Token导航 LogoToken导航

Agent Skill · 图像处理

og-image-creator(og 图像创建者)

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。

og-image-creator 是图像处理类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为操作浏览器;许可证信息还不明确,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 1,545。

总安装 188周安装 8GitHub Stars 23下载 66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
unknown

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:og-image-creator(og 图像创建者)
来源仓库:https://github.com/chongdashu/cc-skills
仓库路径:skills/og-image-creator
安装命令:
npx skills add https://github.com/chongdashu/cc-skills --skill og-image-creator
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 npx skills 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。

skills.shnpx skills
npx skills add https://github.com/chongdashu/cc-skills --skill og-image-creator

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。

用于辅助图像生成、图片编辑和视觉素材处理。适合根据文本生成图片、处理背景或整理视觉提示词。使用时需确认输入图片、版权来源、输出格式和模型限制。涉及人物、品牌或公开展示素材时应核对授权与合规边界。og-image-creator 属于图像处理类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/og-image-creator

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 Codex、Claude、Cursor、Gemini CLI;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 npx skills add https://github.com/chongdashu/cc-skills --skill og-image-creator;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

    它需要操作浏览器,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

og-image-creator 怎么安装?

在终端执行 npx skills add https://github.com/chongdashu/cc-skills --skill og-image-creator。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

og-image-creator 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。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。

og-image-creator 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

og-image-creator 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

og-image-creator 用的人多吗?

公开数据是累计安装 188 次,近一周安装 8 次,下载 66 次,GitHub 23 星。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。

og-image-creator 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

  • 同时适配 4 种工具,换客户端不用重新找替代品。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 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 Codex,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。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 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
og-image-creator

og 图像创建者 · 本页

18823操作浏览器未明确有提醒

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

OG Image Creator

Generate authentic, brand-aligned Open Graph images by understanding your codebase first, then creating contextually appropriate images for each route using Playwright and your existing components.

Philosophy: Authentic Over Template

Before generating OG images, ask:

  • What is the brand identity of this site (colors, fonts, logo, design language)?
  • What routes/pages exist and what are their purposes?
  • What components and assets can be reused for consistency?
  • What page types need different visual treatments?

Core principles:

  1. Study before generate: Understand the codebase structure, routing, and brand before creating anything
  2. Authenticity over templates: Use actual brand colors, fonts, and components—not generic social card templates
  3. Context-aware design: Landing pages, articles, products, and docs need different visual approaches
  4. Smart automation: Auto-detect routes and extract brand, but allow user customization

Workflow

1. Discover and Analyze

Route discovery:

  • Analyze project structure to determine framework (Next.js, Astro, React, etc.)
  • Auto-detect routes based on framework conventions:

- Next.js: app/*/page.tsx or pages/*.tsx - Astro: src/pages/*.astro - React Router: Analyze route configuration

  • Extract route metadata (title, description) from pages
  • Categorize page types (landing, article, product, about, etc.)

Brand extraction:

  • Find logo/icon files (SVG preferred)
  • Extract color palette from CSS/Tailwind config
  • Identify primary fonts from stylesheets or config
  • Capture design patterns from existing components

Use scripts/analyze_codebase.py to automate discovery.

2. Design Strategy

Determine appropriate layouts per page type:

  • Landing pages: Bold, minimal, emphasize value proposition
  • Articles/Blog posts: Title + excerpt, date/author if relevant
  • Product pages: Product name, key feature, visual if available
  • Documentation: Topic + description, organized/structured feel
  • About/Company: Brand-forward, professional

Layout principles:

  • Generous whitespace (avoid clutter)
  • High contrast for readability when shared
  • Brand colors as accents, not overwhelming
  • Typography hierarchy that matches site
  • 1200x630px (standard OG dimensions)

3. Generate Images

Component-based generation:

  • Reuse React components when possible for consistency
  • Use Playwright to render components to images
  • Apply brand colors, fonts, and assets
  • Inject actual page metadata (not placeholders)

Technical approach:

  1. Create minimal HTML template with brand styles
  2. For each route, generate page-specific markup
  3. Use Playwright to capture screenshot at 1200x630
  4. Save to public/og/ or appropriate static directory
  5. Update page metadata to reference OG image

Use scripts/generate_og_images.py for automation.

