Whatsapp Automation

作者 claude-office-skills9c4c7d5cd281MIT499 个星标收录于 2026年10月8日更新于 2026年10月8日仓库8个月前更新

WhatsApp Business automation - customer support, notifications, chatbots, and broadcast messaging

AI 生成的概览

指导 WhatsApp Business 自动化,涵盖客服聊天机器人、订单通知、预约提醒和群发活动。

功能
该技能为 WhatsApp Business 通信自动化提供参考指导,涵盖 Meta Business API 配置、消息模板类别与审批、聊天机器人对话流程、订单与物流通知工作流、预约提醒以及群发活动管理。它还概述了带升级触发条件的 AI 回复配置,并给出一个 n8n 工作流示例。其产出是配置与模板规范,而非可执行代码。
适用场景
适用于规划或记录 WhatsApp Business 消息自动化时,例如设计客服聊天机器人、订单通知序列、预约提醒或群发活动。适合在 Meta Business Manager 或 n8n 中构建之前需要模板与工作流规范的团队。
运行要求
需要 Meta Business 账户、WhatsApp Business API 访问权限(Phone Number ID 和访问令牌)、webhook 配置以及已审批的消息模板。该技能引用名为 whatsapp-mcp 的 MCP 服务器及其发送和媒体工具,并提及 n8n 和用于聊天机器人回复的 AI 模型。它不附带脚本,仅为说明文档。

WhatsApp Automation

Automate WhatsApp Business communications including customer support, notifications, chatbots, and broadcast messaging. Based on n8n's WhatsApp integration templates.

Overview

This skill covers:

  • WhatsApp Business API setup
  • Automated chatbot flows
  • Order/shipping notifications
  • Customer support automation
  • Broadcast messaging

Setup & Configuration

WhatsApp Business API

yaml
setup_requirements:  1. meta_business_account:      - Create at business.facebook.com      - Verify business        2. whatsapp_business_api:      - Access through Meta Developer Portal      - Create WhatsApp Business App      - Get Phone Number ID      - Get Access Token        3. webhook_configuration:      - Set callback URL      - Verify token      - Subscribe to messages        4. message_templates:      - Create templates in Business Manager      - Wait for approval (24-48h)

Message Templates

yaml
template_categories:  transactional:    - order_confirmation    - shipping_update    - delivery_notification    - appointment_reminder      marketing:    - promotional_offer    - product_launch    - newsletter      utility:    - account_verification    - password_reset    - payment_reminder    template_example:  name: "order_confirmation"  language: "en"  category: "TRANSACTIONAL"  components:    - type: HEADER      format: TEXT      text: "Order Confirmed! 🎉"          - type: BODY      text: |        Hi {{1}},                Your order #{{2}} has been confirmed!                Items: {{3}}        Total: {{4}}                We'll notify you when it ships.            - type: FOOTER      text: "Reply HELP for support"          - type: BUTTONS      buttons:        - type: URL          text: "Track Order"          url: "https://example.com/track/{{1}}"

Chatbot Flows

Customer Support Bot

yaml
chatbot_flow:  name: "Customer Support"    welcome:    trigger: first_message    response: |      👋 Welcome to {Company} Support!            How can I help you today?            1️⃣ Track Order      2️⃣ Return/Exchange      3️⃣ Product Questions      4️⃣ Speak to Agent            Reply with a number to continue.        flows:    track_order:      trigger: "1" OR "track" OR "order"      steps:        - ask: "Please enter your order number:"        - validate: order_number_format        - lookup: order_status        - respond: |            📦 *Order #{order_number}*                        Status: {status}            {if_shipped: Tracking: {tracking_number}}            Estimated delivery: {eta}                        [Track Package]({tracking_url})                return_exchange:      trigger: "2" OR "return" OR "exchange"      steps:        - ask: "Order number for return?"        - check: return_eligibility        - if_eligible:            respond: return_instructions        - if_not:            respond: contact_support                product_questions:      trigger: "3" OR "product" OR "question"      steps:        - ai_response:            model: gpt-4            context: product_faq_knowledge_base                agent_handoff:      trigger: "4" OR "agent" OR "human"      steps:        - check: agent_availability        - if_available:            transfer: to_live_agent            notify: "#support-queue"        - if_unavailable:            respond: "Our team is currently unavailable. Leave a message and we'll get back to you within 24 hours."            create: support_ticket

AI-Powered Responses

yaml
ai_chatbot:  model: gpt-4    system_prompt: |    You are a WhatsApp customer support agent for {Company}.        Guidelines:    - Be helpful and friendly    - Keep responses concise (WhatsApp prefers short messages)    - Use emojis appropriately    - If you can't help, offer to connect with a human agent    - Never make up information about orders or products        You have access to:    - Order lookup    - Product catalog    - FAQ knowledge base      tools:    - lookup_order: by order number    - search_products: by keyword    - check_inventory: by product id    - create_ticket: for complex issues      escalation_triggers:    - sentiment: very_negative    - keywords: ["complaint", "refund", "angry", "sue"]    - loop_detected: same_question_3x

