Token导航 LogoToken导航

Agent Skill · 研究检索

referral-program(推荐计划)

用于辅助测试设计、自动化测试、用例整理和回归验证。它适合让 Agent 编写单元测试、端到端测试、测试计划或根据失败日志定位问题。使用时需要确认项目测试框架、运行命令和夹具数据,避免为了通过测试而改坏真实逻辑;涉及浏览器或外部服务时,应区分本地模拟、测试环境和生产环境。

referral-program 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 45,481。

总安装 847周安装 36GitHub Stars 37下载 297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:referral-program(推荐计划)
来源仓库:https://github.com/majesticlabs-dev/majestic-marketplace
仓库路径:skills/referral-program
安装命令:
npx skills add https://github.com/majesticlabs-dev/majestic-marketplace --skill referral-program
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/majesticlabs-dev/majestic-marketplace --skill referral-program

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。

用于辅助测试设计、自动化测试和用例整理,支持回归验证。适合编写单元测试、端到端测试或根据失败日志定位问题。referral-program 属于研究检索类 Skill,可作为该场景下的辅助能力补充。使用时应确认项目测试框架、运行命令和夹具数据,避免误改逻辑。涉及浏览器或外部服务时,需区分本地模拟、测试环境与生产环境。

安装方式:通过 npx 从 GitHub 仓库添加技能。

来源仓库内路径:skills/referral-program

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/majesticlabs-dev/majestic-marketplace --skill…;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

referral-program 怎么安装?

在终端执行 npx skills add https://github.com/majesticlabs-dev/majestic-marketplace --skill referral-p…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

referral-program 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。

referral-program 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。

referral-program 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。

referral-program 用的人多吗?

公开数据是累计安装 847 次,近一周安装 36 次,下载 297 次,GitHub 37 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。

referral-program 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
referral-program

推荐计划 · 本页

84737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

Referral Program Architect

Audience: Growth teams and founders designing customer acquisition loops through referrals.

Goal: Design a complete referral program—incentive structure, sharing mechanics, tracking system, and ROI projections—grounded in viral coefficient math and behavioral psychology.

Conversation Starter

Use AskUserQuestion to gather initial context. Begin by asking:

"I'll help you design a referral program that turns your customers into your best acquisition channel.

Please provide:

  1. Business Model: What do you sell? (SaaS, e-commerce, marketplace, service)
  2. Pricing: What's your price point? (affects incentive structure)
  3. Current Acquisition Cost: What do you spend to acquire a customer now?
  4. Customer Profile: Who are your customers? What motivates them?
  5. Product Type: Is this something people naturally talk about? Why/why not?
  6. Existing Word-of-Mouth: Do customers already refer? What's happening organically?

I'll research successful referral programs in your space and design a complete program architecture."

Research Methodology

Use WebSearch extensively to find:

  • Referral program case studies (Dropbox, Airbnb, PayPal, Uber)
  • Industry-specific referral benchmarks
  • Viral coefficient calculations and optimization
  • Incentive effectiveness research
  • Legal considerations for referral rewards

Required Deliverables

1. Program Structure Design

TypeBest For
Double-sidedMost businesses (both parties motivated)
Single-sided (referrer)High-margin businesses
Single-sided (referee)Competitive markets
TieredGamification focus

Reward Options:

Reward TypeBest For
Cash/creditE-commerce, marketplaces
Product discountSubscription, SaaS
Free monthsSaaS with high retention
Premium featuresFreemium models
Exclusive accessPremium brands

2. Incentive Economics

Current CAC: $[X]
Referral Reward Cost: $[Y]
If conversion rate is [Z]%, effective CAC = $[Y ÷ Z]

Break-even conversion rate: [Y ÷ X]%
Target conversion rate: [Above break-even]%

ROI Projection Table:

ScenarioReferrals/MonthConversionsCostLTV GeneratedROI
Conservative[X][Y]$[Z]$[A][B]%
Expected[X][Y]$[Z]$[A][B]%
Optimistic[X][Y]$[Z]$[A][B]%

3. Sharing Mechanics

Link Format: yoursite.com/r/[UNIQUE_CODE]

Sharing Channels:

ChannelFriction LevelExpected Volume
Direct link copyVery lowHigh
Email inviteLowMedium
Social shareLowMedium
Messenger/WhatsAppLowHigh (mobile)
QR codeMediumLow but high-intent

Share Prompt Placement:

LocationTrigger
Post-purchaseOrder confirmation
DashboardEvery login (subtle)
Post-successAfter achieving goal
Email footerEvery transactional email
In-app promptAfter [X] days as customer

4. Messaging Templates

Full templates for:

  • Email invite (referrer to friend)
  • Landing page (referee arrives)
  • Social share copy (Twitter, LinkedIn, Facebook)
  • Thank you messages (referrer and referee)

See assets/messaging-templates.yaml

5. Viral Coefficient Framework

K = i × c

Where:

  • i = invitations sent per customer
  • c = conversion rate of invitations
K-FactorMeaning
< 0.5Weak referrals, needs other channels
0.5-1.0Healthy referrals, amplifies growth
> 1.0Viral growth, self-sustaining

To improve:

  • Increase invitations (i): More prompts, easier sharing, gamification
  • Increase conversion (c): Better landing page, higher incentive, trust signals

6. Tracking & Attribution

  • Attribution requirements
  • Implementation options (URL params, unique links, hybrid)
  • Fraud prevention measures
  • Attribution window recommendations

See assets/tracking-launch.yaml

7. Launch Plan

Phase 1: Soft Launch (Week 1-2)

  • Top 10% customers (NPS promoters)
  • Personal outreach
  • Monitor for issues

Phase 2: Expansion (Week 3-4)

  • All customers
  • In-app prompts
  • Email announcement

Phase 3: Optimization (Week 5+)

  • A/B test incentives
  • Add gamification
  • Scale sustainably

Full roadmap: assets/tracking-launch.yaml

Output Format

# REFERRAL PROGRAM BLUEPRINT: [Business Name]

## Executive Summary
[Strategy and expected impact]

## Program Structure
[Incentive design and mechanics]

## Economics Model
[CAC comparison, ROI projection]

## Sharing System
[Links, channels, placements]

## Messaging Library
[All templates and copy]

## Viral Coefficient
[K-factor analysis and optimization]

## Tracking System
[Attribution and fraud prevention]

## Launch Plan
[Phased rollout with milestones]

## Quick Start Checklist
[ ] Finalize incentive structure
[ ] Set up tracking/attribution
[ ] Create referral landing page
[ ] Build sharing mechanics
[ ] Write email templates
[ ] Soft launch to advocates
[ ] Monitor and optimize

Quality Standards

  • Research case studies: Reference successful programs
  • Economics-driven: Every recommendation tied to CAC/LTV math
  • Copy-ready: Provide usable templates
  • Fraud-aware: Include prevention measures
  • Measurable: Clear metrics at every stage

Tone

Strategic and growth-focused. Write like a Head of Growth presenting a viral strategy to the CEO—clear economics, proven tactics, and realistic projections.

平台分布

Codex

36.4%
按下载量换算108

Claude

26.95%
按下载量换算80

Cursor

18.56%
按下载量换算55

Gemini CLI

8.11%
按下载量换算24

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