gpt-image-2-skill

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

GPT Image 2 Skill

GPT Image 2 技能

Skill by ara.so — Daily 2026 Skills collection.
A prompt gallery, CLI, and agentic skill for OpenAI's
gpt-image-2
model. Provides 162 curated prompts across categories (research figures, UI mockups, typography, photography, anime, maps, product shots), a full-featured CLI, and skill integrations for Claude Code, Codex, and other agent runtimes.

ara.so开发的技能 — 2026每日技能合集。
一款面向OpenAI
gpt-image-2
模型的提示词图库、CLI工具及Agent技能。提供162个分类精选提示词(涵盖科研图表、UI原型、排版设计、摄影、动漫、地图、产品实拍)、功能完整的CLI工具,以及可与Claude Code、Codex等Agent运行时集成的技能插件。

Install

安装

CLI (fastest)

CLI(最快方式)

bash
undefined
bash
undefined

Run without installing

无需安装直接运行

uvx --from git+https://github.com/wuyoscar/gpt_image_2_skill gpt-image -p "a cat astronaut"
uvx --from git+https://github.com/wuyoscar/gpt_image_2_skill gpt-image -p "a cat astronaut"

Install to PATH permanently

永久安装至系统PATH

uv tool install git+https://github.com/wuyoscar/gpt_image_2_skill gpt-image -p "a cat astronaut"
undefined
uv tool install git+https://github.com/wuyoscar/gpt_image_2_skill gpt-image -p "a cat astronaut"
undefined

Claude Code

Claude Code

text
/plugin marketplace add wuyoscar/gpt_image_2_skill
/plugin install gpt-image@wuyoscar-skills
text
/plugin marketplace add wuyoscar/gpt_image_2_skill
/plugin install gpt-image@wuyoscar-skills

Codex

Codex

text
$skill-installer install https://github.com/wuyoscar/gpt_image_2_skill/tree/main/skills/gpt-image
text
$skill-installer install https://github.com/wuyoscar/gpt_image_2_skill/tree/main/skills/gpt-image

Manual agent-skill install

手动安装Agent技能

bash
git clone https://github.com/wuyoscar/gpt_image_2_skill.git
cd gpt_image_2_skill

export AGENT_SKILLS_DIR="/path/to/your/agent/skills"
mkdir -p "$AGENT_SKILLS_DIR"
ln -s "$PWD/skills/gpt-image" "$AGENT_SKILLS_DIR/gpt-image"

bash
git clone https://github.com/wuyoscar/gpt_image_2_skill.git
cd gpt_image_2_skill

export AGENT_SKILLS_DIR="/path/to/your/agent/skills"
mkdir -p "$AGENT_SKILLS_DIR"
ln -s "$PWD/skills/gpt-image" "$AGENT_SKILLS_DIR/gpt-image"

Configuration

配置

The CLI and skill read your OpenAI key from the environment or
~/.env
:
bash
export OPENAI_API_KEY="sk-..."
No other configuration is required.

CLI工具和技能会从环境变量或
~/.env
文件中读取你的OpenAI密钥:
bash
export OPENAI_API_KEY="sk-..."
无需其他配置。

CLI Reference

CLI参考文档

Text → Image (generation)

文本→图像(生成)

bash
undefined
bash
undefined

Basic generation

基础生成

gpt-image -p "a photorealistic convenience store at 10pm"
gpt-image -p "a photorealistic convenience store at 10pm"

With size, quality, and explicit output file

指定尺寸、画质及输出文件

gpt-image -p "a neon-lit Tokyo alley at midnight"
--size portrait --quality high -f tokyo-alley.png
gpt-image -p "a neon-lit Tokyo alley at midnight"
--size portrait --quality high -f tokyo-alley.png

Square, low quality (cheap draft)

方形低画质(低成本草稿)

gpt-image -p "watercolor mountains at sunrise"
--size 1k --quality low -f draft.png
gpt-image -p "watercolor mountains at sunrise"
--size 1k --quality low -f draft.png

Batch: generates 4 variants, saved as out_0.png … out_3.png

批量生成:生成4个变体,保存为out_0.png … out_3.png

gpt-image -p "product shot of a ceramic mug on white"
--size square --quality medium -n 4 -f out.png
undefined
gpt-image -p "product shot of a ceramic mug on white"
--size square --quality medium -n 4 -f out.png
undefined

