Token导航 LogoToken导航

Agent Skill · 图像处理

product-image-generator(产品图片生成器)

用于辅助图像生成、图片编辑、视觉素材处理或图像模型工作流。它适合让 Agent 根据文本生成图片、处理背景、整理视觉提示词或调用相关图像工具。使用时需要确认输入图片、版权来源、输出格式和模型限制;涉及人物、品牌、商品或公开展示素材时,应额外核对授权、真实性和内容合规边界。

product-image-generator 是图像处理类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为只读;许可证为 MIT-0,第三方扫描有提醒项,按安装量在图像处理类的 1,562 个 Skill 里排第 259。

总安装 12,677周安装 539GitHub Stars 公开资料未说明下载 4,441核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.0.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:product-image-generator(产品图片生成器)
来源仓库:https://github.com/boyd4y/product-image-generator
安装命令:
openclaw skills install product-image-generator
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install product-image-generator

这个 Skill 是做什么的

图像处理类 Skill 处理生成、裁剪、压缩和批量转换图片的任务。

为电商平台生成专业级产品图片,支持多风格与场景。优化商品展示效果,提升点击转化与购买决策效率。内置背景去除与构图建议,适配不同渠道尺寸。product-image-generator 属于图像处理类 Skill,可作为该场景下的辅助能力补充。

安装命令:openclaw skills install product-image-generator。涉及品牌露出时需确认肖像权与商标使用许可。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install product-image-generator;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

product-image-generator 怎么安装?

在终端执行 openclaw skills install product-image-generator。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

product-image-generator 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

product-image-generator 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

product-image-generator 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

product-image-generator 用的人多吗?

公开数据是累计安装 12,677 次,近一周安装 539 次,下载 4,441 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

product-image-generator 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理图像处理类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

图像处理类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
ai-image-generation

AI 图像生成 · Codex、Claude

4,365,552382调用外部服务未明确有提醒
openai-image-gen

OpenAI 图像生成 · OpenClaw

1,906,88432需要联网MIT-0有提醒
image

图像 · OpenClaw

525,99926需要联网MIT-0通过
baoyu-cover-image

宝玉封面图 · Codex、Claude

474,24016,779只读MIT有提醒
baoyu-image-gen

宝玉图像生成器 · Codex、Claude

468,00716,778接触敏感数据MIT通过
product-image-generator

产品图片生成器 · 本页

12,677—只读MIT-0有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
product-image-generator
description
Generates professional product images for e-commerce platforms (Amazon, Shopify, eBay, etc.). Supports 8 visual styles and 6 scene types optimized for different product categories. Use when user mentions "商品图", "product images", "Amazon listing", "电商图片", or needs e-commerce product photography.
metadata
version
1.0.0

E-commerce Product Image Generator

Create professional product images optimized for e-commerce platforms with platform-specific requirements and multiple visual styles.

Usage

# Auto-select style based on product
/product-image-generator product-description.md

# Specify style
/product-image-generator product.md --style minimal

# Specify platform (auto-adjusts image requirements)
/product-image-generator product.md --platform amazon

# Specify scene type
/product-image-generator product.md --scene studio

# Combine options
/product-image-generator product.md --style premium --platform shopify --scene lifestyle

# Direct input
/product-image-generator
[paste product description]

# Direct input with options
/product-image-generator --style minimal --scene studio
[paste product description]

# With reference image (for style consistency)
/product-image-generator product.md --ref brand-style.png

# Multiple reference images
/product-image-generator product.md --ref style.png --ref competitor.jpg

Options

OptionDescription
--style <name>Visual style (see Style Gallery)
--scene <type>Scene type (see Scene Gallery)
--platform <name>E-commerce platform (auto-adjusts requirements)
--ref <path>Reference image(s) for style consistency

Three Dimensions

DimensionControlsOptions
StyleVisual aesthetics: mood, color treatmentminimal, premium, lifestyle, bold, soft, tech, natural, luxury
SceneBackground and contextstudio, lifestyle, contextual, exploded, comparison, infographic
PlatformTechnical requirementsamazon, shopify, ebay, etsy, taobao, jd, pinduoduo

Style × Scene × Platform can be freely combined. Example: --style premium --scene lifestyle --platform amazon creates a high-end lifestyle shot meeting Amazon's image requirements.