Notification Workflows

Order Notifications

yaml
order_notifications:  order_placed:    trigger: shopify_order_created    template: order_confirmation    variables:      - customer_name      - order_number      - items_summary      - total_amount        payment_received:    trigger: payment_captured    message: |      ✅ Payment received for order #{order_number}!            Amount: {amount}            We're preparing your order now.        shipped:    trigger: fulfillment_created    template: shipping_update    variables:      - tracking_number      - carrier      - estimated_delivery        out_for_delivery:    trigger: tracking_status_change    condition: status == "out_for_delivery"    message: |      🚚 Your order is out for delivery!            Expected today by {eta}.            Track live: {tracking_url}        delivered:    trigger: tracking_status_change    condition: status == "delivered"    message: |      📦 Your order has been delivered!            We hope you love it!             [Leave a Review]({review_url})      [Need Help?]({support_url})

Appointment Reminders

yaml
appointment_reminders:  schedule:    - 24_hours_before    - 1_hour_before      template: appointment_reminder    24h_message: |    📅 *Appointment Reminder*        Hi {name},        You have an appointment tomorrow:        📍 {location}    🕐 {time}    👤 {provider}        Reply:    ✅ CONFIRM to confirm    ❌ CANCEL to cancel    📅 RESCHEDULE to change time      1h_message: |    ⏰ Your appointment is in 1 hour!        {provider} at {location}        See you soon!

Broadcast Messaging

Campaign Management

yaml
broadcast_campaigns:  setup:    - import_contacts: from_crm    - segment_audience: by_criteria    - create_template: get_approval    - schedule_send      targeting_options:    - all_contacts    - by_tag: [vip, new_customer, inactive]    - by_purchase_history    - by_location    - custom_segment      compliance:    - opt_in_required: true    - opt_out_easy: reply_STOP    - frequency_limits: per_contact_per_day    - quiet_hours: respect_timezone      tracking:    - delivered    - read    - replied    - clicked_link    - opted_out

Campaign Templates

yaml
campaign_templates:  promotional:    template: |      🎉 *Special Offer for You!*            Hi {name},            {offer_description}            Use code: {promo_code}      Valid until: {expiry}            [Shop Now]({url})            Reply STOP to unsubscribe        product_launch:    template: |      🚀 *New Arrival Alert*            {name}, check out our latest:            {product_name}      {product_description}            {price}            [View Product]({url})        reengagement:    template: |      👋 *We Miss You, {name}!*            It's been a while since your last visit.            Here's 15% off to welcome you back:            Code: COMEBACK15            [Shop Now]({url})

n8n Integration

Workflow Example

yaml
workflow: "WhatsApp Order Bot"
nodes:  - name: "WhatsApp Trigger"    type: "n8n-nodes-base.whatsAppTrigger"    parameters:      events: ["messages"]        - name: "Parse Message"    type: "n8n-nodes-base.code"    parameters:      code: |        const message = $input.first().json;        return {          from: message.from,          text: message.text.body,          type: message.type        };          - name: "AI Response"    type: "n8n-nodes-base.openAi"    parameters:      model: "gpt-4"      messages: [        { role: "system", content: "..." },        { role: "user", content: "{{ $json.text }}" }      ]        - name: "Send Reply"    type: "n8n-nodes-base.whatsApp"    parameters:      operation: "sendMessage"      phoneNumberId: "{{ $env.WHATSAPP_PHONE_ID }}"      recipientPhoneNumber: "{{ $json.from }}"      message: "{{ $json.response }}"

Output Example

Request: "Create WhatsApp order notification flow"

Output:

markdown
# WhatsApp Order Notifications
## Message Templates (Need Meta Approval)
### 1. Order Confirmation

Hi {{1}}! 🎉

Your order #{{2}} is confirmed!

Items: {{3}} Total: {{4}}

We'll update you when it ships.

Track Order [blocked]


### 2. Shipping Update

📦 Your order is on the way!

Order: #{{1}} Carrier: {{2}} Tracking: {{3}}

Expected delivery: {{4}}

Track Package [blocked]


### 3. Delivered

✅ Delivered!

Your order #{{1}} has arrived.

Love it? Leave a review! Need help? Reply to this message.

Review [blocked]


## n8n Workflow
```yamltrigger: Shopify Order Createdsteps:  1. Get customer phone  2. Format order details  3. Send WhatsApp template  4. Log to CRM

Automation Rules

  • Order placed → Confirmation (immediate)
  • Shipped → Tracking (immediate)
  • Out for delivery → Alert (immediate)
  • Delivered → Review request (1 day later)

---
*WhatsApp Automation Skill - Part of Claude Office Skills*

来源与署名

来源:claude-office-skills/skills位于whatsapp-automation提交9c4c7d5

许可证: MIT

内容归原作者所有。SourceWeft 从公开仓库中收录这些内容。

举报或申请下架