Prompt Template Categories in Awesome-GPT-Image-2: A Complete Guide

Awesome-GPT-Image-2 ships with 13 industrial-grade prompt template categories defined in docs/templates.md, spanning visual domains from UI mockups and e-commerce photography to historical period art and document layouts.

The awesome-gpt-image-2 repository by freestylefly standardizes AI image generation through a curated taxonomy of prompt templates. These categories—identified by stable anchor IDs like tpl-ui and tpl-product—provide structured vocabularies for producing consistent, high-fidelity visuals across 13 distinct professional domains. Each template specification lives in the docs/templates.md file and is consumed by the React frontend components to guide user input.

The 13 Prompt Template Categories

UI & Interface (tpl-ui)

Templates for high-fidelity app screens, mobile/web UI mock-ups, and JSON-ready UI specifications. This category targets product designers needing consistent interface generation.

为[产品类型]生成一张[平台，如 iOS/Android/Web]界面图。
核心功能：[功能点A]、[功能点B]、[功能点C]。
视觉风格：[极简/科技/拟物]，主色[颜色]，强调色[颜色]。
布局：[顶部导航/双栏/卡片流]，信息层级清晰，留白充足。
输出：高保真UI截图，文字清晰可读，比例[9:16/16:9]。

Reference: docs/templates.md#tpl-ui

Infographic & Data Visualization (tpl-infographic)

Structured prompts for charts, timelines, and data-driven graphics. Use this for educational content and analytical reports.

生成[主题：明确、具体，避免宽泛。例如：“老年人日常健康管理指南”而非“健康”]信息图，目标读者为[人群:细化人群特征，如年龄段、职业、兴趣等]。
结构：标题区 + [3-5]个模块（每模块含图标、短标题、1-2句说明,模块间逻辑：可用箭头、颜色区分或连接线提示信息流或关系等适当的方式)。
图表类型：[流程图/对比图/关系图/时间线]。
风格：[专业报告/科普插画/儿童教育等]，主色[颜色]，背景[浅色/深色]。
输出：信息层级清晰、可读性高的中文信息图。

Reference: docs/templates.md#tpl-infographic

Poster & Layout (tpl-poster)

General poster designs, campaign ads, and conceptual typography posters. Enforces strict layout dominance for text elements.

Create ONE finished premium conceptual typography poster for the exact title:
"[标题/词语/短句]"
Single poster only. No moodboard, grid, presentation board, mockup, captions, prompt text, process sheet, or sample labels.
The title must be the dominant visual structure of the poster: huge, readable, powerful, and spelled exactly.

Reference: docs/templates.md#tpl-poster

Product & E-commerce (tpl-product)

Product hero images, e-commerce listings, and beauty-recommendation reports. Optimized for commercial display standards.

生成[商品名]电商主图，卖点为[卖点1]、[卖点2]。
场景：[纯色棚拍/生活方式场景]，镜头：[特写/半身/全景]。
材质细节：[材质关键词]，灯光：[柔光/侧光/轮廓光]。
附加元素：[价格角标/卖点icon/促销文案]。
输出：电商平台可直接使用的商品展示图。

Reference: docs/templates.md#tpl-product

Brand & Logo (tpl-brand)

Brand identity systems, logo concepts, and brand-touchpoint visual boards. Focuses on cohesive visual language generation.

为[品牌名]设计品牌视觉方案。品牌关键词：[关键词1]、[关键词2]、[关键词3]。
包含：Logo方向[几何/字标/图形]、辅助图形、主辅色、应用示意。
风格：[现代/高级/亲和]，行业：[行业]，受众：[受众]。
输出：统一风格的品牌识别视觉图。

Reference: docs/templates.md#tpl-brand

Architecture & Space (tpl-architecture)

Architectural renderings and interior visualizations. Includes material specifications and lighting scenarios.

生成[空间类型]设计效果图，功能定位为[用途]。
风格：[现代简约/工业/新中式]，材质：[木/石/金属/玻璃]。
空间结构：[开敞/分区]，动线：[主通道说明]。
光线：[自然采光/人工照明方案]，时间：[白天/夜景]。
输出：写实建筑空间渲染图。

Reference: docs/templates.md#tpl-architecture

Photography & Realism (tpl-photo)

Photo-style prompts covering subjects, lenses, lighting, and mood. Emphasizes camera parameters and realistic texture.

拍摄主题：[人物/物品/街景]，场景为[地点]。
摄影参数风格：[35mm/85mm]，[浅景深/深景深]，[纪实/电影感]。
光线：[自然光/夜景霓虹/逆光]，情绪：[情绪词]。
细节要求：[肤质/材质/颗粒感]。
输出：高写实摄影风格图像。

Reference: docs/templates.md#tpl-photo

Illustration & Art (tpl-illustration)

Stylized illustration briefs spanning anime, watercolor, flat design, and oil painting. Targets artistic rather than photorealistic outputs.

