Token导航 LogoToken导航TokenDH.com
待分类需要联网github未标认证来源可访问许可证需确认审计通过

cdp-domain-knowledgeCDP 领域知识

Agent Skill

cdp-domain-knowledge 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

13,328

周安装

522

GitHub Stars

公开资料未说明

下载量

5,526
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

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:cdp-domain-knowledge(CDP 领域知识)
来源仓库:https://github.com/danhvb/my-ba-skills
仓库路径:skills/cdp-domain-knowledge
安装命令:
npx skills add https://github.com/danhvb/my-ba-skills --skill 'CDP Domain Knowledge'
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/danhvb/my-ba-skills --skill 'CDP Domain Knowledge'

简介

cdp-domain-knowledge 提供客户数据平台领域知识库,涵盖数据整合、分群和营销激活核心概念。

  • 适用于在 Codex、Claude、Cursor、Gemini CLI 中分析 CDP 需求或对比 DMP/CRM 差异时调用。
  • 区分第一方与第三方数据类型,强调身份解析与匿名数据处理规范。
  • 包含典型功能矩阵对比表,但不提供具体厂商产品评测。
  • 建议结合业务场景判断是否采用 CDP 架构而非传统 ETL 方案。

SKILL.md

CDP Domain Knowledge Skill

Purpose

Provide comprehensive CDP domain knowledge for analyzing requirements in customer data unification, segmentation, and marketing activation.

What is a CDP?

A Customer Data Platform (CDP) is a packaged software that:

  • Collects customer data from multiple sources
  • Creates unified customer profiles (360-degree view)
  • Enables segmentation and audience building
  • Activates data to marketing channels

CDP vs DMP vs CRM

CapabilityCDPDMPCRM
Data TypeFirst-party + Third-partyThird-party cookiesFirst-party
IdentityKnown + AnonymousMostly anonymousKnown customers
Data PersistenceLong-termShort-term (90 days)Long-term
Primary UsePersonalization, AnalyticsAdvertisingSales/Service
User ManagementIndividualsSegments/AudiencesAccounts/Contacts

CDP Architecture

Data Flow

Data Sources → Ingestion → Identity Resolution → Profile Unification →
Segmentation → Audience Building → Channel Activation

Core Components

1. Data Collection:

  • Website/app events (page views, clicks, transactions)
  • CRM data (contacts, accounts, opportunities)
  • Transaction data (orders, returns, support tickets)
  • Marketing data (email opens, campaign responses)
  • Offline data (in-store purchases, call center)
  • Third-party data (demographics, intent)

2. Identity Resolution:

  • Deterministic matching (email, phone, customer ID)
  • Probabilistic matching (device fingerprinting, behavior)
  • Cross-device stitching
  • Household linkage
  • Anonymous to known conversion

3. Profile Unification:

  • Golden record creation
  • Attribute standardization
  • Conflict resolution (which source wins?)
  • Profile enrichment
  • Historical event storage

4. Segmentation:

  • Rule-based segments
  • AI/ML-based segments
  • Dynamic vs. static segments
  • Lookalike audiences
  • Predictive segments

5. Activation:

  • Real-time personalization
  • Email/SMS marketing
  • Advertising platforms
  • Customer service systems
  • Analytics/BI tools

Key CDP Concepts

Identity Resolution

Matching Keys:

  • Email address (primary)
  • Phone number (hashed)
  • Customer ID / Loyalty ID
  • Device IDs (IDFA, GAID, cookie)
  • Address + Name combination

Identity Graph:

            Cookie_A
               ↓
Email_1 ← Customer_123 → Device_B
               ↓
          Phone_555

Confidence Levels:

  • High: Email match, customer ID
  • Medium: Phone + first name
  • Low: Cookie + behavior patterns

Customer Profile

Profile Attributes:

## John Smith - Customer Profile

### Identity
- Customer ID: CUST-123456
- Email: john.smith@email.com
- Phone: +1-555-123-4567
- Device IDs: [device_a, device_b]

