image. ``` #### Example Output <img width=

is grouped by theme and includes the full text for direct use. All created by [@aimikoda](https://x.com/aimikoda) ## Table of Contents 1. [Portraits & Identity](#1-portraits-identity) 2. [Style & Transformation](#2-style-transformation) 3. [Fashion & Product Imaging](#3-fashion-product-imaging) 4. [Advertising & Branding Concepts](#4-advertising-branding-concepts) 5. [Movie Making & Storytelling & Comics](#5-movie-making-storytelling-comics) 6. [Cities & Architecture](#6-cities-architecture) 7. [Worlds & Dioramas](#7-worlds-dioramas) 8. [Games & Maps](#8-games-maps) 9. [Food & Culture](#9-food-culture) 10. [Holidays & Humor](#10-holidays-humor) ## 1. Portraits & Identity ### 1.1. Name Meaning Portrait Turn Your Name’s Meaning Into an Iconic Portrait **Prompt:**

在 assets/images/ 文件夹中有6个 杂志封面, 分别是: forbes-cover.jpg, national-geographic-cover.jpg, rolling-stone-cover.jpg, science-cover.jpg, time-cover.jpg, vogue-cover.jpg 用户可以上传自己的图片, 并根据这些杂志封面组成新的图片, 杂志封面对应的 prompt 如下: Use the magazine cover [image 1] as a template. Replace its original subject and elements with the person and objects from [image 2]. The final image must retain the style, typography, composition, and the subject's pose from the template [image 1].