Text + Reference Image → Image (edit / restyle)

文本+参考图→图像(编辑/重绘)

bash
undefined
bash
undefined

Single reference restyle

单参考图重绘

gpt-image -p "Make it a winter evening with heavy snowfall"
-i chess.png --quality high -f chess-winter.png
gpt-image -p "Make it a winter evening with heavy snowfall"
-i chess.png --quality high -f chess-winter.png

Multi-reference composite: dog from image 2, scene from image 1

多参考图合成:图像2的狗,图像1的场景

gpt-image -p "Place the dog from image 2 next to the woman in image 1.
Match the same lighting, composition, and background."
-i woman.png -i dog.png --size portrait --quality medium -f woman-with-dog.png
undefined
gpt-image -p "Place the dog from image 2 next to the woman in image 1.
Match the same lighting, composition, and background."
-i woman.png -i dog.png --size portrait --quality medium -f woman-with-dog.png
undefined

Mask-based Inpainting

基于蒙版的图像修复

bash
undefined
bash
undefined

opaque pixels = keep, transparent pixels = regenerate

不透明像素=保留,透明像素=重新生成

gpt-image -p "replace sky with aurora borealis"
-i photo.jpg -m sky_mask.png -f aurora.png
undefined
gpt-image -p "replace sky with aurora borealis"
-i photo.jpg -m sky_mask.png -f aurora.png
undefined

Full Parameter Reference

完整参数参考

FlagValuesDefaultNotes
-p, --prompt
stringrequiredFull prompt text
-f, --file
pathauto-timestamped
.png
Output file path
-i, --image
path (repeatable)Triggers
/v1/images/edits
; pass multiple for multi-ref
-m, --mask
path (PNG with alpha)Requires
-i
; transparent = regenerate
--size
1k
2k
4k
portrait
landscape
square
wide
tall
or
1024x1024
1024x1024
Literals must be 16-px multiples, max edge 3840
--quality
auto
low
medium
high
high
Budget dial:
low
=drafts,
high
=final/text-heavy
-n, --n
int1Batch count; suffixes files
_0
,
_1
, …
--background
auto
opaque
API default
opaque
disables transparency
--moderation
auto
low
low
low
for broader exploration
--format
png
jpeg
webp
png
Response encoding format
--compression
0–100JPEG/WebP only
Exit codes:
0
success ·
1
API/refusal error ·
2
bad args or missing key

参数标识可选值默认值说明
-p, --prompt
字符串必填完整提示词文本
-f, --file
文件路径自动生成带时间戳的
.png
输出文件路径
-i, --image
文件路径(可重复)触发
/v1/images/edits
接口;可传入多张参考图
-m, --mask
文件路径(带透明通道的PNG)需配合
-i
使用;透明区域将被重新生成
--size
1k
2k
4k
portrait
landscape
square
wide
tall
1024x1024
1024x1024
尺寸值需为16像素的倍数,最大边长3840像素
--quality
auto
low
medium
high
high
成本控制:
low
=草稿,
high
=最终版本/含大量文本的图像
-n, --n
整数1批量生成数量;文件名将添加
_0
,
_1
…后缀
--background
auto
opaque
API默认值
opaque
将禁用透明背景
--moderation
auto
low
low
low
模式支持更广泛的创作探索
--format
png
jpeg
webp
png
响应编码格式
--compression
0–100仅适用于JPEG/WebP格式
退出码:
0
成功 ·
1
API拒绝/错误 ·
2
参数错误或密钥缺失

Python SDK Usage

Python SDK 使用示例

Text → Image

文本→图像

python
from openai import OpenAI

client = OpenAI()  # reads OPENAI_API_KEY from environment

result = client.images.generate(
    model="gpt-image-2",
    prompt="A photorealistic ceramic mug on a white studio background, "
           "soft directional light, light shadow beneath",
    size="1024x1024",   # square
    quality="high",
)
python
from openai import OpenAI

client = OpenAI()  # 从环境变量读取OPENAI_API_KEY

result = client.images.generate(
    model="gpt-image-2",
    prompt="A photorealistic ceramic mug on a white studio background, "
           "soft directional light, light shadow beneath",
    size="1024x1024",   # 方形
    quality="high",
)

