Token导航 LogoToken导航

Agent Skill · 研究检索

using-content-driven-development(使用内容驱动开发)

用于辅助文档、README、Markdown、说明文和内容稿件的整理与改写。它适合让 Agent 提炼结构、补齐章节、统一术语、检查链接或把零散材料整理成可读文档。使用时应保留项目已有事实、命令和路径,不要把未确认的信息写成确定结论;涉及对外文案时,还需要控制语气,避免过度营销或夸大能力。

using-content-driven-development 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 7,653。

总安装 18,408周安装 460GitHub Stars 40下载 5,408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:using-content-driven-development(使用内容驱动开发)
来源仓库:https://github.com/adobe/helix-website
仓库路径:skills/using-content-driven-development
安装命令:
npx skills add https://github.com/adobe/helix-website --skill 'Using Content Driven Development'
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/adobe/helix-website --skill 'Using Content Driven Development'

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于辅助文档、README、Markdown 和内容稿件的整理与改写。适合提炼结构、补齐章节、统一术语或检查链接。使用时需保留项目已有事实和路径,避免写成确定结论。涉及对外文案时应控制语气,避免过度营销或夸大能力。

using-content-driven-development 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/using-content-driven-development

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/adobe/helix-website --skill 'Using Content Dri…;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

using-content-driven-development 怎么安装?

在终端执行 npx skills add https://github.com/adobe/helix-website --skill 'Using Content Driven Develo…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

using-content-driven-development 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。

using-content-driven-development 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。

using-content-driven-development 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。

using-content-driven-development 用的人多吗?

公开数据是累计安装 18,408 次,近一周安装 460 次,下载 5,408 次,GitHub 40 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。

using-content-driven-development 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。
  • 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

  • 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。
  • 需要处理研究检索类任务、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。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
using-content-driven-development

使用内容驱动开发 · 本页

18,40840只读未明确有提醒

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

Using Content Driven Development (CDD)

Content Driven Development is a mandatory process for AEM Edge Delivery Services development that prioritizes content and author needs over developer convenience. This skill orchestrates the development workflow to ensure code is built against real content with author-friendly content models.

Why Content-First Matters

Author needs come before developer needs. When building for AEM Edge Delivery, authors are the primary users of the structures we create. Content models must be intuitive and easy to work with, even if that means more complex decoration code.

Efficiency through preparation. Creating or identifying test content before coding provides:

  • Immediate testing capability: No need to stop development to create test content
  • Better PR workflows: Test content doubles as PR validation links for PSI checks
  • Living documentation: Test content often serves as author documentation and examples
  • Fewer assumptions: Real content reveals edge cases code-first approaches miss

NEVER start writing or modifying code without first identifying or creating the content you will use to test your changes.

When to Apply This Skill

Apply Content Driven Development principles to ALL AEM development tasks:

  • ✅ Creating new blocks
  • ✅ Modifying existing blocks (structural or functional changes)
  • ✅ Changes to core decoration functionality
  • ✅ Bug fixes that require validation
  • ✅ Any code that affects how authors create or structure content

Skip CDD only for:

  • ⚠️ Trivial CSS-only styling tweaks (but still identify test content for validation)
  • ⚠️ Configuration changes that don't affect authoring

When in doubt, follow the CDD process. The time invested pays dividends in quality and efficiency.

Related Skills

This skill orchestrates other skills at the appropriate stages:

  • content-modeling: Invoked when new content models need to be designed or existing models modified
  • building-blocks: Invoked during implementation phase for block creation or modification
  • testing-blocks: Referenced during validation phase for comprehensive testing
  • block-collection-and-party: Used to find similar blocks and reference implementations

The Content-First Process

Follow these phases in order. Do not skip steps.

Phase 1: Content Discovery and Modeling

The first phase establishes what content you're working with and ensures the content model is author-friendly.

Step 1.1: Determine Content Availability

For new blocks:

Skip to Step 1.2 (Content Model Design). Searching for content that doesn't exist is a waste of time.

