AI SummaryGenerate e-commerce images using GPT-Image-2 via Codex CLI. Match user intent to structured JSON prompt templates, assemble concise prompts, and invoke image generation. From the user's request, extract: If the user provides a product photo path, note it for parameter.
Install
Copy this and paste it into Claude Code, Cursor, or any AI assistant:
I want to install the "ecom-image2" skill in my project. Please run this command in my terminal: # Install skill into your project mkdir -p .claude/skills/gpt-image2-ecommerce && curl --retry 3 --retry-delay 2 --retry-all-errors -o .claude/skills/gpt-image2-ecommerce/SKILL.md "https://raw.githubusercontent.com/buluslan/gpt-image2-ecommerce/main/SKILL.md" Then restart Claude Code (or reload the window in Cursor) so the skill is picked up.
Description
Use when generating e-commerce product images, advertising materials, or commercial photography using GPT-Image-2 via Codex CLI. Triggers on requests for product photography, promotional banners, social media assets, UGC-style images, packaging design, flat lay, model shots, livestream scenes, exploded views, ghost mannequin, magazine editorial, seasonal campaigns, luxury atmospherics, device mockups, storefront photography, sports campaigns, and other e-commerce visual content.
Overview
Generate e-commerce images using GPT-Image-2 via Codex CLI. Match user intent to structured JSON prompt templates, assemble concise prompts, and invoke image generation.
Step 1: Intent Recognition
From the user's request, extract: • Scene type: hero image, lifestyle, flat lay, macro detail, poster/banner, social media, UGC, model showcase, before/after, packaging, infographic, creative concept, size spec, multi-product, livestream, virtual try-on, exploded view, ghost mannequin, multi-angle grid, magazine editorial, seasonal campaign, luxury atmospherics, device mockup, storefront, sports campaign • Product info: category (beauty/electronics/food/fashion/home/jewelry/sports), description, material, key selling points • Style preference: luxury, fresh, tech, minimal, or other variant • Reference image: whether user provided a product photo path If the user provides a product photo path, note it for --image parameter.
Step 2: Template Matching
Read the matching template from references/templates/. Match by scanning keywords and trigger_phrases in each template: | Trigger Words | Template File | |---|---| | 白底图, 主图, hero image, packshot | 01-hero-image.json | | 场景图, 生活图, lifestyle | 02-lifestyle-scene.json | | 平铺图, flat lay, 俯拍 | 03-flat-lay.json | | 细节图, 微距, macro, 特写 | 04-detail-macro.json | | 海报, poster, banner, 促销 | 05-poster-banner.json | | 社交媒体, 小红书, Instagram, TikTok | 06-social-media.json | | UGC, 买家秀, GRWM | 07-ugc-style.json | | 模特, model, 人物展示 | 08-model-showcase.json | | 对比, before after, 前后 | 09-before-after.json | | 包装, packaging, 礼盒 | 10-packaging.json | | 信息图, A+, 详情页 | 11-infographic.json | | 创意, 概念, creative | 12-creative-concept.json | | 尺寸, 规格, 使用步骤 | 13-size-spec.json | | 套装, 组合, bundle | 14-multi-product.json | | 直播, livestream | 15-livestream.json | | 试穿, 融入, try on | 16-try-on-virtual.json | | 拆解图, 爆炸图, exploded view, 内部结构 | 17-exploded-view.json | | 隐形模特, ghost mannequin, 3D服装 | 18-ghost-mannequin.json | | 多角度, 网格, grid, 多色展示 | 19-multi-angle-grid.json | | 杂志, 封面, editorial, magazine | 20-magazine-editorial.json | | 季节, 四季, campaign, 春夏秋冬 | 21-seasonal-campaign.json | | 奢华, 氛围, 烟雾, luxury, atmospheric | 22-luxury-atmospherics.json | | 设备模型, 界面, mockup, SaaS, APP | 23-device-mockup.json | | 店铺, 门面, 空间, storefront, 实体店 | 24-storefront.json | | 运动, 健身, sports, fitness | 25-sports-campaign.json | No match → default to 01-hero-image.json. Only read the matched template file (progressive disclosure). Do not load all templates.
Step 3: Prompt Assembly
From the matched JSON template: • Take prompt_template as the base structure • Replace {variables} with user-provided info • If user specified a style variant → apply variants.<name>.overrides • If product category known → apply category_tips.<category> • Simplify: keep only core fields with values, remove empty/null fields • Output a concise JSON object (not the full template metadata) Key principle: keep prompts simple. Only include essential information. Image2 performs best with concise, focused prompts rather than overly complex ones. Example assembled prompt for a beauty hero image: `json { "type": "product photography", "subject": "frosted glass serum bottle with matte white cap", "background": "clean white background", "lighting": "soft diffused studio lighting", "composition": "centered, front view", "quality": "8K, commercial e-commerce photography", "category_note": "emphasize texture and glow" } `
Discussion
Health Signals
My Fox Den
Community Rating
Sign in to rate this booster