Save result

保存结果

import base64 from pathlib import Path
image_bytes = base64.b64decode(result.data[0].b64_json) Path("mug.png").write_bytes(image_bytes) print("Saved mug.png")
undefined
import base64 from pathlib import Path
image_bytes = base64.b64decode(result.data[0].b64_json) Path("mug.png").write_bytes(image_bytes) print("Saved mug.png")
undefined

Portrait / Tall Generation

竖版/长图生成

python
result = client.images.generate(
    model="gpt-image-2",
    prompt="Minimalist event poster: 'Boston Spring Jazz Festival · April 2026' "
           "in bold serif, pastel cherry-blossom watercolor background, centered layout",
    size="1024x1536",   # portrait (3:4)
    quality="high",
)
python
result = client.images.generate(
    model="gpt-image-2",
    prompt="Minimalist event poster: 'Boston Spring Jazz Festival · April 2026' "
           "in bold serif, pastel cherry-blossom watercolor background, centered layout",
    size="1024x1536",   # 竖版(3:4)
    quality="high",
)

Image Edit (single reference)

图像编辑(单参考图)

python
result = client.images.edit(
    model="gpt-image-2",
    image=open("chess.png", "rb"),
    prompt="Make it a winter evening with heavy snowfall, keep the chess pieces identical",
    size="1024x1024",
    quality="high",
)
python
result = client.images.edit(
    model="gpt-image-2",
    image=open("chess.png", "rb"),
    prompt="Make it a winter evening with heavy snowfall, keep the chess pieces identical",
    size="1024x1024",
    quality="high",
)

Multi-Reference Edit

多参考图编辑

python
result = client.images.edit(
    model="gpt-image-2",
    image=[open("woman.png", "rb"), open("dog.png", "rb")],
    prompt="Place the dog from image 2 next to the woman in image 1. "
           "Match the same lighting, composition, and background. "
           "Do not change anything else.",
    size="1024x1536",
    quality="medium",
)
python
result = client.images.edit(
    model="gpt-image-2",
    image=[open("woman.png", "rb"), open("dog.png", "rb")],
    prompt="Place the dog from image 2 next to the woman in image 1. "
           "Match the same lighting, composition, and background. "
           "Do not change anything else.",
    size="1024x1536",
    quality="medium",
)

Mask-Based Inpainting

基于蒙版的图像修复

python
result = client.images.edit(
    model="gpt-image-2",
    image=open("photo.jpg", "rb"),
    mask=open("sky_mask.png", "rb"),   # transparent = regenerate
    prompt="Replace the sky with dramatic aurora borealis, keep everything below the horizon identical",
    size="1024x1024",
    quality="high",
)
python
result = client.images.edit(
    model="gpt-image-2",
    image=open("photo.jpg", "rb"),
    mask=open("sky_mask.png", "rb"),   # 透明区域=重新生成
    prompt="Replace the sky with dramatic aurora borealis, keep everything below the horizon identical",
    size="1024x1024",
    quality="high",
)

Batch Generation with Saving

批量生成并保存

python
import base64
from pathlib import Path
from openai import OpenAI

def generate_batch(prompt: str, n: int = 4, size: str = "1024x1024",
                   quality: str = "medium", out_prefix: str = "variant") -> list[Path]:
    client = OpenAI()
    result = client.images.generate(
        model="gpt-image-2",
        prompt=prompt,
        size=size,
        quality=quality,
        n=n,
    )
    paths = []
    for i, item in enumerate(result.data):
        path = Path(f"{out_prefix}_{i}.png")
        path.write_bytes(base64.b64decode(item.b64_json))
        paths.append(path)
        print(f"Saved {path}")
    return paths
python
import base64
from pathlib import Path
from openai import OpenAI

def generate_batch(prompt: str, n: int = 4, size: str = "1024x1024",
                   quality: str = "medium", out_prefix: str = "variant") -> list[Path]:
    client = OpenAI()
    result = client.images.generate(
        model="gpt-image-2",
        prompt=prompt,
        size=size,
        quality=quality,
        n=n,
    )
    paths = []
    for i, item in enumerate(result.data):
        path = Path(f"{out_prefix}_{i}.png")
        path.write_bytes(base64.b64decode(item.b64_json))
        paths.append(path)
        print(f"Saved {path}")
    return paths

Usage

使用示例

variants = generate_batch( prompt="product shot of a blue glass water bottle, white background, studio lighting", n=4, quality="low", # cheap sweep; rerun winner at high )