Style Gallery

StyleDescriptionBest For
minimal (Default)Clean, white background, focus on productElectronics, accessories, professional products
premiumSophisticated, elegant lighting, subtle shadowsLuxury goods, jewelry, high-end cosmetics
lifestyleProduct in natural use contextHome goods, fashion, outdoor products
boldHigh contrast, vibrant, attention-grabbingSports products, gaming, youth-oriented items
softGentle lighting, pastel tones, warmBaby products, skincare, wellness items
techFuturistic, sleek, modernElectronics, gadgets, software, AI products
naturalOrganic, eco-friendly, earthy tonesSustainable products, food, natural cosmetics
luxuryGold accents, dramatic lighting, rich colorsPremium brands, jewelry, high-end fashion

Detailed style definitions: references/presets/<style>.md

Scene Gallery

SceneDescriptionImage Count
studio (Default)Pure white/gradient background, professional studio lighting1-3 images
lifestyleProduct in real-life usage scenario2-5 images
contextualProduct in environment, staged setting2-4 images
explodedComponent breakdown, features highlighted3-6 images
comparisonBefore/after, product vs alternatives2-4 images
infographicFeatures, specs, benefits with text overlays2-5 images

Detailed scene definitions: references/elements/scene-guide.md

Platform Requirements

PlatformMain ImageAdditional ImagesNotes
amazon1000x1000+, pure whiteUp to 9 imagesMain image must be on pure white background
shopify1024x1024+ recommendedUnlimitedFlexible, but 1:1 or 4:3 recommended
ebay500x500+ minimumUp to 12 freeWhite background preferred
etsy760x760+Up to 10 imagesLifestyle shots perform well
taobao800x800+Up to 15 imagesInfographic style popular
jd800x800+Up to 15 imagesClean, professional style
pinduoduo750x750+Up to 10 imagesValue-focused presentation

Platform-specific guidelines: references/platforms/<platform>.md

Auto Selection

Product CategoryStyleScene
Electronics, gadgetstech or minimalstudio + exploded
Fashion, accessoriespremium or lifestylelifestyle + contextual
Beauty, cosmeticssoft or premiumstudio + lifestyle
Home, furniturelifestyle or naturalcontextual + lifestyle
Sports, outdoorbold or lifestylelifestyle + contextual
Jewelry, luxuryluxury or premiumstudio + lifestyle
Baby, kidssoft or naturallifestyle + contextual
Food, supplementsnatural or minimalstudio + infographic
Tools, hardwarebold or techexploded + infographic

Outline Strategies

Strategy A: Product-Focused (产品聚焦型)

AspectDescription
ConceptProduct as hero, clean presentation
FeaturesMultiple angles, detail shots, pure backgrounds
Best forElectronics, accessories, luxury items
StructureHero shot → Angle variations → Detail close-ups → Scale/context

Strategy B: Lifestyle-Focused (场景融入型)

AspectDescription
ConceptProduct in natural use environment
FeaturesContext-rich, emotional connection, aspirational
Best forHome goods, fashion, outdoor products
StructureLifestyle hero → Usage scenario → Benefits in context → Social proof

Strategy C: Information-Focused (信息传达型)

AspectDescription
ConceptFeatures and benefits clearly communicated
FeaturesText overlays, callouts, comparisons
Best forComplex products, tools, supplements
StructureHero with key benefit → Feature breakdown → Comparison → Specs/usage

File Structure

Each session creates an independent directory named by product slug:

product-images/{product-slug}/
├── source-{slug}.{ext}             # Source files (description, reference images)
├── analysis.md                     # Product analysis + positioning
├── outline-strategy-a.md           # Strategy A: Product-focused
├── outline-strategy-b.md           # Strategy B: Lifestyle-focused
├── outline-strategy-c.md           # Strategy C: Information-focused
├── outline.md                      # Final selected/merged outline
├── prompts/
│   ├── 01-hero-[slug].md
│   ├── 02-detail-[slug].md
│   └── ...
├── 01-hero-[slug].png
├── 02-detail-[slug].png
└── NN-infographic-[slug].png

Slug Generation:

  1. Extract product name (2-4 words, kebab-case)
  2. Example: "无线蓝牙耳机" → wireless-earbuds

Conflict Resolution: If product-images/{product-slug}/ already exists:

  • Append timestamp: {product-slug}-YYYYMMDD-HHMMSS

Workflow

Progress Checklist

Product Image Generation Progress:
- [ ] Step 0: Check preferences (EXTEND.md) ⛔ BLOCKING
- [ ] Step 1: Analyze product → analysis.md
- [ ] Step 2: Confirmation 1 - Product understanding ⚠️ REQUIRED
- [ ] Step 3: Generate 3 outline + style variants
- [ ] Step 4: Confirmation 2 - Outline & style & platform ⚠️ REQUIRED
- [ ] Step 5: Generate images (sequential)
- [ ] Step 6: Completion report

Flow

Input → [Step 0: Preferences] ─┬─ Found → Continue
                               │
                               └─ Not found → First-Time Setup ⛔ BLOCKING
                                              │
                                              └─ Complete setup → Save EXTEND.md → Continue
                                                                                      │
        ┌─────────────────────────────────────────────────────────────────────────────┘
        ↓
Analyze → [Confirm 1] → 3 Outlines → [Confirm 2: Outline + Style + Platform] → Generate → Complete

Step 0: Load Preferences (EXTEND.md) ⛔ BLOCKING

Purpose: Load user preferences or run first-time setup.

CRITICAL: If EXTEND.md not found, MUST complete first-time setup before ANY other steps.

Use Bash to check EXTEND.md existence:

# Check project-level first
test -f .teamclaw-skills/product-image-generator/EXTEND.md && echo "project"

# Then user-level
test -f "$HOME/.teamclaw-skills/product-image-generator/EXTEND.md" && echo "user"
ResultAction
FoundRead, parse, display summary → Continue to Step 1
Not found⛔ BLOCKING: Run first-time setup → Save EXTEND.md → Step 1

First-Time Setup (when EXTEND.md not found):

Use AskUserQuestion with ALL questions in ONE call:

  1. Default platform preference: amazon / shopify / ebay / etsy / taobao / jd / pinduoduo / no preference
  2. Default style preference: minimal / premium / lifestyle / bold / soft / tech / natural / luxury
  3. Watermark: Enable / Disable (if enable, specify text and position)
  4. Language: Chinese / English / Auto-detect

Schema: references/config/preferences-schema.md

Step 1: Analyze Product → analysis.md

Actions:

  1. Save source content:

- If file path provided: use as-is - If pasted content: save to source.md - Backup rule: If exists, rename to source-backup-YYYYMMDD-HHMMSS.md

  1. Read and analyze product information
  2. Product analysis:

- Category identification - Target audience - Key features and selling points - Competitive positioning - Visual opportunities

  1. Detect product language
  2. Determine recommended image count and types
  3. Generate clarifying questions (see Step 2)
  4. Save to analysis.md

Step 2: Confirmation 1 - Product Understanding ⚠️

Purpose: Validate understanding + collect missing info.

Display summary:

  • Product category identified
  • Key features extracted
  • Target audience
  • Recommended platform match

Use AskUserQuestion for:

  1. Primary selling point (multiSelect: true):

- Design/aesthetics - Functionality/features - Price/value - Quality/durability - Brand/status - Innovation/technology - Sustainability/eco-friendly - Convenience/ease of use

  1. Target customer:

- Budget-conscious - Quality-focused - Luxury/premium - Tech-savvy - Eco-conscious - Family-oriented - Professional/business - Youth/trend-focused

  1. Main use scenario:

- Indoor/home - Outdoor - Office/work - Travel - Sports/fitness - Social/events - Auto

  1. Additional context (optional)

After response: Update analysis.md → Step 3

Step 3: Generate 3 Outline + Style Variants

Create three distinct strategy variants, each with outline structure and visual style recommendation.

StrategyFilenameFocusRecommended Style
Aoutline-strategy-a.mdProduct-focusedminimal, tech
Boutline-strategy-b.mdLifestyle-focusedlifestyle, soft, natural
Coutline-strategy-c.mdInformation-focusedbold, infographic-style

Outline format (YAML front matter + content):

---
strategy: a  # a, b, or c
name: Product-Focused
style: minimal
style_reason: "Clean presentation highlights product design and build quality"
scene: studio
platform: amazon
image_count: 5
---

## P1 Hero Shot
**Type**: hero
**Purpose**: Main product image, first impression
**Visual**: Product on pure white background, professional lighting
**Platform**: Amazon main image compliant (1000x1000, pure white)

## P2 Angle Variation
**Type**: angle
**Purpose**: Show product from different perspective
**Visual**: 45-degree angle, slight shadow for depth

## P3 Detail Close-up
**Type**: detail
**Purpose**: Highlight key feature or quality detail
**Visual**: Macro shot of texture/material/connection point

...

After response: Update analysis.md → Step 3

Step 4: Confirmation 2 - Outline & Style & Platform Selection ⚠️

Purpose: User chooses strategy, confirms style and platform.

Display each strategy:

  • Strategy name + image count + recommended style
  • Image-by-image summary

Use AskUserQuestion with questions:

Question 1: Outline Strategy

  • Strategy A (Product-focused)
  • Strategy B (Lifestyle-focused)
  • Strategy C (Information-focused)
  • Combine: specify images from each

Question 2: Visual Style

  • Use strategy's recommended style
  • Or select from: minimal / premium / lifestyle / bold / soft / tech / natural / luxury
  • Or type custom style description

Question 3: Platform

  • Use strategy's recommended platform
  • Or select: amazon / shopify / ebay / etsy / taobao / jd / pinduoduo
  • Custom: specify requirements

Question 4: Scene Type

  • Use strategy's recommended scene
  • Or select: studio / lifestyle / contextual / exploded / comparison / infographic

After response:

