\r \r

Insurance Agent Intelligent Trainer / 保险代理人智能陪练系统\r

\r

English: AI-powered insurance agent coaching system — parses product documents, generates\r personalized question banks, assesses agent competency levels, schedules daily training based on\r client visits, and runs interactive role-play drills. Benchmarked against AIA, Ping An, and\r Alibaba Cloud insurance training systems.\r \r 中文: 保险代理人智能陪练系统——解析产品文档、自动生成问题库、评估代理人能力等级、\r 结合当日客户拜访行程安排个性化训练、进行情景对练。对标友邦保险、平安保险、阿里云智能陪练水平。\r \r ---\r \r

Trigger Keywords / 触发关键词\r

\r Immediately activate when user mentions:\r \r

• 保险陪练 / 产品陪练 / 智能陪练 / 代理人训练\r
• 代理人培训 / 新人培训 / 保险话术训练\r
• 产品演练 / 客户异议处理 / 保险销售训练\r
• agent training / insurance coaching / product drill / sales training\r
• skill assessment / competency evaluation / agent profiling\r
• daily training plan / training schedule / personalized coaching\r \r ---\r \r

Core System Architecture / 核心系统架构\r

\r

┌─────────────────────────────────────────────────────────────────┐\r
│                   Insurance Agent Intelligent Trainer            │\r
├─────────────────────────────────────────────────────────────────┤\r
│  ┌──────────────┐  ┌──────────────┐  ┌──────────────────────┐  │\r
│  │ Product Doc  │  │ Agent Profile│  │ Daily Schedule/Routes│ │\r
│  │ Parser       │  │ Engine       │  │ Integration          │ │\r
│  │ (PDF/Word/   │  │ (Skill Level │  │ (Today's Visits &    │ │\r
│  │  Images)     │  │  Assessment) │  │  Client Profiles)    │ │\r
│  └──────┬───────┘  └──────┬───────┘  └──────────┬───────────┘  │\r
│         │                  │                      │              │\r
│         ▼                  ▼                      ▼              │\r
│  ┌──────────────────────────────────────────────────────────┐    │\r
│  │            Question Bank Generation Engine                │    │\r
│  │  Product Knowledge │ Objection Handling │ Case Analysis   │    │\r
│  │  [5 difficulty tiers × 3 categories = 15 question types] │    │\r
│  └──────────────────────────┬───────────────────────────────┘    │\r
│                             │                                     │\r
│                             ▼                                     │\r
│  ┌──────────────────────────────────────────────────────────┐    │\r
│  │            Personalized Training Scheduler               │    │\r
│  │  [Skill Level + Schedule + Product Priority = Daily Plan]│    │\r
│  └──────────────────────────┬───────────────────────────────┘    │\r
│                             │                                     │\r
│                             ▼                                     │\r
│  ┌──────────────────────────────────────────────────────────┐    │\r
│  │            Interactive Training Engine                   │    │\r
│  │  Role-play │ Real-time Feedback │ Progress Tracking      │    │\r
│  └──────────────────────────────────────────────────────────┘    │\r
└─────────────────────────────────────────────────────────────────┘\r
```\r
\r
---\r
\r
## Core Capabilities / 核心能力\r
\r
### 1. Product Document Parser / 产品文档解析引擎\r
\r
**Supported formats:** PDF, Word (.docx), scanned images (with OCR), plain text\r
\r
**Parsing pipeline:**\r
\r
```\r
Document Upload\r
      │\r
      ▼\r
[Format Detection] → PDF / Word / Image / Text\r
      │\r
      ▼\r
[Text Extraction] → Raw text content\r
      │\r
      ▼\r
[Structure Analysis]\r
  ├─ Product name, type, target customers\r
  ├─ Coverage scope (death, medical, annuity, critical illness, etc.)\r
  ├─ Premium levels & payment periods\r
  ├─ Policy terms & exclusions\r
  ├─ Sales pitch key points\r
  ├─ Competitive advantages vs. similar products\r
  └─ Compliance notes & regulatory requirements\r
      │\r
      ▼\r
[Structured Product Profile] → Ready for question generation\r
```\r
\r
**Output: Structured Product Profile JSON**\r
\r
```json\r
{\r
  "product_name": "XX福享人生终身寿险(万能型)",\r
  "product_type": "whole-life insurance with universal account",\r
  "insurer": "国联人寿",\r
  "target_customers": ["30-50岁中高收入人群", "有财富传承需求"],\r
  "coverage": {\r
    "death_benefit": "100%-160%账户价值",\r
    "annuity_option": "60岁起可转换为年金",\r
    "waiver": "可选投保人保费豁免"\r
  },\r
  "premium": {\r
    "min_annual": 12000,\r
    "payment_periods": ["3年", "5年", "10年", "20年"],\r
    "min_coverage_years": "终身"\r
  },\r
  "key_selling_points": [\r
    "复利增值，万能账户历史结算利率4.5%-5.2%",\r
    "灵活追加，额外资金可随时进入万能账户",\r
    "身故保障与财富传承双重功能"\r
  ],\r
  "competitive_edges": ["结算利率优于同类竞品", "追加无上限"],\r
  "exclusions": ["投保人对被保险人的故意伤害", "2年内自杀(无民事行为能力人除外)"],\r
  "compliance_notes": ["需双录(录音录像)", "犹豫期15天", "等待期90天"],\r
  "difficulty_tags": ["新人友好", "需强化健康告知", "财务规划综合能力"]\r
}\r
```\r
\r
---\r
\r
### 2. Agent Profile & Skill Assessment / 代理人画像与能力评估\r
\r
**Three skill tiers:**\r
\r
| Tier | Level | Description | Training Focus |\r
|------|-------|-------------|----------------|\r
| 🌱 **L1 - 入门级** | Beginner | \x3C 1 year experience, struggles with product details and objection handling | Foundation: product knowledge, basic sales scripts, simple objection responses |\r
| ⚡ **L2 - 进阶级** | Intermediate | 1-3 years, solid product knowledge but inconsistent closing rate | Application: complex scenarios, multi-product combination, competitive replacement, high-net-worth clients |\r
| 🎯 **L3 - 专家级** | Advanced | 3+ years, high performance, needs strategy for complex cases | Mastery: enterprise/group clients, tax planning, estate planning, competitive stealing, mentoring skills |\r
\r
**Profile structure:**\r
\r
```json\r
{\r
  "agent_id": "AG20240001",\r
  "name": "张明",\r
  "level": "L2",\r
  "level_label": "进阶级",\r
  "tenure_years": 2.5,\r
  "certifications": ["保险代理人资格证", "健康险销售资质"],\r
  "performance": {\r
    "monthly_premium_target": 50000,\r
    "monthly_premium_actual": 42000,\r
    "closing_rate": 0.32,\r
    "avg_policy_size": 18500,\r
    "new_customer_rate": 0.45\r
  },\r
  "product_mastery": {\r
    "term_life": 0.85,\r
    "whole_life": 0.72,\r
    "critical_illness": 0.58,\r
    "medical_insurance": 0.80,\r
    "annuity": 0.45,\r
    "investment_linked": 0.38\r
  },\r
  "weak_points": [\r
    "健康险异议处理不够熟练",\r
    "不了解高端客户的税务筹划需求",\r
    "组合产品销售话术单一"\r
  ],\r
  "strong_points": [\r
    "老客户维护能力强",\r
    "缘故市场开拓优秀"\r
  ],\r
  "daily_schedule": [\r
    {"time": "09:00-10:00", "activity": "晨会", "location": "营业部"},\r
    {"time": "10:30-12:00", "activity": "拜访客户A（国企中层，有养老需求）", "location": "客户公司"},\r
    {"time": "14:00-15:30", "activity": "拜访客户B（私企业主，健康险需求）", "location": "客户公司"},\r
    {"time": "16:00-17:30", "activity": "缘故客户C（教育金规划）", "location": "咖啡厅"}\r
  ]\r
}\r
```\r
\r
---\r
\r
### 3. Question Bank Generation / 问题库自动生成\r
\r
**Generated from product profile + agent level + training objectives**\r
\r
#### Question Types (15 categories across 3 dimensions)\r
\r
**By Category:**\r
\r
| Category | Description | Example |\r
|----------|-------------|---------|\r
| **产品知识** | Product features, terms, coverage | "XX福的等待期是多久？" |\r
| **客户画像** | Target customer identification | "什么样的客户适合购买这款产品？" |\r
| **异议处理** | Objection handling scripts | "客户说'我已经有社保了，不需要商业保险'，如何回应？" |\r
| **案例分析** | Real case discussion | "40岁国企中层，年薪50万，如何用这款产品做养老规划？" |\r
| **合规话术** | Compliance-approved scripts | "如何向客户解释犹豫期和退保损失？" |\r
| **竞品对比** | vs. competitors | "相比平安福，这款产品的核心优势是什么？" |\r
| **促成话术** | Closing techniques | "客户表现出购买意向，如何自然促成？" |\r
| **交叉销售** | Multi-product combination | "如何将主险与医疗险组合销售？" |\r
\r
**By Difficulty (5 tiers):**\r
\r
| Level | Target Audience | Question Complexity |\r
|--------|----------------|---------------------|\r
| ⭐ 基础 | L1新人 | 单一产品，单一问题，直接答案 |\r
| ⭐⭐ 入门 | L1-L2 | 单一产品，1-2个知识点，需要解释 |\r
| ⭐⭐⭐ 进阶 | L2 | 单一产品，3-5个知识点，需组合分析 |\r
| ⭐⭐⭐⭐ 高阶 | L2-L3 | 多产品组合，竞争替换，高净值客户 |\r
| ⭐⭐⭐⭐⭐ 专家 | L3 | 综合方案，税务筹划，财富传承 |\r
\r
#### Question Bank Generation Prompt:\r
\r
```\r
Based on the product profile provided, generate a question bank with:\r
\r
1. For each difficulty tier (基础/入门/进阶/高阶/专家):\r
   - 5 multiple choice questions (产品知识)\r
   - 3 case analysis questions\r
   - 3 objection handling scenarios\r
   - 2 competitive comparison questions\r
   - 1 closing technique exercise\r
\r
2. Total: 65+ questions per product\r
\r
3. For each question, provide:\r
   - Question text\r
   - Difficulty level (1-5)\r
   - Category (产品知识/异议处理/案例分析/竞品对比/促成话术)\r
   - Ideal answer / model response\r
   - Evaluation criteria (excellent/good/needs-improvement)\r
   - Coaching tips for the trainer\r
```\r
\r
---\r
\r
### 4. Personalized Training Scheduler / 个性化训练调度引擎\r
\r
**Input factors:**\r
\r
```\r
Agent Profile (Level + Weak Points)\r
         +\r
Today's Client Schedule (Who → What need → What product)\r
         +\r
Product Priority Matrix\r
         =\r
Personalized Daily Training Plan\r
```\r
\r
**Scheduling Algorithm:**\r
\r
```python\r
def generate_daily_training_plan(agent_profile, daily_schedule, products):\r
    """\r
    Generate personalized training plan for the day.\r
    """\r
    # Step 1: Identify today's client visit products\r
    today_products = extract_products_from_schedule(daily_schedule)\r
    \r
    # Step 2: Get agent's weakness areas for these products\r
    weakness_map = get_weakness_for_products(\r
        agent_profile.weak_points, \r
        today_products\r
    )\r
    \r
    # Step 3: Calculate training time available\r
    available_minutes = calculate_available_training_time(daily_schedule)\r
    \r
    # Step 4: Prioritize by impact × weakness × product value\r
    training_queue = prioritize_training(\r
        weakness_map,\r
        today_products,\r
        agent_profile.level,\r
        time_constraint=available_minutes\r
    )\r
    \r
    # Step 5: Generate session plan\r
    sessions = split_into_sessions(training_queue, available_minutes)\r
    \r
    return {\r
        "date": today,\r
        "agent": agent_profile.name,\r
        "total_minutes": available_minutes,\r
        "sessions": sessions,\r
        "focus_products": today_products,\r
        "key_objectives": get_key_objectives(training_queue)\r
    }\r
```\r
\r
**Example Daily Training Plan:**\r
\r
```json\r
{\r
  "date": "2026-05-05",\r
  "agent": "张明",\r
  "level": "L2",\r
  "total_minutes": 90,\r
  "sessions": [\r
    {\r
      "time": "08:00-08:20",\r
      "duration": 20,\r
      "type": "晨间快练",\r
      "mode": "快问快答",\r
      "focus": "年金险产品知识（高频问题5题）",\r
      "product": "福享人生终身寿险",\r
      "objective": "巩固年金转换权的计算逻辑"\r
    },\r
    {\r
      "time": "12:30-13:00",\r
      "duration": 30,\r
      "type": "午间强化",\r
      "mode": "情景对练",\r
      "focus": "健康险异议处理",\r
      "scenario": "客户："我有社保，不需要商业医疗险"",\r
      "product": "康健医疗保险",\r
      "level": "⭐⭐⭐ 进阶",\r
      "coaching_tips": "引导客户认识到社保报销比例上限，用自费药比例对比引发需求"\r
    },\r
    {\r
      "time": "17:30-18:30",\r
      "duration": 40,\r
      "type": "晚间复盘",\r
      "mode": "案例分析 + 角色扮演",\r
      "focus": "私企业主综合保障方案",\r
      "scenario": "45岁私企老板，年收入200万，已有多份保单，如何做加保方案？",\r
      "products": ["终身寿险+万能账户", "高端医疗", "企业财产险"],\r
      "level": "⭐⭐⭐⭐ 高阶",\r
      "model_response_guide": "从家庭资产与企业资产隔离角度切入，引出终身寿险的债务隔离和传承功能"\r
    }\r
  ],\r
  "key_metrics_to_track": [\r
    "异议处理响应时间（目标\x3C30秒）",\r
    "产品知识点正确率（目标>85%）",\r
    "方案组合完整性（3单以上产品覆盖）"\r
  ]\r
}\r
```\r
\r
---\r
\r
### 5. Interactive Training Session / 智能陪练对话引擎\r
\r
**Session modes:**\r
\r
| Mode | Description | Duration | Best For |\r
|------|-------------|----------|----------|\r
| **快问快答** | Rapid-fire Q&A | 5-10 min | Pre-meeting warmup |\r
| **情景对练** | Role-play (client vs. agent) | 15-30 min | Skill practice |\r
| **案例研讨** | Real case analysis | 20-40 min | Advanced agents |\r
| **异议攻关** | Objection busting focus | 10-15 min | Weak point training |\r
| **综合考核** | Full simulation exam | 30-60 min | Level assessment |\r
\r
**Real-time coaching during training:**\r
\r
```\r
Agent Response\r
      │\r
      ▼\r
[Natural Language Understanding] → Extract key claims, tone, strategy\r
      │\r
      ▼\r
[Evaluation Engine]\r
  ├─ Product knowledge accuracy ✓/✗\r
  ├─ Objection handling effectiveness (1-5)\r
  ├─ Compliance adherence ✓/✗\r
  ├─ Closing attempt timing (good/early/late/missing)\r
  ├─ Client empathy signals ✓/✗\r
  └─ Product combination logic ✓/✗\r
      │\r
      ▼\r
[Real-time Coaching Feedback]\r
  ├─ Immediate tip (if struggling): "💡 提示：可以先问客户目前的保障缺口..."\r
  ├─ Completion praise (if excellent): "🌟 完美！您已经很好地识别了客户需求"\r
  └─ Post-question summary: "本轮得分 85/100。建议加强：竞品对比环节"\r
```\r
\r
**Training session flow:**\r
\r
```\r
1. 导入 (5%)     → 介绍训练目标和产品背景\r
2. 暖场 (10%)   → 快问快答热身，激活产品知识\r
3. 主体 (60%)   → 情景对练：客户角色扮演 + 实时点评\r
4. 复盘 (20%)   → AI给出详细反馈：优点/不足/改进建议\r
5. 行动 (5%)   → 下次拜访的具体行动计划\r
```\r
\r
---\r
\r
### 6. Effect Assessment & Progress Tracking / 效果评估与进度追踪\r
\r
**Metrics tracked per session:**\r
\r
| Metric | Definition | Target |\r
|--------|------------|--------|\r
| **产品知识得分** | 知识点正确率 | L1: ≥70%, L2: ≥80%, L3: ≥90% |\r
| **异议处理时效** | 从异议提出到满意回答的时间 | \x3C 30秒 |\r
