Token导航 LogoToken导航

Agent Skill · 研究检索

referral-program(推荐计划)

referral-program 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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。

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,权限为接触敏感数据;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 56,981。

总安装 442周安装 19GitHub Stars 19下载 155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/finsilabs/awesome-ecommerce-skills
仓库路径:skills/referral-program
安装命令:
npx skills add https://github.com/finsilabs/awesome-ecommerce-skills --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/finsilabs/awesome-ecommerce-skills --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。

referral-program 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。适用于推荐计划相关的资料搜索与内容筛选,可结合来源仓库和原始 README 进一步核验具体用法。

通过 npx skills add 命令从 GitHub 仓库安装,支持主流 AI 宿主环境集成调用。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操作。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

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/finsilabs/awesome-ecommerce-skills --skill ref…;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

    它需要接触敏感数据,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

referral-program 怎么安装?

在终端执行 npx skills add https://github.com/finsilabs/awesome-ecommerce-skills --skill referral-prog…。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?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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 442 次,近一周安装 19 次,下载 155 次,GitHub 19 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。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 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。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 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

44219接触敏感数据未明确有提醒

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

Overview

A referral program turns your existing customers into an acquisition channel by rewarding them for introducing new customers. Referral apps like ReferralCandy and Referral Hero handle unique link generation, double-sided rewards (referrer and referee), fraud detection, and post-purchase email triggers without custom code. Only build a custom referral system if your tiering logic, CRM integration, or fraud rules exceed what these tools support.

When to Use This Skill

  • When building a "give $10, get $10" refer-a-friend program
  • When adding tiered rewards that escalate with the number of successful referrals
  • When fraud from self-referrals or multiple accounts is draining referral reward budget
  • When measuring referral program CAC versus other acquisition channels
  • When integrating referral tracking with post-purchase email flows in Klaviyo

Core Instructions

Step 1: Determine platform and choose the right referral tool

PlatformRecommended ToolWhy
ShopifyReferralCandyPurpose-built for e-commerce; auto-generates referral links per customer, handles double-sided rewards, post-purchase email trigger, and fraud detection
ShopifyReferral HeroMore flexible reward types (cash, gift cards, store credit, custom); deep Klaviyo integration for referral email flows
ShopifySmile.ioCombines loyalty points with referral mechanics in one app — best when you want both
WooCommerceReferralCandy or Referral HeroBoth offer WooCommerce plugins; connect via REST API to track orders and issue rewards
BigCommerceReferralCandyAvailable on the BigCommerce App Marketplace
Custom / HeadlessBuild referral tracking + reward logicRequired when platform integrations or fraud rules don't match your needs

Step 2: Platform-specific setup


Shopify

Option A: ReferralCandy (recommended — full referral suite)

  1. Install ReferralCandy from the Shopify App Store
  2. Connect your Shopify store — ReferralCandy automatically imports existing customers
  3. Configure your reward structure:

- Go to ReferralCandy → Rewards → Set Rewards - Choose reward type for referrer: cash via PayPal, store discount, or custom gift - Choose reward type for referee: discount code applied at checkout - Example: Referrer gets $10 store credit; referee gets $10 off their first order

  1. Configure referral email timing:

- Go to Post-Purchase Email → Settings - Set send delay to 1–3 days after purchase (customer is in the honeymoon phase) - Customize the email template with your brand and the referral link

  1. ReferralCandy automatically:

- Generates a unique referral link per customer - Tracks clicks and attributions with a 30-day cookie window - Detects self-referrals and blocks reward issuance - Sends the referee a welcome email with their discount code when they sign up via the link

  1. View performance in ReferralCandy → Analytics:

- Referrals sent, clicks, signups, and purchases - Revenue attributed to referrals - Top referrers (for VIP outreach)

Option B: Referral Hero (Klaviyo-first)

  1. Install Referral Hero from the App Store
  2. Configure in Referral Hero → Campaigns → Create Campaign
  3. Referral Hero integrates with Klaviyo — when a referral converts, it triggers a Klaviyo flow for reward notification emails
  4. Reward types include store credit (issued via Shopify discount codes), cash (PayPal), or gift cards

Adding a referral entry point in post-purchase emails (Klaviyo):

