[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-81396":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":11,"openIssues":13,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":14,"stars7d":14,"stars30d":14,"stars90d":14,"forks30d":14,"starsTrendScore":14,"compositeScore":15,"rankGlobal":9,"rankLanguage":9,"license":16,"archived":17,"fork":17,"defaultBranch":18,"hasWiki":19,"hasPages":17,"topics":20,"createdAt":9,"pushedAt":9,"updatedAt":41,"readmeContent":42,"aiSummary":43,"trendingCount":14,"starSnapshotCount":14,"syncStatus":44,"lastSyncTime":45,"discoverSource":46},81396,"image2-ads-studio","kwistzzqq-byte\u002Fimage2-ads-studio","kwistzzqq-byte","Advertising prompt agent for Image2: turns business briefs and reference images into debranded, production-ready image prompts.",null,"TypeScript",36,3,1,0,41.81,"Apache License 2.0",false,"main",true,[21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40],"ads-studio","advertising","advertising-automation","ai-agent","bilingual","creative-automation","design-tools","ecommerce","ecommerce-tools","gpt-image-2","image-generation","image-prompts","image2","llm","marketing-tools","openai","prompt-agent","prompt-engineering","prompt-optimizer","typescript","2026-06-12 04:01:33","# Image2 Ads Studio\n\n[中文文档](README.zh-CN.md) | [English gallery](examples\u002Fgallery\u002Fcases.md) | [中文案例库](examples\u002Fgallery\u002Fcases.zh-CN.md)\n\nAdvertising prompt agent for Image2: turn business briefs and reference images into debranded, production-ready image prompts.\n\n中文定位：面向 Image2 广告作图的开源 Prompt Agent，把客户白话需求、文案、行业和参考图转成可测试、可追踪、可集成的优化提示词。\n\nThis is not another prompt dump. Prompt libraries are useful for browsing examples; Image2 Ads Studio is a working prompt pipeline for advertising production. It parses a customer brief, retrieves advertising templates and visual recipes, rewrites references through an LLM prompt brain, validates deterministic execution details, and returns an optimized prompt that can be manually tested in Image2.\n\nThis repository is the **Community Edition**. It opens the agent framework, local Web workbench, all current 320 business templates, 240 visual recipes, LLM prompt brain interface, deterministic prompt checks, reference image policy, and a public case gallery.\n\n![Workflow](docs\u002Fassets\u002Fgithub-workflow.svg)\n\n## Why This Is Different\n\nMost Image2 prompt repositories answer: \"What prompt can I copy?\"\n\nImage2 Ads Studio answers: \"Given this business brief, copywriting, industry, aspect ratio, and reference image, what prompt should the agent produce?\"\n\n| Prompt libraries | Image2 Ads Studio |\n| --- | --- |\n| Curated examples for inspiration | Runnable advertising prompt agent |\n| Start from existing prompts | Start from customer brief and reference images |\n| General creative coverage | Advertising vertical: storefronts, posters, ecommerce, product ads, signage, brand walls, commercial photography |\n| Prompt text is the product | Parser, retriever, compiler, LLM rewrite, deterministic checks, and JSON records are the product |\n| Usually copy\u002Fpaste driven | Produces parsed intent, matched templates, visual recipes, design plan, optimized prompt, and source policy |\n| References may stay close to source | Public gallery prompts are regenerated, debranded, attributed, and made reusable |\n| Hard to connect to business systems | Designed for later ERP fields, template governance, adapters, and production workflows |\n\n## Prompt Gallery\n\nThe public gallery contains 100 image-to-prompt cases from project-generated concepts and multiple in-repo source libraries. Each case includes one preview image, English and Chinese optimized prompts, source attribution, and an in-repo composite source made from matched advertising templates plus visual recipes.