---
variants = generate_batch( prompt="product shot of a blue glass water bottle, white background, studio lighting", n=4, quality="low", # 低成本快速生成;选中最优后再用high画质重新生成 )

---

Prompt Engineering Patterns

提示词设计范式

Structure template

结构模板

[background/scene] → [subject] → [key details] → [constraints/intended use]
[背景/场景] → [主体] → [关键细节] → [约束条件/使用场景]

Research paper figure

科研论文图表

bash
gpt-image -p "Clean scientific diagram: transformer architecture overview. \
White background, labeled encoder/decoder blocks with arrows, \
color-coded attention heads in teal and orange, \
sans-serif labels, publication-ready, 4K resolution" \
--size landscape --quality high -f transformer-diagram.png
bash
gpt-image -p "Clean scientific diagram: transformer architecture overview. \
White background, labeled encoder/decoder blocks with arrows, \
color-coded attention heads in teal and orange, \
sans-serif labels, publication-ready, 4K resolution" \
--size landscape --quality high -f transformer-diagram.png

UI mockup

UI原型

bash
gpt-image -p "Mobile app UI mockup, iOS style, dark mode. \
Fitness tracking dashboard: circular progress ring in neon green, \
daily steps '8,432', heart rate '74 bpm', \
bottom nav with 4 icons, pixel-perfect, no lorem ipsum" \
--size portrait --quality high -f fitness-app.png
bash
gpt-image -p "Mobile app UI mockup, iOS style, dark mode. \
Fitness tracking dashboard: circular progress ring in neon green, \
daily steps '8,432', heart rate '74 bpm', \
bottom nav with 4 icons, pixel-perfect, no lorem ipsum" \
--size portrait --quality high -f fitness-app.png

Typography poster

排版海报

bash
gpt-image -p "Event poster. Text: 'SUMMER SONIC 2026' in bold condensed sans-serif. \
Subtext: 'Tokyo · August 9–10'. Vivid sunset gradient background (magenta to amber). \
Geometric grid overlay, high contrast, print-ready" \
--size portrait --quality high -f poster.png
bash
gpt-image -p "Event poster. Text: 'SUMMER SONIC 2026' in bold condensed sans-serif. \
Subtext: 'Tokyo · August 9–10'. Vivid sunset gradient background (magenta to amber). \
Geometric grid overlay, high contrast, print-ready" \
--size portrait --quality high -f poster.png

Photorealistic product shot

写实产品实拍

bash
gpt-image -p "Photorealistic product photo: matte black insulated coffee thermos, \
condensation droplets, placed on dark slate surface, \
single soft key light from upper-left, shallow depth of field, \
shot on Canon 5D, 85mm lens, commercial quality" \
--size square --quality high -f thermos.png
bash
gpt-image -p "Photorealistic product photo: matte black insulated coffee thermos, \
condensation droplets, placed on dark slate surface, \
single soft key light from upper-left, shallow depth of field, \
shot on Canon 5D, 85mm lens, commercial quality" \
--size square --quality high -f thermos.png

Put required text in quotes

需精准呈现的文本加引号

python
undefined
python
undefined

Any text that must appear verbatim in the image — put in straight quotes in the prompt

所有需要在图像中精准呈现的文本——在提示词中用直引号包裹

prompt = '''Storefront sign reading "OPEN 24/7" in red neon. Below it: "Est. 1987" in smaller white block letters. Realistic neon glow, night scene, rain-slicked pavement.'''

---
prompt = '''Storefront sign reading "OPEN 24/7" in red neon. Below it: "Est. 1987" in smaller white block letters. Realistic neon glow, night scene, rain-slicked pavement.'''

---

Quality / Budget Strategy

画质/成本策略

Stage
--quality
When to use
Exploration sweep
low
Generating 8–16 variants to find direction
Normal iteration
medium
Style probing, layout checks
Final / shipping
high
In-image text, dense diagrams, posters, paper figures
Rule of thumb: start every new concept at
low
, run 4 variants, pick the best, then rerun at
high
.
bash
undefined
阶段
--quality
参数
使用场景
探索阶段
low
生成8–16个变体以确定创作方向
常规迭代
medium
风格探索、布局校验
最终交付
high
含文本的图像、密集图表、海报、科研论文配图
经验法则: 所有新创意先以
low
画质生成4个变体,选中最优后再用
high
画质重新生成。
bash
undefined