  1. In Klaviyo, go to the Post-Purchase flow
  2. Add an email at the 3-day step with the referral link
  3. ReferralCandy and Referral Hero both provide a Klaviyo integration that injects {{customer.referral_link}} as a merge tag

WooCommerce

ReferralCandy for WooCommerce:

  1. Install the ReferralCandy plugin from WordPress.org or the ReferralCandy integrations page
  2. Connect your WooCommerce store with your ReferralCandy API credentials
  3. ReferralCandy tracks WooCommerce orders and attributes purchases to referrers automatically
  4. Configure rewards and email timing in the ReferralCandy dashboard (same as Shopify above)

Referral Hero for WooCommerce:

  1. Install the Referral Hero WooCommerce plugin
  2. Configure a campaign and set reward type to WooCommerce coupon code
  3. Referral Hero issues a unique coupon code to each referee upon signup and notifies the referrer when they convert

Manual referral program with AutomateWoo:

  • If you prefer full control, use AutomateWoo to trigger referral emails and issue discount codes
  • AutomateWoo has a built-in Referral workflow template under AutomateWoo → Workflows → Referrals

BigCommerce

ReferralCandy for BigCommerce:

  1. Go to Apps → Search "ReferralCandy" and install
  2. Configuration is the same as the Shopify version
  3. ReferralCandy tracks BigCommerce orders and issues BigCommerce discount codes as rewards

Custom / Headless

For headless storefronts, build referral tracking with tiered rewards and fraud detection:

// lib/referrals.ts
import { randomBytes } from 'crypto';

// Generate a unique, human-friendly referral code per customer
export async function getOrCreateReferralCode(customerId: string): Promise<string> {
  const existing = await db.referralCodes.findFirst({ where: { customerId } });
  if (existing) return existing.code;

  const customer = await db.customers.findUnique({ where: { id: customerId } });
  const prefix = (customer!.firstName ?? 'USER').slice(0, 4).toUpperCase();
  const suffix = randomBytes(3).toString('hex').toUpperCase();
  const code = `${prefix}${suffix}`; // e.g., JANE3A9F

  await db.referralCodes.create({ data: { customerId, code } });
  return code;
}

// Middleware: capture referral code from ?ref= URL param into a 30-day cookie
export function referralTrackingMiddleware(req: Request, res: Response, next: NextFunction) {
  const code = req.query.ref as string;
  if (code && !req.cookies.ref_code) {
    res.cookie('ref_code', code, { maxAge: 30 * 86400000, httpOnly: true, secure: true, sameSite: 'lax' });
  }
  next();
}

// Tiered rewards: reward amount increases as referrer accumulates successful referrals
const REFERRAL_TIERS = [
  { minReferrals: 0,  referrerRewardCents: 1000, refereeRewardCents: 1000 }, // $10/$10
  { minReferrals: 5,  referrerRewardCents: 1500, refereeRewardCents: 1000 }, // $15/$10 after 5
  { minReferrals: 10, referrerRewardCents: 2500, refereeRewardCents: 1500 }, // $25/$15 after 10
];

function getReferralReward(successfulReferrals: number) {
  return [...REFERRAL_TIERS].reverse().find(t => successfulReferrals >= t.minReferrals)!;
}

// On new account creation: attribute referral, run fraud checks
export async function onCustomerCreated(customerId: string, refCode: string | undefined) {
  if (!refCode) return;

  const referralCode = await db.referralCodes.findFirst({ where: { code: refCode } });
  if (!referralCode || referralCode.customerId === customerId) return; // block self-referral

  // Fraud check: same shipping address as referrer
  const [referee, referrer] = await Promise.all([
    db.customers.findUnique({ where: { id: customerId }, include: { addresses: true } }),
    db.customers.findUnique({ where: { id: referralCode.customerId }, include: { addresses: true } }),
  ]);

  const refereeAddr = referee?.addresses[0]?.addressHash;
  const referrerAddr = referrer?.addresses[0]?.addressHash;
  if (refereeAddr && refereeAddr === referrerAddr) return; // same address = block

  await db.referrals.create({ data: { referrerId: referralCode.customerId, refereeId: customerId, code: refCode, status: 'pending' } });
}