\n\nView the full gallery: [English cases](examples\u002Fgallery\u002Fcases.md) \u002F [中文案例](examples\u002Fgallery\u002Fcases.zh-CN.md)\n\n| Preview | Optimized Prompt Excerpt |\n| --- | --- |\n| \u003Cimg src=\"docs\u002Fassets\u002Fgallery\u002Fcases\u002Fowner-beverage-ad-concept.jpg\" alt=\"Beverage product ad concept\" width=\"260\"> | **Beverage Product Ad Concept**\u003Cbr>\u003Cbr>`Create a commercially usable 16:9 beverage advertising hero image for a new debranded sparkling drink. Composition: place a tall matte-gloss red can slightly right of center...` |\n| \u003Cimg src=\"docs\u002Fassets\u002Fgallery\u002Fcases\u002Fowner-product-ad-concept.jpg\" alt=\"Premium appliance product ad concept\" width=\"260\"> | **Premium Appliance Product Ad Concept**\u003Cbr>\u003Cbr>`Create a 16:9 premium consumer-electronics product advertising hero image for a debranded high-end airflow appliance. Use a black-and-champagne-gold cylindrical device...` |\n| \u003Cimg src=\"docs\u002Fassets\u002Fgallery\u002Fcases\u002Fgallery-03-e-commerce-main-image-luxury-amber-perfume-ad.jpg\" alt=\"Luxury amber perfume ad prompt case\" width=\"260\"> | **Luxury Amber Perfume Ad**\u003Cbr>\u003Cbr>`Create a square 1:1 luxury beauty product advertisement for a debranded amber perfume bottle. Show one classic rectangular glass bottle as the hero object...` |\n| \u003Cimg src=\"docs\u002Fassets\u002Fgallery\u002Fcases\u002Fgallery-05-e-commerce-main-image-tropical-citrus-soda-ad-poster.jpg\" alt=\"Tropical citrus soda ad prompt case\" width=\"260\"> | **Tropical Citrus Soda Ad Poster**\u003Cbr>\u003Cbr>`Create a 9:16 vertical commercial beverage poster for a tropical citrus soda campaign. Show one large transparent plastic bottle as the hero object...` |\n| \u003Cimg src=\"docs\u002Fassets\u002Fgallery\u002Fcases\u002Fgallery-32-moss-radio-brand-identity-showcase-board.jpg\" alt=\"Brand identity showcase prompt case\" width=\"260\"> | **Moss Radio Brand Identity Showcase Board**\u003Cbr>\u003Cbr>`Create a square 1:1 brand-identity showcase board for a fictional audio, retail, or culture-focused business. Build a dense editorial presentation...` |\n| \u003Cimg src=\"docs\u002Fassets\u002Fgallery\u002Fcases\u002Fgallery-36-zoo-visitor-wayfinding-map.jpg\" alt=\"Wayfinding map prompt case\" width=\"260\"> | **Zoo Visitor Wayfinding Map**\u003Cbr>\u003Cbr>`Create a polished 16:9 advertising-style wayfinding map board for a fictional wildlife park campaign, designed as a commercially usable tourism graphic...` |\n| \u003Cimg src=\"docs\u002Fassets\u002Fgallery\u002Fcases\u002Fgallery-47-e-commerce-product-detail-page-layout.jpg\" alt=\"Ecommerce product detail prompt case\" width=\"260\"> | **E-Commerce Product Detail Page Layout**\u003Cbr>\u003Cbr>`Create a 9:16 vertical e-commerce product detail poster for a futuristic consumer electronics hero product. Use a dense marketplace layout...` |\n\nSource and license notes are tracked in [gallery attribution](examples\u002Fgallery\u002FATTRIBUTION.md). The gallery is not a copied prompt collection: upstream prompts\u002Fimages are used as structure references, then this project regenerates, debrands, and normalizes the final prompt through the same agent flow used by the Web workbench.\n\n## Why It Exists\n\nGeneral image-generation prompts are unstable for advertising production. A local shop owner may say \"make a milk tea storefront signboard\", but production-ready output needs structure: task type, industry, copywriting, aspect ratio, material constraints, reference image policy, and clear composition.