anbeime/skill / skills/gpt-image-2-prompt-engine
gpt-image-2-prompt-engine
Verified面向电商设计师、海报美工、品牌视觉、UI设计师、信息图编辑、商业摄影师、内容创作者等需要高质量可控出图的角色,在需要用 GPT-Image-2 生成电商主图、电影海报、信息图、品牌视觉、UI截图、古籍国风、角色IP等场景时,通过「Prompt as Code」原子化Schema+20+工业JSON模板+四步工作流,产出结构化、可复用、可批量的生图提示词,再调用 image_generation 出图。不适用于随意生图或简单风景照。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 9.1 KB |
| references/style-library.md | markdown | 25.7 KB |
| scripts/query_templates.py | script | 5.5 KB |
scripts/query_templates.py
View on GitHub#!/usr/bin/env python3
"""
GPT-Image-2 Prompt Engine - Template & Case Query Script
零依赖辅助脚本,按类别/关键词检索本地克隆仓库的模板与案例。
未克隆时优雅降级,输出在线 raw 链接与克隆命令。
Usage:
python3 query_templates.py --category ecommerce
python3 query_templates.py --keyword "香水 海报"
python3 query_templates.py --list
"""
import argparse
import os
import sys
# 上游仓库信息
REPO_URL = "https://github.com/freestylefly/awesome-gpt-image-2"
RAW_BASE = "https://raw.githubusercontent.com/freestylefly/awesome-gpt-image-2/main"
ONLINE_RESOURCES = {
"templates": f"{RAW_BASE}/docs/templates.md",
"gallery_part1": f"{RAW_BASE}/docs/gallery-part-1.md",
"gallery_part2": f"{RAW_BASE}/docs/gallery-part-2.md",
"gallery": f"{RAW_BASE}/docs/gallery.md",
"style_library": f"{RAW_BASE}/agents/skills/gpt-image-2-style-library/references/style-library.md",
}
# 模板类别映射
CATEGORIES = {
"ui": {"id": "ui-screenshot-system", "name": "UI截图系统", "anchor": "tpl-ui"},
"infographic": {"id": "infographic-engine", "name": "信息图引擎", "anchor": "tpl-infographic"},
"poster": {"id": "poster-campaign", "name": "海报与排版", "anchor": "tpl-poster"},
"ecommerce": {"id": "ecommerce-hero", "name": "电商主图", "anchor": "tpl-ecommerce"},
"brand": {"id": "brand-identity", "name": "品牌视觉", "anchor": "tpl-brand"},
"photo": {"id": "commercial-photo", "name": "商业摄影", "anchor": "tpl-photo"},
"character": {"id": "character-design", "name": "角色与IP", "anchor": "tpl-character"},
"narrative": {"id": "narrative-illustration", "name": "叙事插画", "anchor": "tpl-narrative"},
"classical": {"id": "classical-chinese", "name": "古籍国风", "anchor": "tpl-classical"},
"3d": {"id": "3d-render", "name": "3D渲染", "anchor": "tpl-3d"},
}
def find_local_repo():
"""尝试在常见路径查找本地克隆的仓库"""
candidates = [
os.path.join(os.path.dirname(__file__), "..", "..", "awesome-gpt-image-2"),
os.path.expanduser("~/awesome-gpt-image-2"),
"/tmp/awesome-gpt-image-2",
]
for path in candidates:
if os.path.isdir(os.path.join(path, "docs")):
return os.path.abspath(path)
return None
def print_online_links():
"""输出在线资源链接(降级模式)"""
print("Local repo not found. Using online search mode.\n")
print("Online resources:")
for name, url in ONLINE_RESOURCES.items():
print(f" - {name}: {url}")
print(f"\nClone for full search:")
print(f" git clone {REPO_URL}.git")
def list_categories():
"""列出所有模板类别"""
print("Available template categories:\n")
for key, info in CATEGORIES.items():
print(f" {key:15s} -> {info['name']} (ID: {info['id']})")
print(f"\nTemplates doc: {ONLINE_RESOURCES['templates']}")
def search_category(category):
"""按类别检索模板"""
cat = category.lower().strip()
if cat not in CATEGORIES:
print(f"Unknown category: {category}")
print(f"Available: {', '.join(CATEGORIES.keys())}")
return
info = CATEGORIES[cat]
local_repo = find_local_repo()
print(f"Category: {info['name']} (ID: {info['id']})\n")
if local_repo:
templates_path = os.path.join(local_repo, "docs", "templates.md")
if os.path.isfile(templates_path):
with open(templates_path, "r", encoding="utf-8") as f:
content = f.read()
anchor = f'<a name="{info["anchor"]}"></a>'
if anchor in content:
section = content.split(anchor)[1].split('<a name=')[0]
print(section[:3000])
return
print("Local template file parse failed, falling back to online links.")
print(f"Online templates: {ONLINE_RESOURCES['templates']}#{info['anchor']}")
print(f"Gallery: {ONLINE_RESOURCES['gallery']}")
def search_keyword(keyword):
"""按关键词搜索模板和案例"""
local_repo = find_local_repo()
print(f"Keyword: {keyword}\n")
if local_repo:
found = False
for doc_name in ["templates.md", "gallery.md", "gallery-part-1.md", "gallery-part-2.md"]:
doc_path = os.path.join(local_repo, "docs", doc_name)
if os.path.isfile(doc_path):
with open(doc_path, "r", encoding="utf-8") as f:
for i, line in enumerate(f, 1):
if keyword.lower() in line.lower():
print(f" [{doc_name}:{i}] {line.strip()[:200]}")
found = True
if found:
return
print("No local matches, falling back to online links.")
print(f"Online templates: {ONLINE_RESOURCES['templates']}")
print(f"Gallery Part1: {ONLINE_RESOURCES['gallery_part1']}")
print(f"Gallery Part2: {ONLINE_RESOURCES['gallery_part2']}")
def main():
parser = argparse.ArgumentParser(description="GPT-Image-2 Template & Case Query")
parser.add_argument("--category", "-c", help="Search by category")
parser.add_argument("--keyword", "-k", help="Search by keyword")
parser.add_argument("--list", "-l", action="store_true", help="List all categories")
args = parser.parse_args()
if args.list:
list_categories()
elif args.category:
search_category(args.category)
elif args.keyword:
search_keyword(args.keyword)
else:
parser.print_help()
print()
print_online_links()
if __name__ == "__main__":
main()
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation). | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK). | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |