SuperAPI AI Directory
Partner中文AI newsFavorites
Skillssmart-product-matching

smart-product-matching

Marketing & growth

View repository

Updated 10-16 08:54

About this skill

Intelligent Product Matching

Available Tools

This skill can call the following MCP data services, to be selected as needed during the execution process:

Yingmi Financial Data (qieman)

  • Service address:https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse
  • Core capabilities: fund search/diagnosis, portfolio analysis/backtesting, asset allocation solutions, CFP toolchain, chart rendering
  • Main tools of this skill:SearchFunds, GetPopularFund, BatchGetFundsDetail, GetBatchFundPerformance, AnalyzeFundRisk, GetFundDiagnosis, RenderEchart

Hundsun Juyuan Financial Data (gildata-aidata)

  • Service address: obtained after activating the Hundsun Juyuan MCP service, in the format https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp
  • Core capabilities: individual stock research (A/H/US), financial statements, capital flows, research reports and public sentiment, wealth management products, macroeconomic data
  • Main tools of this skill:FundMultipleFactorFilter, FinancialProductFilter, ProductBasicInfoList, CreditBondBaseInfo

Core Principles

Charts first, concise text. The performance comparison data of recommended products must be generated through RenderEchart to create visual charts, allowing the account manager to see product strengths and weaknesses at a glance, with text focusing on matching rationale and sales pitches.

Input Requirements

Required Information

  • Customer risk level (R1-R5)
  • Requirement description (product type/configuration gap/investment amount, at least one)

Optional Information

  • Current holdings overview (to avoid recommending duplicate or highly correlated products)
  • AUM, customer tier
  • Exclusion conditions (such as "no closed-end" "no new funds")
  • Product category preference (funds/wealth management products/bonds)

If the user only says "recommend a few funds", ask follow-up questions about risk level and product type requirements.

Execution Process

Step 1: Requirement Analysis

  • Determine screening dimensions: product type, risk constraints, scale requirements, liquidity needs
  • Determine product category: funds / wealth management products / bonds / comprehensive matching

Step 2: Product Screening (routed by category)

Fund-type requirements (qieman + gildata-aidata):

  • SearchFunds(qieman): search funds by conditions
  • GetPopularFund(qieman): obtain popular funds as supplementary candidates
  • FundMultipleFactorFilter(gildata-aidata): multi-factor screening, precisely screening funds from multiple dimensions such as return, risk, and style

Wealth management product requirements (gildata-aidata):

  • FinancialProductFilter: screen bank wealth management products by conditions such as risk level, term, and yield
  • ProductBasicInfoList: obtain basic information on wealth management products (issuer, term, performance benchmark, etc.)

Bond-type requirements (gildata-aidata):

  • CreditBondBaseInfo: obtain basic information on credit bonds (rating, coupon rate, maturity date, etc.)

Step 3: Due Diligence Verification of Candidate Products (3-5)

Fund-type verification (qieman):

  • BatchGetFundsDetail: detailed information
  • GetBatchFundPerformance: performance data
  • AnalyzeFundRisk: risk assessment
  • GetFundDiagnosis: diagnostic information

Wealth management product verification (gildata-aidata):

  • ProductBasicInfoList: product details, historical performance, fee structure

Step 4: Deduplicate with Current Holdings (if holdings data is available)

  • Check the overlap between recommended products and current holdings

Step 5: Generate visualization charts (mandatory, cannot be skipped)

MustCall RenderEchart to generate the following charts:

Chart 1: Recommended product performance comparison bar chart (must be generated)

Use the candidate products' trailing 1-year return to generate a bar chart, visually showing each product's performance. ECharts option reference:

{
  "title": { "text": "推荐产品业绩对比", "left": "center" },
  "tooltip": { "trigger": "axis" },
  "grid": { "left": "30%", "right": "10%" },
  "xAxis": { "type": "value", "axisLabel": { "formatter": "{value}%" } },
  "yAxis": { "type": "category", "data": ["广发纯债A", "招商中证白酒", "易方达蓝筹精选"] },
  "series": [{
    "type": "bar",
    "data": [3.5, 12.8, 18.2],
    "label": { "show": true, "position": "right", "formatter": "{c}%" },
    "itemStyle": { "borderRadius": [0, 4, 4, 0], "color": "#5470c6" }
  }]
}

Replace with the actual recommended product names and returns. Sort by return from smallest to largest (highest at the top).

Chart 2: Recommended product multi-dimensional comparison radar chart (recommended to generate, if there are ≥2 products)

Score the recommended products on dimensions such as return, risk, scale, and fees from 1-5, and generate a radar chart. ECharts option reference:

{
  "title": { "text": "推荐产品综合评分", "left": "center" },
  "legend": { "top": "bottom" },
  "radar": {
    "indicator": [
      { "name": "收益", "max": 5 },
      { "name": "风险控制", "max": 5 },
      { "name": "规模稳定", "max": 5 },
      { "name": "费率", "max": 5 },
      { "name": "经理能力", "max": 5 }
    ],
    "shape": "circle"
  },
  "series": [{
    "type": "radar",
    "data": [
      { "value": [4, 3, 4, 3, 5], "name": "推荐一", "areaStyle": { "opacity": 0.2 } },
      { "value": [3, 5, 4, 4, 3], "name": "推荐二", "areaStyle": { "opacity": 0.2 } },
      { "value": [5, 2, 3, 3, 4], "name": "推荐三", "areaStyle": { "opacity": 0.2 } }
    ]
  }]
}

Replace with the actual product names and scores.

Step 6: Generate the recommendation plan and talking points

Output template

Output according to the following structure,charts embedded in the corresponding sections, and the recommendation reason for each product kept concise：

## 产品智能匹配

### 匹配条件
- **风险等级**：{R等级}
- **产品类型**：{类型}
- **配置目的**：{描述}

### 业绩一览

{推荐产品业绩对比柱状图}

### 综合评分

{推荐产品多维对比雷达图}

### 推荐详情

#### 推荐一：{产品名称}({代码})
| 类型 | 规模/期限 | 经理/发行方 | 近1年/业绩基准 | 最大回撤 | 风险等级 |
|------|----------|-----------|--------------|---------|---------|
| {类型} | {X}亿/{X}月 | {姓名/机构} | {X}% | {X}% | {R等级} |

**为什么适合**：{1-2句}
> **话术**："{推荐话术}"

#### 推荐二：{产品名称}({代码})
{同上格式}

#### 推荐三：{产品名称}({代码})
{同上格式}

---
*产品推荐基于量化筛选和历史数据，不构成投资建议。基金有风险，投资需谨慎。*

Notes

  • Charts are mandatory: The recommended product performance comparison bar chart is a must-generate item
  • Compliance red line: Do not use "recommend buying"; use "for reference" and "worth watching"
  • Suitability matching: The risk level of recommended products must not be higher than the client's risk level
  • Objective presentation: Present both advantages and risk points for each product
  • Number of recommendations: Generally recommend 3 products (provide choice space but do not cause choice difficulty)
  • Concise text: Keep the full text within 800-1200 characters (excluding charts)
  • Category labeling: Clearly label the category of each recommended product (fund/wealth management product/bond) to facilitate differentiation by client managers