### Demographics
- Age: 35
- Gender: Male
- Location: San Francisco, CA
- Household Income: $100-150K

### Behavioral
- First Purchase: 2024-01-15
- Last Purchase: 2026-01-10
- Total Orders: 12
- LTV: $1,245
- Average Order Value: $103.75
- Favorite Category: Electronics
- Last Page Viewed: Laptops
- Email Engagement: High (45% open rate)

### Segments
- VIP Customer
- Tech Enthusiast
- High Email Engager
- Likely to Churn (ML score: 0.65)

### Consent
- Email Marketing: Opted In
- SMS: Opted Out
- Third-party Sharing: Opted Out
- Last Updated: 2026-01-01

Segmentation

Segment Types:

Static Segments: Fixed list, doesn't update automatically

Customers who purchased in Black Friday 2025

Dynamic Segments: Updates as data changes

Customers who purchased in last 30 days
AND total spend > $500
AND category = "Electronics"

Predictive Segments: ML-based predictions

Customers with churn risk > 70%
Customers likely to purchase in next 7 days

Lookalike Segments: Similar to seed audience

Customers similar to "VIP Customers" segment

Data Activation

Activation Channels:

  • Email: Personalized campaigns, triggered emails
  • SMS/Push: Time-sensitive notifications
  • Advertising: Facebook, Google, programmatic
  • Website: Real-time personalization
  • Customer Service: Agent context
  • Analytics: Reporting and insights

Real-time vs Batch:

  • Real-time: Website personalization, triggered emails
  • Batch: Daily email campaigns, weekly reports

Privacy & Consent

Consent Management

Consent Types:

  • Marketing email opt-in/out
  • SMS opt-in/out
  • Third-party data sharing
  • Cookie consent
  • Cross-site tracking

Consent Requirements:

  • GDPR (EU): Explicit opt-in required
  • CCPA (California): Opt-out right
  • LGPD (Brazil): Consent required
  • PDPA (Thailand/Singapore): Consent required

Data Rights

User Rights (GDPR/CCPA):

  • Right to access (what data do you have?)
  • Right to rectification (fix incorrect data)
  • Right to erasure (delete my data)
  • Right to portability (give me my data)
  • Right to opt-out (don't sell my data)

Key Metrics

Data Quality

  • Profile completeness (% fields populated)
  • Identity match rate
  • Data freshness (recency of updates)
  • Duplicate rate

Engagement

  • Segment size and growth
  • Activation success rate
  • Channel reach
  • Personalization lift

Business Impact

  • Revenue influenced by CDP
  • Customer lifetime value improvement
  • Campaign ROI improvement
  • Reduced customer acquisition cost

Common CDP Requirements

Data Collection

REQ-001: System shall collect website events including page views, product views, add-to-cart, and purchases
REQ-002: System shall ingest CRM data via API daily
REQ-003: System shall capture email engagement events in real-time

Identity Resolution

REQ-010: System shall match profiles using email as primary key
REQ-011: System shall stitch anonymous cookie to known profile on login
REQ-012: System shall maintain identity graph with confidence scores

Segmentation

REQ-020: System shall support dynamic segment creation using AND/OR logic
REQ-021: System shall recalculate segment membership daily at minimum
REQ-022: System shall support audience size estimation before creation

Activation

REQ-030: System shall sync audiences to Facebook in near real-time
REQ-031: System shall trigger email via Klaviyo when user enters segment
REQ-032: System shall provide personalization API for website with < 100ms latency

Questions for Stakeholders

  • What data sources need to be connected?
  • How do you identify customers today?
  • What segments do you need?
  • Which marketing channels for activation?
  • What personalization use cases?
  • Privacy/consent requirements?
  • Real-time vs batch needs?

Popular CDP Platforms

  • Segment: Developer-friendly, data infrastructure
  • mParticle: Mobile-focused, SDK expertise
  • Treasure Data: Enterprise, data lake integration
  • Tealium: Tag management heritage
  • Adobe Experience Platform: Adobe ecosystem
  • Salesforce CDP: Salesforce ecosystem

References

  • CDP Institute Resources
  • First-party Data Strategy
  • Privacy Regulations (GDPR, CCPA)

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.29%
按下载量换算1,840

Claude

31.15%
按下载量换算1,721

Cursor

18.61%
按下载量换算1,028

Gemini CLI

8.62%
按下载量换算476

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills