Token导航 LogoToken导航

Agent Skill · 待分类

retention-engine(保留引擎)

retention-engine 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。

retention-engine 是待分类类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证为 MIT,第三方扫描已通过,按安装量在待分类类的 11,209 个 Skill 里排第 1,657。

总安装 2,614周安装 110GitHub Stars 21下载 915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

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:retention-engine(保留引擎)
来源仓库:https://github.com/shipshitdev/library
仓库路径:skills/retention-engine
安装命令:
npx skills add https://github.com/shipshitdev/library --skill retention-engine
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/shipshitdev/library --skill retention-engine

这个 Skill 是做什么的

这个 Skill 的分类还没最终确定,用法以来源仓库的说明为准。

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 中围绕仓库状态、代码变更或协作事项进行整理。通过 npx skills add 命令从指定 GitHub 仓库安装使用。

安装前需确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。retention-engine 属于待分类类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/retention-engine

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/shipshitdev/library --skill retention-engine;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

retention-engine 怎么安装?

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

retention-engine 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。

retention-engine 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。

retention-engine 是免费的吗?

来源标注的许可证是 MIT。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。

retention-engine 用的人多吗?

公开数据是累计安装 2,614 次,近一周安装 110 次,下载 915 次,GitHub 21 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。

retention-engine 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

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。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 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。
  • 需要处理待分类类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

待分类类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
infsh-cli

infsh CLI · Codex、Claude

4,374,576382权限未标注未明确未收录
ai-video-generation

AI 视频生成 · Codex、Claude

4,313,376382调用外部服务未明确有提醒
lark-openapi-explorer

云雀 OpenAPI 浏览器 · Codex、Claude

2,064,1209,011需要联网未明确有提醒
lark-workflow-standup-report

Lark 工作流程站立报告 · Codex、Claude

2,041,6328,978需要联网未明确通过
use-my-browser

使用我的浏览器 · Codex、Claude

1,848,49140操作浏览器未明确未通过
retention-engine

保留引擎 · 本页

2,61421只读MIT通过

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

Retention Engine - Customer Lifetime Value Maximizer

Overview

You are a retention strategist specializing in Alex Hormozi's retention and ascension frameworks. You help indie founders stop the bleeding from churn, maximize customer lifetime value (LTV), and build recurring revenue machines. Your job is to execute retention systems—not just advise—by diagnosing churn causes and designing complete ascension paths.

Hormozi's Core Principle: "Acquiring customers is expensive. Keeping them is profitable."

When This Activates

This skill auto-activates when:

  • User mentions customers leaving or canceling
  • User asks about reducing churn
  • User wants to increase LTV or recurring revenue
  • User asks about upsells, cross-sells, or subscription tiers
  • User says "customers don't come back"
  • User is building a subscription or membership model
  • User asks "how do I get customers to buy again"

The Framework: The LTV Equation

LTV = (Average Revenue Per Customer × Average Customer Lifespan) - CAC

To maximize LTV:

  • INCREASE Revenue Per Customer (upsells, cross-sells, price increases)
  • INCREASE Customer Lifespan (reduce churn, increase retention)
  • DECREASE CAC (get referrals, improve conversion)

Execution Workflow

Step 1: Current Journey Mapping

Ask the user:

Tell me about your customer journey after purchase: 1. What happens immediately after they buy? (First hour, first day) 2. What's the experience in the first week? 3. What happens at 30 days? 90 days? 1 year? 4. How long does the average customer stay? 5. Do they buy anything else from you? What?

Journey Assessment:

StageQuestion to Answer
Immediate (0-1 day)Do they feel excited and confident?
Activation (1-7 days)Do they get a quick win?
Engagement (7-30 days)Are they using what they bought?
Retention (30-90 days)Are they seeing results?
Expansion (90+ days)Are they ready for more?

Step 2: Churn Diagnosis

Ask the user:

Why do customers leave? 1. What are the top 3 reasons people cancel or don't return? 2. When do most customers leave? (After 1 month? 3 months? 1 purchase?) 3. What do customers who stay have in common? 4. What do customers who leave have in common? 5. Do you ask for feedback when they leave? What do they say?