Step 1: cheap sweep

第一步:低成本快速探索

gpt-image -p "minimalist logo for a coffee brand" --quality low -n 4 -f logo.png
gpt-image -p "minimalist logo for a coffee brand" --quality low -n 4 -f logo.png

Step 2: pick winner (e.g. logo_2.png), rerun at high

第二步:选中最优版本(如logo_2.png),用high画质重新生成最终版

gpt-image -p "minimalist logo for a coffee brand" --quality high -f logo-final.png

---
gpt-image -p "minimalist logo for a coffee brand" --quality high -f logo-final.png

---

Size Reference

尺寸参考

AliasPixelsRatioBest for
square
/
1k
1024×10241:1Social posts, icons, product shots
portrait
1024×15362:3Mobile UI, posters, stories
landscape
1536×10243:2Web banners, diagrams
wide
1792×10247:4Cinematic, hero sections
tall
1024×17924:7Long-form mobile content
2k
2048×20481:1High-res assets

别名像素尺寸比例适用场景
square
/
1k
1024×10241:1社交帖子、图标、产品实拍
portrait
1024×15362:3移动端UI、海报、故事类内容
landscape
1536×10243:2web横幅、图表
wide
1792×10247:4电影级画面、首页 hero 区域
tall
1024×17924:7移动端长内容
2k
2048×20481:1高分辨率素材

Common Patterns & Recipes

常见场景与示例

Virtual try-on (multi-ref edit)

虚拟试穿(多参考图编辑)

python
undefined
python
undefined

image 1 = person, image 2 = garment

图像1=人物,图像2=服装

result = client.images.edit( model="gpt-image-2", image=[open("person.png", "rb"), open("shirt.png", "rb")], prompt="Dress the person in image 1 wearing the shirt from image 2. " "Keep the person's face, pose, and background identical. " "Natural fabric draping and lighting.", size="1024x1536", quality="high", )
undefined
result = client.images.edit( model="gpt-image-2", image=[open("person.png", "rb"), open("shirt.png", "rb")], prompt="Dress the person in image 1 wearing the shirt from image 2. " "Keep the person's face, pose, and background identical. " "Natural fabric draping and lighting.", size="1024x1536", quality="high", )
undefined

Billboard / signage mockup

广告牌/标识原型

python
result = client.images.edit(
    model="gpt-image-2",
    image=open("billboard_photo.jpg", "rb"),
    mask=open("billboard_mask.png", "rb"),
    prompt='Replace the billboard face with: "SALE ENDS SUNDAY" '
           'in bold white text on solid red background. '
           'Match perspective and lighting of surrounding scene.',
    size="1536x1024",
    quality="high",
)
python
result = client.images.edit(
    model="gpt-image-2",
    image=open("billboard_photo.jpg", "rb"),
    mask=open("billboard_mask.png", "rb"),
    prompt='Replace the billboard face with: "SALE ENDS SUNDAY" '
           'in bold white text on solid red background. '
           'Match perspective and lighting of surrounding scene.',
    size="1536x1024",
    quality="high",
)

Anime / manga style transfer

动漫/漫画风格转换

bash
gpt-image -p "Anime key visual style (Studio Ghibli-inspired): \
young woman standing on a hillside overlooking a coastal town at golden hour, \
painterly backgrounds, soft cel shading, \
detailed environmental storytelling, cinematic composition" \
--size landscape --quality high -f anime-scene.png
bash
gpt-image -p "Anime key visual style (Studio Ghibli-inspired): \
young woman standing on a hillside overlooking a coastal town at golden hour, \
painterly backgrounds, soft cel shading, \
detailed environmental storytelling, cinematic composition" \
--size landscape --quality high -f anime-scene.png

Translation / text replacement edit

翻译/文本替换编辑

python
undefined
python
undefined

Replace text in an existing image in a different language

将现有图像中的文本替换为其他语言

result = client.images.edit( model="gpt-image-2", image=open("menu_english.png", "rb"), prompt='Replace all English text with Japanese translations. ' 'Keep the exact same layout, fonts, colors, and imagery. ' 'Translate "Grilled Salmon" → "グリルサーモン", ' '"Caesar Salad" → "シーザーサラダ".', size="1024x1024", quality="high", )