创作[题材]插画，主角为[角色/主体]。
画风：[日漫/水彩/扁平/厚涂]，线条：[细腻/粗犷]。
配色：[配色方案]，背景：[简洁/复杂场景]。
构图：[近景/中景/远景]，重点表现[细节]。
输出：可用于封面或社媒发布的高质量插画。

Reference: docs/templates.md#tpl-illustration

Character & Role (tpl-character)

Character concept art, pose sheets, and 3-D toy conversion. Defines physical appearance and narrative context.

设计[角色身份]角色设定图。
外观：[年龄/发型/服饰/配件]，性格：[关键词]。
姿态：[站姿/动态动作]，表情：[情绪]。
世界观：[时代/阵营/职业]，标志性元素：[元素]。
输出：角色主视图 + 风格统一的人设图。

Reference: docs/templates.md#tpl-character

Scene & Narrative (tpl-scene)

Narrative scene generation with story beats, camera language, and atmosphere. Built for storytelling and concept art pipelines.

生成[故事主题]场景图，发生在[时间+地点]。
主事件：[事件描述]，主角：[角色]，冲突点：[冲突]。
镜头语言：[广角建立镜头/中景叙事/特写]。
氛围：[紧张/温暖/悬疑]，色调：[冷/暖/高反差]。
输出：具备叙事张力的场景概念图。

Reference: docs/templates.md#tpl-scene

Historical & Period (tpl-history)

Historical and 古风 (classical Chinese style) prompts with era-specific costume and architecture rules. Enforces period accuracy.

生成[朝代/古风设定]题材画面，主题为[主题]。
人物：[身份/服饰/器物]，场景：[宫廷/市井/山水]。
美术风格：[工笔/写意/影视写实]，色调：[色调]。
文化细节：[纹样/礼制/建筑要素]。
输出：历史氛围准确的古风题材图。

Reference: docs/templates.md#tpl-history

Document & Publication (tpl-document)

Layouts for menus, magazines, reports, and corporate brochures. Focuses on typographic hierarchy and grid systems.

制作[文档类型，如菜单/杂志内页/报纸版式]。
版面结构：[栏数/页边距/标题层级]。
内容模块：[封面区/正文区/图表区/脚注]。
字体风格：[衬线/无衬线]，配色：[配色方案]。
输出：可读性强、版式规范的出版物视觉稿。

Reference: docs/templates.md#tpl-document

Other Applications (tpl-other)

Catch-all templates for miscellaneous visual tasks including concept products and custom generation workflows.

任务目标：[你要生成的内容类型]。
输入约束：主体[主体]，场景[场景]，风格[风格]，色彩[配色]。
质量约束：清晰度[高清/4K]，比例[比例]，构图[构图方式]。
输出约束：用于[用途]，需突出[核心信息]。
请输出一版主方案 + 一版备选方案。

Reference: docs/templates.md#tpl-other

Repository Architecture

The template system relies on four critical files that separate taxonomy from implementation:

• docs/templates.md – The authoritative source declaring all 13 categories, their anchor IDs, and the full prompt text for each template.
• data/style-library.json – Defines reusable style tokens (color palettes, aspect ratios, typography scales) referenced by the templates to ensure consistency.
• src/main.jsx – React entry point where users select a template category; consumes the anchor IDs to route parameters.
• src/apimartClient.js – Client library that submits the populated template prompts to the GPT-image generation API.

Summary

• 13 categories cover professional visual domains from UI design (tpl-ui) to historical art (tpl-history).
• Anchor IDs (e.g., tpl-product, tpl-character) provide stable references for programmatic access.
• Source of truth for all templates is docs/templates.md in the repository root.
• Style consistency is enforced through shared tokens in data/style-library.json.
• Implementation spans React components (src/main.jsx) and API clients (src/apimartClient.js).

Frequently Asked Questions

How do I reference a specific template category programmatically?

Each category uses a stable anchor ID (e.g., tpl-ui, tpl-product) defined in docs/templates.md. You can reference these IDs directly in URL fragments or import them via the React components in src/main.jsx to filter the template selection UI.

What distinguishes the Photography category from the Illustration category?

The Photography (tpl-photo) category emphasizes photorealistic parameters like lens focal lengths, lighting setups, and camera film grain, while the Illustration (tpl-illustration) category focuses on artistic media such as watercolor, anime, or oil painting styles. Both are defined in docs/templates.md but target fundamentally different visual outputs.

Where are reusable style definitions stored for these templates?

Shared style tokens—including color palettes, aspect ratios, and typography rules—are centralized in data/style-library.json. This JSON file acts as a vocabulary library that the template prompts in docs/templates.md reference for consistent styling across categories.

How many total templates are available in the repository?

The repository contains 21 industrial-grade prompt templates distributed across the 13 thematic categories. While some domains like UI or Product have multiple specialized variants, all templates follow the anchor ID taxonomy documented in the main templates file.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}