4. Verify and Optimize

  • Validate OG image metadata in HTML
  • Test with OG preview tools (opengraph.xyz, linkedin inspector)
  • Ensure file sizes are reasonable (<200KB preferred)
  • Check image appearance at different scales

Framework-Specific Guidance

Next.js

App Router:

// app/page.tsx
export const metadata = {
  openGraph: {
    images: ['/og/home.png'],
  },
}

Pages Router:

// pages/index.tsx
<Head>
  <meta property="og:image" content="/og/home.png" />
</Head>

Dynamic OG images: Consider Next.js OG Image Generation (@vercel/og) for dynamic content, but use static generation for brand consistency.

Astro

---
// src/pages/index.astro
const ogImage = '/og/home.png';
---
<head>
  <meta property="og:image" content={ogImage} />
</head>

React SPAs

Add OG tags to public/index.html or use react-helmet:

<Helmet>
  <meta property="og:image" content="/og/home.png" />
</Helmet>

Anti-Patterns to Avoid

❌ Generic templates: Don't use stock social card templates that ignore the site's brand

  • Why bad: Creates disconnect between site and social sharing experience
  • Better: Extract actual brand colors, fonts, and use them consistently

❌ One size fits all: Don't use the same OG image design for every page type

  • Why bad: Misses opportunity for context-appropriate presentation
  • Better: Different layouts for landing, articles, products, etc.

❌ Placeholder content: Don't use "Lorem ipsum" or generic text

  • Why bad: Defeats the purpose of preview images
  • Better: Use actual page titles and descriptions from metadata

❌ Overcrowding: Don't cram too much text, too many images, or complex layouts

  • Why bad: Illegible when scaled down in social feeds
  • Better: Focus on 1-2 key elements with generous whitespace

❌ Ignoring components: Don't create OG images from scratch if components exist

  • Why bad: Breaks visual consistency with the site
  • Better: Reuse button styles, card layouts, typography from components

❌ Manual generation: Don't create OG images one by one manually

  • Why bad: Not scalable, hard to maintain consistency
  • Better: Script the generation process to handle all routes

Variation Guidance

IMPORTANT: OG images should vary based on page context, not converge on a single design.

Dimensions to vary:

  • Layout: Centered vs left-aligned, single vs split column, text vs visual focus
  • Color emphasis: Primary brand color vs neutral, gradient vs solid
  • Content hierarchy: Large title vs balanced title+description
  • Visual elements: Logo size/placement, decorative elements, icons
  • Typography: Size ratios, weight distribution, font pairing

Avoid converging on:

  • Same gradient background for everything
  • Identical "title + logo" layout
  • Single font size ratio
  • Fixed logo position

Context drives variation:

  • Technical docs → Clean, organized, icon-based
  • Marketing landing → Bold, colorful, value-focused
  • Blog articles → Reader-friendly, author/date context
  • Product pages → Visual product emphasis

Technical Requirements

Dependencies (install as needed):

npm install playwright sharp
# or
pip install playwright Pillow

Playwright setup:

playwright install chromium

File structure:

project/
├── public/og/          # Generated OG images
│   ├── home.png
│   ├── about.png
│   └── ...
└── scripts/            # Generation scripts
    ├── analyze_codebase.py
    └── generate_og_images.py

Remember

OG images are an extension of your brand, not generic social cards.

Study the codebase to understand the design language. Extract the brand identity. Use actual components and content. Create contextually appropriate images that feel native to the site.

When someone shares your page, the OG image should feel like a natural preview—not a template slapped on top.

You're capable of creating sophisticated, brand-aligned OG images that enhance rather than undermine the site's identity.

平台分布

Codex

36.59%
按下载量换算24

Claude

28.5%
按下载量换算19

Cursor

18.33%
按下载量换算12

Gemini CLI

9.97%
按下载量换算7

安全审计

Gen Agent Trust Hub

可疑

Socket

通过

Snyk

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

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