GPT Image 2AI 提示词
Dark Fantasy Lore Map — Sora 2 Prompt
Generates a detailed, tiered infographic chart mapping a complex dark fantasy hierarchy of pathways, pillars, and lore relationships.
提示词 · EN
type
complex dark fantasy infographic chartstyle
{argument name="theme style" default="dark fantasy, gothic, arcane, gold borders on black background"}header
{argument name="main title" default="诡秘之主·全途径与支柱关系图"}layout
sections
item 1
title
{argument name="tier 1 title" default="第一级：三大支柱 (THE PILLARS)"}position
topcount
3labels
item 1
1. 诡秘之主 (LORD OF THE MYSTERIES)item 2
2. 上帝 / 全知全能者 (GOD ALMIGHTY)item 3
3. 堕落母神 (MOTHER GODDESS OF DEPRAVITY)details
Each box contains a main entity illustration, a 'Sefirah' label, and a 'Pathways' section with circular icons. Box 1 has 3 icons, Box 2 has 5 icons, Box 3 has 2 icons.item 2
title
{argument name="tier 2 title" default="第二级：潜在的第四支柱 (THE 4TH PILLAR)"}position
upper-middlecount
2labels
item 1
A. 永恒之暗 (ETERNAL DARKNESS)item 2
B. 毁灭灾祸 (CALAMITY OF DESTRUCTION)details
Two large boxes connected by a central text block with arrows. Box A has 3 pathway icons. Box B has 2 pathway icons.item 3
title
{argument name="tier 3 title" default="第三级：其他旧日 (GREAT OLD ONES)"}position
lower-middlecount
4labels
item 1
失序之国 (THE ANARCHY)item 2
知识荒野 (DEMON OF KNOWLEDGE)item 3
光之钥 (KEY OF LIGHT)item 4
父神 (FATHER OF DEVILS)details
Four vertical boxes. Box 1 has 2 icons, Box 2 has 2 icons, Box 3 has 1 icon, Box 4 has 2 icons.item 4
title
Legend and Logicposition
bottomcount
4labels
item 1
图解说明与核心逻辑item 2
1. “相邻途径”原则item 3
2. “聚合定律”item 4
3. “支柱”与“旧日”的区别details
Four informational panels containing text, bullet points, and small diagrammatic arrows explaining the chart's rules and hierarchies.关于这个提示词
Generates a detailed, tiered infographic chart mapping a complex dark fantasy hierarchy of pathways, pillars, and lore relationships. 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、style、header、layout 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 full JSON prompt object above.
- Paste it into your preferred AI image or video generation model.
- Customize the argument variables (like
theme styleormain title) to fit your specific lore. - Generate the image, then refine the prompt based on the output if needed.