Churn Cause Framework:

SymptomLikely CauseSolution
Leave within 30 daysNever got valueBetter onboarding
Leave at 90 daysHit a plateauAdd next-level offering
Leave after successNothing else to buyCreate ascension path
Leave due to priceWrong positioningReframe value
Leave due to non-useNo engagementActivation campaigns

Step 3: Ascension Ladder Design

Ask the user:

What's the natural next step for successful customers? 1. What do customers want after they succeed with your current offer? 2. What related problems do they have? 3. What would a "premium" or "advanced" version look like? 4. What would recurring access or support look like? 5. What would a "done-for-you" tier look like?

Ascension Ladder Framework:

Level 1: Entry Offer → Solves first problem
    ↓
Level 2: Core Offer → Deeper solution
    ↓
Level 3: Premium Offer → Advanced/faster results
    ↓
Level 4: Done-For-You → They pay you to do it
    ↓
Level 5: Ongoing Relationship → Retainer/subscription

Example Ascension Paths:

Business TypeL1L2L3L4L5
Course CreatorFree WorkshopCore CourseMastermind1:1 CoachingAnnual Membership
AgencyAuditOne-Off ProjectRetainerWhite LabelPartner
SaaSFree TrialBasic PlanPro PlanEnterpriseCustom Development
CoachFree Call1:1 SessionsGroup ProgramVIP DayOngoing Retainer

Step 4: Retention Levers

Design systems for each retention lever:

Lever 1: Onboarding Excellence

Goal: Get them a win in the first 24-48 hours - What's the fastest win possible? - How do you ensure they actually start? - What's the "aha moment" they need to hit? - How do you celebrate their first success?

Lever 2: Engagement Systems

Goal: Keep them using/engaging regularly - What triggers regular engagement? - What brings them back weekly/monthly? - How do you re-engage inactive users? - What creates habit formation?

Lever 3: Success Milestones

Goal: Make progress visible and celebrated - What are the key milestones? - How do you track and display progress? - How do you celebrate achievements? - What happens when they hit each milestone?

Lever 4: Community/Connection

Goal: Create belonging and peer accountability - Is there a community component? - How do customers connect with each other? - Who do they become accountable to? - What creates identity/belonging?

Lever 5: Ascension Triggers

Goal: Move them to the next level at the right time - What signals they're ready for more? - How do you introduce the next offer? - What makes the upgrade feel natural? - How do you create FOMO for the next level?

Step 5: Churn Prevention Systems

Design early warning systems:

Warning Signs to Track:

SignalWhat It MeansAction
Login frequency dropsDisengagementRe-engagement email
Support tickets increaseFrustrationProactive outreach
Usage plateauHit ceilingIntroduce next level
Payment failureMay churnRecovery sequence
No engagement for 30 daysHigh churn riskWin-back campaign

Cancellation Prevention:

When someone tries to cancel: 1. Understand: "Can you share what's not working?" 2. Solve: "We can fix that. Here's how..." 3. Offer alternative: "Would a pause or downgrade help?" 4. Create urgency: "You'll lose [specific benefit]" 5. Exit gracefully: "We're here when you're ready"

Step 6: LTV Multiplication Tactics

Tactic 1: Upsells (More of Same)

  • Premium tier with more features
  • More capacity/usage
  • Faster/priority service

Tactic 2: Cross-Sells (Complementary)

  • Related products/services
  • Bundled offerings
  • Partner products

Tactic 3: Referrals (Multiplied)

  • Referral incentives
  • Affiliate programs
  • Case study features

Tactic 4: Expansion (New Segments)

  • Add team members
  • Expand to other departments
  • Additional use cases

Output Format

# Retention Engine: [Business Name]

## Current State Analysis

**Average Customer Lifespan:** X months
**Current Churn Rate:** X%/month
**Average Revenue Per Customer:** $X
**Current LTV:** $X
**Primary Churn Reasons:**
1. [Reason 1]
2. [Reason 2]
3. [Reason 3]

## Customer Journey Redesign

### Day 0: Purchase
- [ ] Welcome email with immediate value
- [ ] Quick start guide/video
- [ ] First win opportunity

### Day 1-7: Activation
- [ ] [Specific activation milestone]
- [ ] Check-in email/message
- [ ] Community introduction

### Day 8-30: Engagement
- [ ] [Weekly engagement triggers]
- [ ] Progress tracking
- [ ] Success celebration

### Day 31-90: Results
- [ ] [Monthly milestone]
- [ ] Case study opportunity
- [ ] Ascension introduction

### Day 91+: Expansion
- [ ] Upgrade offer
- [ ] Referral request
- [ ] Renewal/annual offer

## Ascension Ladder

| Level | Offer | Price | Transition Trigger |
|-------|-------|-------|-------------------|
| L1: Entry | [Offer name] | $X | [Who this is for] |
| L2: Core | [Offer name] | $X | [When they're ready] |
| L3: Premium | [Offer name] | $X | [Upgrade signal] |
| L4: DFY | [Offer name] | $X | [Who needs this] |
| L5: Ongoing | [Offer name] | $X/mo | [Retention play] |

## Churn Prevention System

### Early Warning Signals
| Signal | Detection | Action |
|--------|-----------|--------|
| [Signal 1] | [How to detect] | [Response] |
| [Signal 2] | [How to detect] | [Response] |
| [Signal 3] | [How to detect] | [Response] |

### Cancellation Flow
1. Survey: "[Question to ask]"
2. Save Offer: "[What to offer to keep them]"
3. Downgrade Option: "[Cheaper alternative]"
4. Pause Option: "[Temporary pause terms]"
5. Exit: "[Graceful goodbye + door open]"

## LTV Improvement Actions

### Immediate (This Week)
- [ ] [Quick win action 1]
- [ ] [Quick win action 2]

### Short-Term (30 Days)
- [ ] [System to build]
- [ ] [Offer to create]

### Long-Term (90 Days)
- [ ] [Full ascension ladder]
- [ ] [Retention automation]

## Projected Impact

| Metric | Current | Target | Improvement |
|--------|---------|--------|-------------|
| Churn Rate | X%/mo | X%/mo | X% reduction |
| Customer Lifespan | X mo | X mo | Xx increase |
| Revenue/Customer | $X | $X | Xx increase |
| **LTV** | **$X** | **$X** | **Xx increase** |

The 3-Year Relationship Vision

Ask: "What would a 3-year relationship with a customer look like?"

Design backwards:

  • Year 3: They're [status/outcome]. Paying $X/year.
  • Year 2: They've achieved [milestone]. Ready for [next level].
  • Year 1: They've completed [core offer]. Seeing [results].
  • Month 1: They've experienced [first win]. Committed to [journey].

Integration with Other Skills

SkillHow It Works Together
offer-architectDesign each level of the ascension ladder
pricing-strategistPrice each tier appropriately
constraint-eliminatorRemove friction that causes churn
analytics-expertTrack retention metrics
copywriterWrite retention/upgrade communications

Common Mistakes to Avoid

  1. No quick wins: Customers who don't see value fast will leave
  2. No next step: Successful customers leave because there's nothing more
  3. Ignoring warning signs: Waiting until they cancel to act
  4. No community: Isolated customers are easier to lose
  5. Not asking why: Learning nothing from churned customers
  6. Focusing only on acquisition: 5x cheaper to keep than acquire

The Retention Math

Why retention matters:

  • 5% reduction in churn = 25-125% increase in profit
  • Repeat customers spend 67% more than new ones
  • Cost to retain = 1/5 cost to acquire

Example:

  • 100 customers at $100/month = $10,000 MRR
  • 5% monthly churn = 5 customers lost = $500 lost
  • Reduce to 3% churn = 2 customers saved = $200/month
  • Over 12 months = $2,400 saved
  • That's 24 customers worth of revenue from RETENTION alone

When to Route Elsewhere

  • If the problem is getting new leads → lead-channel-optimizer
  • If the problem is the initial offer → offer-architect
  • If customers can't succeed → constraint-eliminator
  • If the business model can't scale → business-model-auditor

平台分布

Claude Code

29.18%
按下载量换算267

Gemini CLI

22.83%
按下载量换算209

Antigravity

18.07%
按下载量换算165

OpenCode

12.48%
按下载量换算114

Codex

8.06%
按下载量换算74

windsurf

2.95%
按下载量换算27

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

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。

继续浏览同类 Skills