For modifications to existing blocks:

Ask the user: "Does content using this block already exist that we can use for testing?"

  • YES → Identify existing content to test against

- Use the scripts/find-block-content.js script to search for pages containing the block - Or ask the user: "What are the path(s) to page(s) with this block?" - Validate the content loads correctly in your local dev environment - Proceed to Phase 2 (skip content modeling if structure isn't changing)

  • NO existing content → Proceed to Step 1.2
Step 1.2: Content Model Design

REQUIRED for:

  • All new blocks
  • Structural changes to existing blocks (adding/removing/modifying sections, variants, or the authoring structure)

Ask the user: "This requires a new content model. Would you like me to use the content-modeling skill to design an author-friendly content model now?"

  • YES → Invoke the content-modeling skill

- Follow the content modeling process completely - Return to this skill when content model is defined - Proceed to Step 1.3

  • NO → The user may want to define it themselves

- Ask: "Please describe the content structure authors will use" - Document their description for reference - Proceed to Step 1.3

Step 1.3: Content Creation

Once the content model is defined (from Step 1.2), you need test content.

Ask the user: "We need test content for development and validation. This content will serve multiple purposes:

  • Testing during development
  • PR validation link for PSI checks
  • Author documentation and examples

Would you like to:

  1. Create this content in the CMS now (Google Drive/SharePoint/DA/Universal Editor)
  2. Create temporary local HTML files for testing (will need CMS content before PR)"

Option 1: CMS Content (Recommended)

  • Guide the user through creating content in their CMS
  • Wait for user confirmation that content is created and published
  • Get the content URL(s) from the user
  • Validate content loads in local dev environment
  • Proceed to Phase 2

Option 2: Local HTML Files (Temporary)

  • Create HTML file(s) in drafts/ folder matching the content model structure
  • Reference the HTML Structure Guide for proper file format
  • Remind user: "Restart your dev server with: aem up --html-folder drafts"
  • Note: "You will need to create actual CMS content before raising a PR"
  • Proceed to Phase 2
Making Test Content Serve as Author Documentation

Test content can often double as author-facing documentation, saving time and keeping documentation current. Consider this when creating test content:

When test content IS sufficient as author documentation:

  • The block is straightforward with clear patterns
  • Test content shows all variants and use cases
  • Content demonstrates best practices authors should follow
  • Examples are realistic and relatable to actual use cases

When separate author documentation is needed:

  • Block has complex configuration or many variants requiring explanation
  • There are edge cases or gotchas authors need to understand
  • Project standards require formal documentation in a specific location/format
  • Block behavior isn't self-evident from examples alone

Structuring test content to serve both purposes:

  1. Create comprehensive examples: Show all variants, edge cases, and common patterns
  2. Use realistic content: Avoid "lorem ipsum" or technical placeholders
  3. Demonstrate best practices: Structure content the way authors should
  4. Consider location: Place content where it can serve as documentation

- Sidekick Library projects: Consider creating in /tools/sidekick/library/ or appropriate library location - Document Authoring: Place in DA Library structure - Simple documentation: Use /drafts/docs/ or /drafts/library/ - Universal Editor: Follow project-specific documentation patterns

Ask the user about documentation approach: "Should this test content also serve as author documentation? If so, we can structure it accordingly and place it in an appropriate location (e.g., /drafts/library/{block-name} or your project's library system)."

If yes, guide content creation with documentation in mind. If no, proceed with test-focused content and note that author documentation will be needed later.

Phase 2: Implementation

CRITICAL: Do not begin Phase 2 until you have confirmed test content exists and is accessible.

Now that test content exists, proceed with implementation:

For Block Development

Invoke the building-blocks skill:

  • Provide the skill with the content model and test content URL(s)
  • Follow the building-blocks process for implementation
  • Return to this skill when implementation is complete
  • Proceed to Phase 3
For Core Functionality Changes

Follow standard development practices:

  • Make changes to scripts, styles, or configuration
  • Test against the identified content throughout development
  • Ensure changes don't break existing blocks or content models
  • Proceed to Phase 3

Phase 3: Validation

The final phase ensures the implementation works correctly with real content.

