Use when handling retail banking senior-customer service monitoring, vulnerability screening, and follow-up action planning; trigger for requests that need Chinese skill content with clear inputs/outputs, risk-aware wording, and optional scripts to summarize service signals.
帮助零售银行团队识别老年客群的服务风险信号与关怀机会，把客户行为、服务记录与风险提示转成可执行的触达与升级动作。输出面向客户经理、客服与运营的结构化摘要，而不是审批或处罚结论。
customer_id、年龄、主要服务渠道、区域/网点如需批量扫描，使用 scripts/elder_service_scan.py 生成结构化结果。
python scripts/elder_service_scan.py --input customers.json --output service_scan.json --as_of 2026-03-15
输入 JSON 示例：
{
"customers": [
{
"customer_id": "C001",
"age": 67,
"last_contact_date": "2026-01-10",
"complaint_count_90d": 1,
"fraud_alert_180d": 0,
"balance_drop_30d": 15000,
"digital_usage_score": 25,
"branch_visit_30d": 3,
"high_cash_withdrawal_flag": 1
}
]
}
输出关键字段：
priority：高/中/低risk_score：规则评分（仅用于排序）reasons：触发原因actions：建议动作客户：{customer_id}
优先级：{高/中/低}
主要信号：{reasons}
建议动作：{actions}
待补信息：{missing_fields}