buluslan/gpt-image2-ecommerce

E-commerce product image generator skill — 39 scenario templates, GPT-Image-2/2.5 (Flare/Sunburst) model routing, campaign consistency lock, platform compliance. 电商做图 Skill：39 场景模板×双模型路由×套图一致性×平台合规。

What it solves

This tool simplifies the creation of professional e-commerce product imagery for cross-border sellers. It eliminates the need for manual prompt engineering by providing a library of 39 scene templates and a structured workflow to generate high-quality, platform-compliant images (e.g., for Amazon, TikTok Shop, Shopify) while maintaining visual consistency across a full set of product listings.

How it works

The system operates as an "Agent-native" skill that can be integrated into AI coding agents like Claude Code or Cursor. It takes product materials and requirements as input and follows these steps:

  1. Template Matching: It matches the request to one of 39 scene templates (ranging from white-background hero shots to lifestyle and marketing banners).
  2. Prompt Assembly: It uses a "five-element assembly rule" (subject, material, light, lens/ratio, and finish) to build production-grade prompts.
  3. Model Routing: It automatically suggests either the gpt-image-2.5-flare model for fast generation or gpt-image-2.5-sunburst for high-precision editing.
  4. Generation: It can generate images via an API (OpenAI-compatible endpoints) or export a prompt package for manual use in ChatGPT.
  5. Technical Pre-check: A Python script (compliance_check.py) verifies if the output meets platform constraints, such as pure white backgrounds, foreground ratio, and OCR text presence.

Who it’s for

Cross-border e-commerce sellers and digital marketers who need professional product visuals without hiring a full studio, and those using AI agents to automate their creative workflows.

Highlights

  • 39 Scene Templates: Covers everything from macro details and flat-lays to A+ modules and motion GIFs.
  • Campaign Style Lock: A 10-field locking protocol and 6-layer funnel to ensure visual consistency across multiple images in a single listing.
  • Platform Compliance: Built-in technical checks for major platforms like Amazon and Temu to prevent listing rejections.
  • Channel Agnostic: Works with any OpenAI-compatible API or can be used entirely offline by exporting prompts.
  • Anti-AI Filtering: Implements text-level and visual-level filters to reduce "AI-looking" artifacts (slop words).

Related

  • Project
    wuyoscar/GPT-Image2-Skill

    A toolkit featuring a prompt gallery, agent skills, and a CLI for GPT Image 2 and 2.5 to simplify image generation, editing, and prompt extraction from reference images.

  • Project
    wzj177/ecommerce-image-suite

    An AI-powered toolkit for e-commerce sellers to automatically analyze product photos and generate a complete set of professional marketing images across various styles and platforms.

  • Project
    feichanggege/ecommerce-visual-copywriting-skill

    An AI‑agent skill that turns product data into a complete, compliance‑checked visual plan for e‑commerce listings (storyboard, per‑image design cards, prompts, and negative prompts).

  • Project
    freestylefly/awesome-gpt-image-2

    A Prompt-as-Code system and template library for GPT-Image2, providing over 500 reverse-engineered cases and structured prompts for stable, controllable image generation.

  • Project
    popopo-99/zy-cinematic-realism

    A cinematic visual direction framework for LLMs that transforms scripts and ideas into precise, consistent prompts for AI image generators.