| **促成成功率** | 能否自然引入促成信号 | ≥ 1次有效尝试 |\r
| **话术合规率** | 合规敏感词使用正确性 | 100% |\r
| **方案完整性** | 保障覆盖广度 | ≥ 3个维度 |\r
\r
**Progress report structure:**\r
\r
```markdown\r
## 📊 代理人张明 训练报告 - 2026-05-05\r
\r
### 综合得分: ⭐⭐⭐⭐ (78/100)\r
\r
| 维度 | 本次得分 | 较上次 | 目标 |\r
|------|---------|--------|------|\r
| 产品知识 | 82/100 | ↑5 | 80+ |\r
| 异议处理 | 71/100 | ↓3 | 75+ |\r
| 促成技巧 | 85/100 | ↑8 | 80+ |\r
| 合规话术 | 95/100 | →0 | 100 |\r
| 方案设计 | 72/100 | ↑12 | 75+ |\r
\r
### 🔥 本次表现亮点\r
1. 养老规划方案逻辑清晰，能结合客户生命周期讲解\r
2. 合规话术使用规范，犹豫期/退保说明完整\r
\r
### ⚠️ 需要加强\r
1. 健康险异议处理：回应"已有社保"时过于被动，应主动算账\r
2. 竞品对比：对中国平安主要产品线不够熟悉\r
\r
### 📅 明日训练重点\r
- 产品：康健医疗保险（健康告知流程）\r
- 场景：竞品替换（平安福 vs. XX福）\r
- 时长：30分钟情景对练 + 10分钟快问快答\r
```\r
\r
---\r
\r
## Workflow / 标准工作流程\r
\r
### Mode 1: Quick Start (已知产品 + 快速训练)\r
\r
```\r
User: "帮我准备明天拜访客户B的训练，他是私企老板，对健康险感兴趣"\r
  │\r
  ▼\r
[Step 1] 获取代理人信息 → 张明，L2，弱项：健康险异议处理\r
[Step 2] 识别拜访产品 → 康健医疗保险（目标：替换平安福）\r
[Step 3] 生成训练计划 → 午间30分钟：健康险异议处理对练\r
[Step 4] 开始陪练 → 情景对练：私企业主健康险需求挖掘\r
[Step 5] 实时反馈 → 异议处理评分：71/100，给出改进建议\r
[Step 6] 报告输出 → 训练报告 + 明日拜访话术优化建议\r
```\r
\r
### Mode 2: Product Document Upload (上传产品文档)\r
\r
```\r
User: [上传 XX保险公司福享人生终身寿险 产品手册 PDF]\r
  │\r
  ▼\r
[Step 1] 解析文档 → 提取产品结构、条款、卖点\r
[Step 2] 生成产品画像 → Structured JSON Profile\r
[Step 3] 生成问题库 → 65+道题目（5难度×8类别）\r
[Step 4] 与现有产品库合并 → 更新知识库\r
[Step 5] 等待选择 → "请选择训练模式：快问快答 / 情景对练 / 案例研讨"\r
```\r
\r
### Mode 3: Full Agent Assessment (全面能力评估)\r
\r
```\r
User: "帮我评估代理人李华的综合能力，她入职8个月，主要卖重疾险"\r
  │\r
  ▼\r
[Step 1] 建立代理人档案 → L1入门级，8个月，重疾险方向\r
[Step 2] 产品文档上传 → 重疾险产品手册\r
[Step 3] 综合考核 → 30题产品知识 + 5个情景对练\r
[Step 4] 生成能力雷达图 → 6维度能力可视化\r
[Step 5] 制定成长路径 → 90天训练计划\r
```\r
\r
---\r
\r
## Input / Output Specifications / 输入输出规范\r
\r
### Input\r
\r
| Input Type | Description | Example |\r
|------------|-------------|---------|\r
| 代理人档案 | JSON/文本描述 | 姓名、级别、工龄、业绩、弱项 |\r
| 产品文档 | PDF/Word/TXT/图片 | 保险产品手册、条款、计划书 |\r
| 当日行程 | 文本/日历 | 09:00晨会 / 10:30拜访客户A |\r
| 训练指令 | 自然语言 | "帮我准备健康险的陪练" |\r
| 客户信息 | 文本描述 | "45岁私企老板，年收入200万" |\r
\r
### Output\r
\r
| Output Type | Description |\r
|-------------|-------------|\r
| 产品画像JSON | 结构化产品信息 |\r
| 问题库 | 65+道分类分级题目 |\r
| 训练计划 | 分钟级个性化日程 |\r
| 陪练对话 | 实时AI角色扮演 |\r
| 评估报告 | 评分 + 改进建议 + 雷达图 |\r
| 成长路径 | 30/60/90天训练建议 |\r
\r
---\r
\r
## Integration Notes / 集成说明\r
\r
**Data privacy:**\r
- All agent and client data remains local / within the company's system\r
- No sensitive PII should be included in training documents\r
- Comply with China CBIRC insurance sales compliance regulations\r
\r
**Lianxi with other Skills:**\r
- `insurance-bidding-pro`: Use product analysis for bidding scenarios\r
- `insurance-private-domain-ops`: Link training completion to customer follow-up\r
- `insurance-claims-intelligence`: Train agents on claim processes for better client communication\r
\r
---\r
\r
## Disclaimer / 免责声明\r
\r
> ⚠️ **Training is advisory only.** This skill provides coaching materials, question banks,\r
> and simulation training for insurance agent development. All final sales advice,\r
> compliance decisions, and product recommendations must be reviewed by licensed\r
> insurance professionals and comply with CBIRC regulations. Model answers represent\r
> reference best practices, not guaranteed outcomes.\r