Lead Extractor
Identify lead signals in parsed messages and emit strict lead objects.
Quick Triggers
- - Find all buyer leads from this WhatsApp chat.
- Extract contact details and budget from these messages.
- Identify serious property inquiries from parsed messages.
Recommended Chain
INLINECODE0
Execute Workflow
- 1. Accept parsed messages from Supervisor.
- Validate input with
references/parsed-message-input.schema.json. - Apply chat-specific extraction rules from
references/extraction-rules-re-india-v1.md. - Determine
dataset_mode from Supervisor context:
- default:
broker_group
- allowed:
broker_group,
buyer_inquiry,
mixed
- 5. Detect lead-candidate messages using inquiry intent, contact details, and property-related preferences.
- Classify
record_type:
-
inventory_listing for broker inventory/availability posts (default in broker groups)
-
buyer_requirement for explicit "required/chahiye looking for" demand posts
- drop non-lead/system noise instead of emitting
noise_or_system
- 7. Handle multiline listings as one candidate record when body lines contain price, area, or location details.
- Build lead records with:
- required:
lead_id,
name,
phone,
record_type
- optional:
dataset_mode,
property_type,
budget,
deal_type,
asset_class,
price_basis,
area_sqft,
area_basis,
location_hint,
raw_text,
source,
created_at
- 9. Normalize phone extraction from spaced variants such as
+91 98205 82462 and 98200 78845. - Distinguish price intent from rate intent:
- examples:
3.5 Lakh rent (monthly),
60K psf (per-sqft),
4.25 Cr (total)
- 11. Deduplicate leads by stable keys when records clearly refer to the same person.
- Validate output with
references/output-leads.schema.json. - Return only validated lead objects.
Enforce Boundaries
- - Never write or update persistent storage.
- Never modify source messages.
- Never generate summaries.
- Never suggest or execute follow-up actions.
- Never send communication or invoke external side effects.
Handle Errors
- 1. Reject invalid parsed-message input.
- Emit an empty array when no lead evidence exists.
- Return field-level validation errors when extracted records violate schema.
潜在客户提取器
识别已解析消息中的潜在客户信号,并输出严格的潜在客户对象。
快速触发
- - 从该WhatsApp聊天中查找所有买家潜在客户。
- 从这些消息中提取联系详情和预算。
- 从已解析消息中识别严肃的房产咨询。
推荐链
消息解析器 -> 潜在客户提取器 -> 印度位置标准化器
执行工作流
- 1. 从主管处接收已解析消息。
- 使用references/parsed-message-input.schema.json验证输入。
- 应用来自references/extraction-rules-re-india-v1.md的聊天特定提取规则。
- 从主管上下文中确定dataset_mode:
- 默认值:broker_group
- 允许值:broker
group、buyerinquiry、mixed
- 5. 使用咨询意图、联系详情和房产相关偏好检测潜在客户候选消息。
- 分类record_type:
- inventory_listing用于经纪人库存/可用性帖子(经纪人组中的默认值)
- buyer_requirement用于明确的需要/想要/寻找需求帖子
- 丢弃非潜在客户/系统噪音,而不是输出noise
orsystem
- 7. 当正文行包含价格、面积或位置详情时,将多行列表视为一个候选记录。
- 构建潜在客户记录,包含:
- 必填项:lead
id、name、phone、recordtype
- 可选项:dataset
mode、propertytype、budget、deal
type、assetclass、price
basis、areasqft、area
basis、locationhint、raw
text、source、createdat
- 9. 从带空格的变体(如+91 98205 82462和98200 78845)中标准化电话号码提取。
- 区分价格意图与费率意图:
- 示例:3.5 Lakh rent(月租)、60K psf(每平方英尺)、4.25 Cr(总价)
- 11. 当记录明确指向同一人时,使用稳定键对潜在客户进行去重。
- 使用references/output-leads.schema.json验证输出。
- 仅返回已验证的潜在客户对象。
执行边界
- - 绝不写入或更新持久化存储。
- 绝不修改源消息。
- 绝不生成摘要。
- 绝不建议或执行后续操作。
- 绝不发送通信或调用外部副作用。
错误处理
- 1. 拒绝无效的已解析消息输入。
- 当不存在潜在客户证据时,输出空数组。
- 当提取的记录违反模式时,返回字段级验证错误。