Step 3.1: Test with Real Content

Mandatory testing:

  • ✅ View test content in local dev environment
  • ✅ Verify all variants render correctly
  • ✅ Check responsive behavior (mobile, tablet, desktop)
  • ✅ Test edge cases revealed by the actual content
  • ✅ Validate accessibility basics (keyboard navigation, screen reader friendly)
Step 3.2: Run Quality Checks

Required before considering implementation complete:

npm run lint

If linting fails, fix issues with:

npm run lint:fix
Step 3.3: Comprehensive Testing

The testing-blocks skill is automatically invoked by building-blocks for block development.

For other code changes, or for additional testing guidance, invoke the testing-blocks skill which provides:

  • Unit testing strategies for logic-heavy utilities
  • Browser testing with Playwright/Puppeteer
  • Linting and code quality checks
  • Performance validation with GitHub checks
  • Guidance on keeper vs throwaway tests
Step 3.4: PR Preparation

Before raising a PR, ensure:

  • ✅ Test content exists in the CMS (not just local HTML)
  • ✅ Test content URL is accessible for PSI checks
  • ✅ All linting passes
  • ✅ Author documentation is updated (if applicable)

The test content URL will be used as the PR validation link.

Anti-Patterns to Avoid

Common mistakes:

  • ❌ Starting with code before understanding the content model
  • ❌ Making assumptions about content structure without seeing real examples
  • ❌ Creating developer-friendly but author-hostile content models
  • ❌ Skipping content creation "to save time" (costs more time later)
  • ❌ Testing against imagined content instead of real content
  • ❌ Treating test content creation as separate from development workflow

Workflow Summary

Quick reference for the CDD process:

1. CONTENT DISCOVERY
   └─ Existing content? → Use it
   └─ New block/structure? → Design content model → Create test content

2. IMPLEMENTATION
   └─ Build code against the real content model
   └─ Test continuously with actual content

3. VALIDATION
   └─ Comprehensive testing with test content
   └─ Quality checks (linting, accessibility)
   └─ PR preparation with test URL

KEY RULE: Never proceed to implementation without test content

Scripts and Tools

Finding Existing Block Content

Use the provided script to search for pages containing a specific block:

# Search on localhost (default)
node .claude/skills/content-driven-development/scripts/find-block-content.js <block-name>

# Search for specific variant
node .claude/skills/content-driven-development/scripts/find-block-content.js <block-name> localhost:3000 <variant>

# Search on live site
node .claude/skills/content-driven-development/scripts/find-block-content.js <block-name> main--repo--owner.aem.live

# Search on preview with variant
node .claude/skills/content-driven-development/scripts/find-block-content.js <block-name> main--repo--owner.aem.page <variant>

Examples:

node .claude/skills/content-driven-development/scripts/find-block-content.js hero
node .claude/skills/content-driven-development/scripts/find-block-content.js hero localhost:3000 dark
node .claude/skills/content-driven-development/scripts/find-block-content.js cards main--site--owner.aem.live three-up

This script queries the site's query-index to find all pages containing the specified block (and optional variant) and returns their URLs. The script uses proper DOM parsing to accurately identify blocks.

Integration with Other Skills

This skill acts as the orchestrator for AEM development workflows:

At Content Modeling stage: → Invoke content-modeling skill for author-friendly design

At Implementation stage: → Invoke building-blocks skill for block development → Reference block-collection-and-party skill for patterns

At Validation stage: → Reference testing-blocks skill for comprehensive testing

Following this orchestration ensures all development follows content-first principles.

平台分布

Codex

37.67%
按下载量换算2,037

Claude

28.43%
按下载量换算1,537

Cursor

16.79%
按下载量换算908

Gemini CLI

9.07%
按下载量换算491

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

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