diff --git a/app/service/agent/implementations/risk_agent.py b/app/service/agent/implementations/risk_agent.py
index 489a972..38e57a6 100644
--- a/app/service/agent/implementations/risk_agent.py
+++ b/app/service/agent/implementations/risk_agent.py
@@ -362,7 +362,8 @@ def _agent_system_prompt(message: str) -> str:
         "8. 查询结果包含 summary 时，客户、产品和规则数量必须依据完整 summary，"
         "不能因为 items 被截断就回答只覆盖部分记录。\n"
         "9. 最终回答使用中文，简洁说明结论、依据和剩余风险，并提醒由风控专员人工复核。\n"
-        f"{context}\n{filter_context}\n{_truncation_instruction()}"
+        f"{context}\n{filter_context}\n{_truncation_instruction()}\n"
+        f"{_field_meaning_instruction()}"
     )
 
 
@@ -374,6 +375,14 @@ def _truncation_instruction() -> str:
     )
 
 
+def _field_meaning_instruction() -> str:
+    return (
+        "11. 工具结果中的 field_meanings 是字段中文含义词典。面向用户回答时必须使用"
+        "这些业务含义组织内容，不要直接罗列 customer_no、ack_status、behavior_score "
+        "等数据库字段名；只有用户明确要求查看原始字段时才保留字段名。"
+    )
+
+
 def _is_disposition_query(message: str) -> bool:
     return any(keyword in message for keyword in (
         "误报",
diff --git a/app/service/risk_tools.py b/app/service/risk_tools.py
index 9237195..4bc7e9d 100644
--- a/app/service/risk_tools.py
+++ b/app/service/risk_tools.py
@@ -16,6 +16,111 @@ from app.service.risk_judgement_service import (
 )
 from app.service.risk_query_service import scope_from_context
 
+FIELD_MEANINGS: dict[str, str] = {
+    "alert_no": "预警编号",
+    "customer_id": "客户编号",
+    "customer_no": "客户编号",
+    "customer_name": "客户姓名（脱敏）",
+    "product_code": "产品代码",
+    "product_name": "产品名称",
+    "alert_type": "预警类型",
+    "risk_level": "风险等级",
+    "rule_codes": "命中规则编号",
+    "triggered_rules": "命中规则编号",
+    "evidence_summary": "证据摘要",
+    "event_status": "事件状态",
+    "status": "业务状态",
+    "ack_status": "预警确认状态",
+    "created_at": "创建时间",
+    "updated_at": "更新时间",
+    "due_at": "处理截止时间",
+    "ack_at": "确认时间",
+    "escalated_at": "升级时间",
+    "priority_score": "优先级分数",
+    "is_escalated": "是否已升级",
+    "evidence_archived": "证据是否已归档",
+    "close_reason": "关闭原因",
+    "disposition_hint": "列表级只读研判建议",
+    "disposition_assessment": "详情级只读研判草案",
+    "name": "客户姓名（脱敏）",
+    "birth_date": "出生日期",
+    "occupation": "职业",
+    "mobile_masked": "手机号（脱敏）",
+    "investor_type": "投资者风险承受等级",
+    "investment_horizon": "投资期限",
+    "preferred_asset_class": "偏好资产类别",
+    "trading_frequency": "交易频率",
+    "total_asset": "总资产",
+    "behavior_score": "行为分",
+    "risk_tags": "风险标签",
+    "trade_account": "交易账号",
+    "product_category": "产品类别",
+    "risk_disclosure_required": "是否要求风险揭示",
+    "second_confirmation_required": "是否要求二次确认",
+    "recording_required": "是否要求双录",
+    "transaction_no": "交易编号",
+    "order_side": "买卖方向",
+    "transaction_type": "交易类型",
+    "amount": "金额",
+    "channel": "交易渠道",
+    "trade_status": "交易状态",
+    "risk_disclosure_signed": "风险揭示签署状态",
+    "second_confirmation": "二次确认状态",
+    "confirmed_at": "确认时间",
+    "executed_at": "成交时间",
+    "flow_no": "资金流水编号",
+    "flow_type": "资金流水类型",
+    "settled_at": "到账时间",
+    "occurred_at": "发生时间",
+    "source_type": "资金来源",
+    "match_status": "资金匹配状态",
+    "shares": "持有份额",
+    "cost_amount": "持仓成本",
+    "current_value": "当前市值",
+    "profit_loss": "浮动盈亏",
+    "profit_loss_ratio": "盈亏比例",
+    "holding_days": "持有天数",
+    "holding_ratio": "持仓占资产比例",
+    "login_at": "登录时间",
+    "login_result": "登录结果",
+    "ip_region": "登录地区",
+    "device_id": "登录设备标识",
+    "is_common_device": "是否常用设备",
+    "failure_reason": "失败原因",
+    "notification_no": "通知编号",
+    "send_status": "发送状态",
+    "receiver_email": "接收邮箱",
+    "sent_at": "发送时间",
+    "handler_id": "处理人编号",
+    "related_transaction_id": "关联交易编号",
+    "related_work_order_id": "关联工单编号",
+    "evidence_truncated": "被截断的证据类型",
+    "data_truncated": "数据是否被截断",
+    "truncated": "结果是否被截断",
+}
+
+
+def with_field_meanings(value: dict[str, Any]) -> dict[str, Any]:
+    """附加当前结果实际出现字段的中文含义，供模型转换后回答。"""
+    names: set[str] = set()
+
+    def collect(item: object) -> None:
+        if isinstance(item, dict):
+            for key, child in item.items():
+                names.add(str(key))
+                collect(child)
+        elif isinstance(item, (list, tuple)):
+            for child in item:
+                collect(child)
+
+    collect(value)
+    meanings = {
+        name: FIELD_MEANINGS[name]
+        for name in sorted(names)
+        if name in FIELD_MEANINGS
+    }
+    return {**value, "field_meanings": meanings}
+
 
 async def search_risk_alerts_tool(
     arguments: RiskAlertQuery,
@@ -61,7 +166,7 @@ def build_alert_search_result(
         hint = item.get("disposition_hint") or {}
         if isinstance(hint, dict) and hint.get("verdict"):
             disposition_counts[str(hint["verdict"])] += 1
-    return {
+    return with_field_meanings({
         "total": len(compact_items),
         "filters": filters or {},
         "summary": {
@@ -76,7 +181,7 @@ def build_alert_search_result(
             "complete": True,
         },
         "items": compact_items,
-    }
+    })
 
 
 def _compact_alert_item(item: dict[str, Any]) -> dict[str, Any]:
@@ -170,7 +275,8 @@ async def get_risk_overview_tool(
 ) -> dict[str, Any]:
     del arguments
     async with SessionFactory() as session:
-        return await RiskRepository(session, scope=scope_from_context(context)).overview()
+        data = await RiskRepository(session, scope=scope_from_context(context)).overview()
+        return with_field_meanings(data)
 
 
 async def get_alert_evidence_tool(
@@ -186,4 +292,4 @@ async def get_alert_evidence_tool(
             return None
         detail = record.to_dict()
         detail["disposition_assessment"] = assess_alert_detail(detail)
-        return detail
+        return with_field_meanings(detail)
diff --git a/tests/contract/test_risk_agent_contract.py b/tests/contract/test_risk_agent_contract.py
index e4020c6..931273d 100644
--- a/tests/contract/test_risk_agent_contract.py
+++ b/tests/contract/test_risk_agent_contract.py
@@ -446,3 +446,11 @@ def test_agent_prompt_requires_truncation_disclosure() -> None:
 
     assert "data_truncated=true" in prompt
     assert "证据不完整" in prompt
+
+
+def test_agent_prompt_requires_business_field_names() -> None:
+    prompt = public_risk_agent._agent_system_prompt("查看当前预警")
+
+    assert "field_meanings" in prompt
+    assert "customer_no" in prompt
+    assert "ack_status" in prompt
diff --git a/tests/unit/service/test_risk_search_result.py b/tests/unit/service/test_risk_search_result.py
index 7ebd69c..9fc1682 100644
--- a/tests/unit/service/test_risk_search_result.py
+++ b/tests/unit/service/test_risk_search_result.py
@@ -25,3 +25,6 @@ def test_search_result_summary_contains_all_customers_and_products() -> None:
     assert result["summary"]["complete"] is True
     assert result["summary"]["disposition_counts"] == {"可考虑放行": 12}
     assert len(result["items"]) == 12
+    assert result["field_meanings"]["customer_no"] == "客户编号"
+    assert result["field_meanings"]["risk_level"] == "风险等级"
+    assert result["field_meanings"]["disposition_hint"] == "列表级只读研判建议"