  • Single strategy → copy to outline.md with confirmed settings
  • Combination → merge specified images
  • Update outline.md frontmatter with final settings

Step 5: Generate Images

With confirmed outline + style + scene + platform:

Visual Consistency — Reference Image Chain:

  1. Check for user-provided reference images (--ref option)

- If provided: Use as primary style reference - If not provided: Use internal reference chain (see below)

  1. Internal Reference Chain (when no user ref provided):

- Generate image 1 (hero) FIRST — without --ref - Use image 1 as --ref for ALL remaining images (2, 3, ..., N)

For each image:

  1. Save prompt to prompts/NN-{type}-[slug].md

- Backup rule: If exists, rename to prompts/NN-{type}-[slug]-backup-YYYYMMDD-HHMMSS.md - Include reference image paths in prompt

  1. Generate image:

- With user ref: --ref <user-reference.png> for all images - Image 1 (no user ref): Generate without --ref (establishes visual anchor) - Images 2+ (no user ref): Generate with --ref <image-01-path> - Backup rule: If image exists, rename with timestamp

  1. Report progress after each generation

Reference Image Best Practices:

Reference TypeUsageEffect
Style referenceBrand guideline imagesMaintains brand consistency
Competitor referenceSimilar product imagesMatches category standards
Internal chain (image 1)First generated imageEnsures series consistency

Platform Compliance:

  • Apply platform-specific requirements automatically
  • Amazon: Pure white background for main image
  • Size adjustments based on platform

Session Management: Use consistent session ID for all images in the set: product-{slug}-{timestamp}

Step 6: Completion Report

Product Image Set Complete!

Product: [product name]
Strategy: [A/B/C/Combined]
Style: [style name]
Scene: [scene type]
Platform: [platform] (requirements applied)
Location: [directory path]
Images: N total

✓ analysis.md
✓ outline-strategy-a.md
✓ outline-strategy-b.md
✓ outline-strategy-c.md
✓ outline.md (selected: [strategy])

Files:
- 01-hero-[slug].png ✓ Main image ([platform] compliant)
- 02-angle-[slug].png ✓ Angle variation
- 03-detail-[slug].png ✓ Detail shot
...

Image Modification

ActionSteps
EditUpdate prompt file FIRST → Regenerate with same session ID
AddSpecify position → Create prompt → Generate → Renumber files
DeleteRemove files → Renumber subsequent → Update outline

Platform-Specific Guidelines

Amazon

  • Main image: Pure white background (RGB 255,255,255)
  • Minimum 1000x1000 pixels for zoom feature
  • Product must fill 85%+ of frame
  • No text, logos, or watermarks on main image
  • Additional images: lifestyle, infographic, comparison allowed

Shopify

  • Recommended 1024x1024 or 4:3 aspect ratio
  • More flexible styling
  • Lifestyle images perform well
  • Consistent style across all product images

eBay

  • Minimum 500x500 pixels
  • White background preferred but not required
  • Up to 12 free images per listing

Content Breakdown Principles

  1. Hero Image (Image 1): Main product shot, platform-compliant
  2. Angle Variations: Show product from multiple perspectives
  3. Detail Shots: Highlight quality, features, materials
  4. Lifestyle/Context: Product in use, emotional connection
  5. Infographic: Features, specs, benefits with callouts
  6. Comparison: Before/after, vs alternatives
  7. Social Proof: Reviews, testimonials (if applicable)

References

Detailed templates in references/ directory:

Elements:

  • elements/scene-guide.md - Scene types and compositions
  • elements/lighting.md - Lighting setups and moods
  • elements/composition.md - Framing and composition rules

Presets:

  • presets/<style>.md - Style definitions and prompts

Workflows:

  • workflows/analysis-framework.md - Product analysis
  • workflows/outline-template.md - Outline templates
  • workflows/prompt-assembly.md - Prompt assembly

Platforms:

  • platforms/amazon.md - Amazon requirements
  • platforms/shopify.md - Shopify requirements
  • platforms/ebay.md - eBay requirements
  • platforms/etsy.md - Etsy requirements
  • platforms/taobao.md - Taobao requirements
  • platforms/jd.md - JD.com requirements
  • platforms/pinduoduo.md - Pinduoduo requirements

Config:

  • config/preferences-schema.md - EXTEND.md schema
  • config/watermark-guide.md - Watermark configuration

Notes

  • Auto-retry once on failure
  • Platform requirements auto-applied
  • Two confirmation points required (Steps 2 & 4)
  • Reference image chain ensures visual consistency

Extension Support

Custom configurations via EXTEND.md. See Step 0 for paths.

平台分布

OpenClaw

77.71%
按下载量换算3,451

安全审计

VirusTotal

可疑

ClawScan

可疑

Static analysis

未展示

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills