GPT Image 2AI 提示词
Doki Doki Horror Visual Novel Prompt — Sora 2
Create a psychological horror anime visual novel composite with torn paper edges, a central silhouette, and four distinct character portraits.

提示词 · EN
type
anime visual novel composite illustrationtheme
psychological horror mixed with cute dating simcenterpiece
top
Circular logo featuring a blue stitched heart, colorful 3D text '{argument name="logo text top" default="DOKI DOKI"}', a green pen with a pink heart charm, and a white banner reading '{argument name="logo text bottom" default="Literature Club!"}'bottom
Silhouette of a person from behind, sitting at a desk, illuminated by a glowing {argument name="monitor glow color" default="pink"} computer monitorlayout
style
torn paper edges framing four distinct character portraits in the cornerssections
item 1
position
top-leftdescription
Anime girl with long brown hair, white hair bow, green eyes, gentle smile, wearing a {argument name="uniform color" default="gray"} school uniformitem 2
position
top-rightdescription
Anime girl with long dark purple hair, purple eyes, worried expression, wearing a {argument name="uniform color" default="gray"} school uniformitem 3
position
bottom-leftdescription
Anime girl with short pink hair, red hairclip, pink eyes, pouting expression, wearing a {argument name="uniform color" default="gray"} school uniformitem 4
position
bottom-rightdescription
Anime girl with short peach hair, red hair bow, blue eyes, cheerful smile, wearing a {argument name="uniform color" default="gray"} school uniformfloating_elements
item 1
count
3description
white question marks scattered around the central silhouetteitem 2
count
4description
pieces of lined paper with blurry handwritten text floating in the foregroundbackground
{argument name="atmosphere" default="dark, shadowy, slightly grungy and glitchy"}关于这个提示词
Create a psychological horror anime visual novel composite with torn paper edges, a central silhouette, and four distinct character portraits. 它更适合作为 GPT Image 2 的概念艺术起点:先保留画面结构、主体关系和风格约束,再替换成你的品牌、人物或场景。
这条提示词使用结构化 JSON 组织信息,包含 type、theme、centerpiece、layout、floating elements、background 等字段。保留这种层级能让模型更清楚地区分画面主题、布局、界面元素和细节约束。
使用时建议先小幅修改主体、场景、镜头和色调,再生成多个版本对比构图与细节。这样页面内容对用户有实际帮助,也避免把模型名或标签机械堆在正文里。
如何使用这个提示词
- Copy the entire prompt JSON into your AI image generator's input field.
- Select a model that excels at anime style, like Sora 2 or a compatible diffusion model.
- Customize the variables (e.g.,
{argument name="logo text top"}) to change text, colors, or character details. - Generate the image and review the composite layout with its central logo and corner portraits.
- Regenerate or tweak arguments for variations in mood, character expression, or color scheme.