// On referee's first purchase: issue rewards to both parties
export async function onRefereeFirstPurchase(refereeId: string, orderId: string) {
  const referral = await db.referrals.findFirst({ where: { refereeId, status: 'pending' } });
  if (!referral) return;

  const order = await db.orders.findUnique({ where: { id: orderId } });
  if (!order || order.subtotalCents < 2500) return; // require minimum $25 order

  const referrerStats = await db.referralCodes.findFirst({ where: { customerId: referral.referrerId } });
  const tier = getReferralReward(referrerStats?.totalReferrals ?? 0);

  await Promise.all([
    issueStoreCredit(referral.referrerId, tier.referrerRewardCents, `Referral reward`),
    issueStoreCredit(refereeId, tier.refereeRewardCents, 'Welcome reward — referred by a friend'),
  ]);

  await db.referrals.update({ where: { id: referral.id }, data: { status: 'rewarded', orderId } });
  await db.referralCodes.update({ where: { customerId: referral.referrerId }, data: { totalReferrals: { increment: 1 } } });
}

Referral program analytics SQL:

-- Monthly referral program performance
SELECT
  DATE_TRUNC('month', r.created_at) AS month,
  COUNT(*) AS total_referrals,
  COUNT(CASE WHEN r.status = 'rewarded' THEN 1 END) AS successful_referrals,
  ROUND(100.0 * COUNT(CASE WHEN r.status = 'rewarded' THEN 1 END) / NULLIF(COUNT(*), 0), 1) AS conversion_rate_pct,
  SUM(CASE WHEN r.status = 'rewarded' THEN rc.total_rewards_cents END) / 100.0 AS rewards_paid_usd
FROM referrals r
JOIN referral_codes rc ON r.referrer_id = rc.customer_id
GROUP BY 1
ORDER BY 1 DESC;

Step 3: Promote the referral program

The referral link needs to be visible — customers won't find it if it's buried:

  1. Post-purchase confirmation page — add a referral CTA to the order confirmation page: "Love your order? Share with a friend and you both get $10"
  2. Post-purchase email — include the referral link in the 3-day post-purchase email (ReferralCandy and Klaviyo both support this automatically)
  3. Customer account page — add a "Refer a Friend" section showing their link and referral history
  4. Transactional emails — include a subtle referral CTA in shipping confirmation emails

Step 4: Measure referral program ROI

MetricHow to Measure
Referral conversion rateSuccessful referrals / total referral clicks
Referral CACTotal rewards paid / successful referrals
Revenue attributed to referralsTag referred orders and compare total revenue
Referred customer CLV vs. non-referredCohort comparison at 90 days and 12 months

Compare referral CAC against your paid acquisition channels. Referral programs typically produce 20–30% lower CAC than paid channels when optimized.

Best Practices

  • Use ReferralCandy before building from scratch — it handles link generation, attribution, fraud detection, and reward issuance in minutes
  • Require a minimum first-order value to qualify for the reward — prevents reward farming with $1 test orders
  • Use store credit, not one-time discount codes — store credit requires the customer to return and creates a second purchase occasion
  • Apply a 30-day attribution window — shorter windows miss customers who take time to convert
  • Send a reward notification email immediately when the referee purchases — the referrer often doesn't know the reward was granted; this also prompts further referrals
  • Block referrals to the same shipping address — this is the clearest fraud signal for household self-referrals

Common Pitfalls

ProblemSolution
Self-referral fraud via multiple email accountsBlock same shipping address referrals (strongest signal); ReferralCandy detects this automatically
Referral link shared on coupon forumsMonitor referred cohort CLV; if it's significantly lower than organic, apply a higher minimum order value
Store credit issued before 30-day return windowDelay reward issuance by 30 days after the referee's order to account for returns
Referral cookie overwritten by later ad clickStore the first referral click separately; ReferralCandy uses first-click attribution for referrals
Low referral email open ratesMove the referral ask from the receipt email to a dedicated 3-day post-purchase email when customer satisfaction is highest

Related Skills

  • @affiliate-program
  • @customer-lifetime-value
  • @customer-segmentation
  • @email-marketing-automation

平台分布

Codex

37.4%
按下载量换算58

Claude

27.49%
按下载量换算43

Cursor

16.72%
按下载量换算26

Gemini CLI

9.18%
按下载量换算14

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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