---
result = client.images.edit( model="gpt-image-2", image=open("menu_english.png", "rb"), prompt='Replace all English text with Japanese translations. ' 'Keep the exact same layout, fonts, colors, and imagery. ' 'Translate "Grilled Salmon" → "グリルサーモン", ' '"Caesar Salad" → "シーザーサラダ".', size="1024x1024", quality="high", )

---

Troubleshooting

故障排查

OPENAI_API_KEY
not found

未找到
OPENAI_API_KEY

bash
export OPENAI_API_KEY="sk-..."
bash
export OPENAI_API_KEY="sk-..."

or add to ~/.env — the CLI reads it automatically

或添加至~/.env文件——CLI会自动读取

undefined
undefined

Refusal / content policy error (exit code 1)

内容政策拒绝错误(退出码1)

  • Full API response is echoed to stderr
  • Try rephrasing: be more descriptive and less ambiguous about intent
  • Switch
    --moderation auto
    --moderation low
    for broader exploration (already the CLI default)
  • API完整响应会输出至stderr
  • 尝试重新表述提示词:更清晰地描述创作意图,减少歧义
  • --moderation auto
    改为
    --moderation low
    以支持更广泛的创作探索(CLI默认已为该值)

Text in image is garbled or wrong

图像中文本模糊或错误

  • Always use
    --quality high
    for any prompt containing required text
  • Wrap required text in straight quotes inside the prompt string
  • Keep required text short (under ~6 words per element)
  • 任何包含需精准呈现文本的提示词,务必使用
    --quality high
  • 在提示词中用直引号包裹需精准呈现的文本
  • 需呈现的文本尽量简短(每个元素不超过约6个单词)

Multi-ref edit ignores one image

多参考图编辑忽略其中一张图

  • Be explicit: "the subject in image 1", "the object in image 2"
  • Reduce to two input images; more than two can cause ambiguity
  • 明确指定:“图像1中的主体”、“图像2中的物体”
  • 最多使用两张输入图;超过两张可能导致歧义

Size rejected by API

API拒绝尺寸参数

  • Dimensions must be multiples of 16
  • Max edge: 3840px
  • Max aspect ratio: 3:1
  • Total pixels: 655,360–8,294,400
  • 尺寸必须为16像素的倍数
  • 最大边长:3840像素
  • 最大宽高比:3:1
  • 总像素数:655,360–8,294,400

gpt-image-2
rejects
--input-fidelity

gpt-image-2
拒绝
--input-fidelity
参数

  • input-fidelity
    is a
    gpt-image-1
    /
    1.5
    parameter; the CLI drops it automatically for
    gpt-image-2
  • input-fidelity
    gpt-image-1
    /
    1.5
    的参数;CLI会自动为
    gpt-image-2
    忽略该参数

Slow generation at
high
quality

high
画质生成速度慢

Expected —
high
quality is significantly slower. Use
low
for drafts,
high
only for finals.

这是正常现象——
high
画质生成速度明显较慢。用
low
画质生成草稿,仅在最终交付时使用
high
画质。

Prompt Gallery Categories

提示词图库分类

The skill ships 162 prompts split across category files under
skills/gpt-image/references/
:
  • gallery-research-paper-figures.md
    — diagrams, charts, architecture visuals
  • gallery-ui-ux-mockups.md
    — mobile/web UI, dashboards, design systems
  • gallery-product-and-food.md
    — product shots, food styling, e-commerce
  • gallery-typography.md
    — posters, signage, lettering
  • gallery-photography.md
    — portrait, landscape, macro, street
  • gallery-anime-manga.md
    — key visuals, character design, backgrounds
  • gallery-maps.md
    — illustrated maps, infographic cartography
  • Start with
    gallery.md
    as a routing index to pick the right category file
本技能内置162个提示词,按类别存放在
skills/gpt-image/references/
目录下:
  • gallery-research-paper-figures.md
    — 图表、示意图、架构可视化
  • gallery-ui-ux-mockups.md
    — 移动端/web端UI、仪表盘、设计系统
  • gallery-product-and-food.md
    — 产品实拍、美食摄影、电商素材
  • gallery-typography.md
    — 海报、标识、字体设计
  • gallery-photography.md
    — 人像、风景、微距、街拍
  • gallery-anime-manga.md
    — 关键视觉图、角色设计、背景
  • gallery-maps.md
    — 插画地图、信息图式制图
  • 可先查看
    gallery.md
    作为索引,选择合适的分类文件