\n\nImage2 Ads Studio turns that loose brief into a structured advertising plan and an optimized prompt that can be manually tested in Image2 or another image-generation tool.\n\n## Core Features\n\n- Advertising vertical coverage: storefront signboards, posters, roll-up banners, event backdrops, wayfinding signage, brand walls, local promotions, ecommerce main images, product ads, and commercial photography.\n- Open template library: 320 business templates and 240 visual recipes.\n- LLM prompt brain: uses a Responses-compatible LLM endpoint to refine the rule prompt.\n- Reference image policy: user uploads can be read by the LLM and marked for later Image2 editing; upstream references are only used for visual understanding.\n- Deterministic prompt validation: removes vague expressions and requires composition, lighting, typography, material, and image-reference policy.\n- Local Web UI: prompt workbench with brief input, optimized prompt output, preview placeholder, and retrieval signals.\n\n## Quick Start\n\n```bash\npnpm install\npnpm --filter ad-image-agent-core test\npnpm --filter ad-image-agent build\nexport OPENAI_API_KEY=\"your_api_key\"\npnpm --filter ad-image-agent serve:llm\n```\n\nOpen:\n\n```text\nhttp:\u002F\u002F127.0.0.1:5174\n```\n\nThe app does not call an image-generation API in this edition. Copy the optimized prompt into your image-generation tool and upload user reference images when the app marks them for Image2 participation.\n\n## Architecture\n\n```mermaid\nflowchart LR\n  A[User Brief] --> B[Intent Parser]\n  B --> C[Template Retriever]\n  C --> D[Visual Recipe Retriever]\n  D --> E[LLM Prompt Brain]\n  E --> F[Optimized Prompt]\n  F --> G[Image2 Manual Test]\n  G --> H[Case Library]\n  H -. Template \u002F Recipe Update .-> C\n  H -. Template \u002F Recipe Update .-> D\n```\n\n## Community vs Commercial\n\n| Capability | Community Edition | Commercial Edition |\n| --- | --- | --- |\n| Local Web UI | Included | Included |\n| Core prompt agent framework | Included | Included |\n| Business templates | 320 open templates | Custom\u002Fprivate templates and governance |\n| Visual recipes | 240 open recipes | Custom\u002Fprivate recipes and scored evaluation |\n| LLM prompt brain | Interface + local server | Production setup |\n| Image generation | Manual adapter | Real adapters and storage |\n| ERP integration | Overview docs | Connectors, field mapping, deployment |\n| Case gallery | Public samples | Private scored case library |\n\nCommercial work can add ERP connectors, real image-generation adapters, deployment support, custom private templates, and production case evaluation.\n\n## Repository Layout\n\n```text\napps\u002Fad-image-agent              Local Web workbench\npackages\u002Fad-image-agent-core     Parser, retrievers, compiler, adapters, tests\nexamples                         Sample briefs and generated records\ndocs                             Architecture, adapter, ERP, authoring, roadmap\nscripts                          Community validation and asset counting\n```\n\n## Development\n\n```bash\npnpm --filter ad-image-agent-core test\npnpm --filter ad-image-agent-core build\npnpm --filter ad-image-agent typecheck\npnpm --filter ad-image-agent build\n```\n\n## License\n\nApache-2.0. See [LICENSE](LICENSE).\n","Image2 Ads Studio 是一个面向广告制作的开源Prompt Agent，能够将商业简报和参考图片转化为去品牌化、可直接投入生产的图像提示。该项目采用TypeScript编写，通过解析客户需求、调用广告模板与视觉配方，并利用LLM进行重写优化，最终生成适用于电商、产品广告等场景的设计方案。其核心功能包括需求解析、模板匹配、视觉元素编译以及确定性检查等，确保输出高质量且符合业务要求的图像提示词。适合于需要高效自动化创意设计流程的企业或个人使用，尤其在电子商务、品牌宣传等领域具有广泛应用前景。",2,"2026-06-11 04:04:54","CREATED_QUERY"]