60b72874b5
本次提交围绕成员画像插件进行了较大升级，核心目标是把原来偏单次、偏近期的成员交互摘要，升级为可随时间沉淀的分层画像能力。 主要功能变更如下： 1. 新增成员分层摘要存储表 t_member_digest，并提供对应的数据库操作层，支持按成员、按群、按摘要类型（daily/weekly/monthly）持久化周期性摘要结果。 2. 在 member_context 插件内新增 MemberDigestService，把画像生成拆分为日摘要、周摘要、月摘要三级处理流程，再由最终画像服务消费这些分层摘要，减少直接反复处理大量原始消息带来的成本和失真。 3. 新增提示词构建模块，分别为日级观察、周级归纳、月级归纳以及最终画像整理提供独立提示词，强调中性、克制、避免敏感推断，并将长期特征与近期状态明确分层。 4. 重写成员最终画像生成逻辑，优先基于日/周/月摘要融合出长期特征、习惯模式、长期回复偏好、近期状态等信息，再用 AI 对分层摘要做最终整理，避免仅依赖近 30 天消息得出偏短期结论。 5. 保留并增强长期画像融合逻辑，通过打分、衰减和重复证据累积，使长期特征随着时间逐步稳定，而不会被单次刷新完全覆盖。 6. 在消息存储层补充成员按时间增量获取、按活跃日期统计、按天取消息等查询方法，为后续分层摘要生成提供数据支撑。 7. 扩展 member_context 插件配置，增加日级摘要消息上限、日摘要最小消息数、单次回填的日摘要数量上限、最终画像使用的日/周/月摘要数量等参数，便于在准确性和系统负载之间做平衡。 8. 后台成员摘要详情页新增长期沟通倾向、长期特征、习惯模式、长期回复偏好、近期状态、历史样本数、分层摘要数量等展示字段，方便观察画像沉淀程度。 9. 优化后台查看成员摘要接口逻辑：首次打开如果还没有摘要，不再同步阻塞生成，而是返回未就绪状态，配合后台手动异步刷新，降低页面卡顿和接口阻塞风险。 10. 增强刷新日志，单成员和群级刷新会输出当前刷新模式以及日/周/月摘要数量，便于排查画像构建进度。 11. 调整当前日、当前周、当前月摘要的重算逻辑，确保新增日摘要写入后，本周和本月摘要不会长期停留在旧版本。 本次提交后，成员画像能力从“基于近期样本的单层摘要”升级为“基于时间沉淀的分层画像管线”，为后续把画像稳定接入 AI 自动回复上下文打下基础，同时尽量保持现有群权限控制和后台异步刷新方式不变。
148 lines
7.8 KiB
Python
148 lines
7.8 KiB
Python
# -*- coding: utf-8 -*-
|
||
import json
|
||
from typing import Dict, List
|
||
|
||
|
||
class MemberContextPromptBuilder:
|
||
"""成员分层画像提示词构建器"""
|
||
|
||
@staticmethod
|
||
def build_daily_digest_prompt(chatroom_id: str, wxid: str, display_name: str,
|
||
digest_date: str, messages: List[Dict]) -> str:
|
||
lines = []
|
||
for msg in messages[-80:]:
|
||
ts = str(msg.get("timestamp", ""))[11:16]
|
||
content = (msg.get("content") or "").replace("\n", " ").strip()[:180]
|
||
if content:
|
||
lines.append(f"[{ts}] {content}")
|
||
|
||
return (
|
||
"你是微信群后台的成员日观察摘要生成器。\n"
|
||
"请仅基于给定的当日公开聊天记录，提取对后续互动有帮助的中性行为观察。\n"
|
||
"不要做人格诊断、隐私猜测、负面评价，不要脑补群外信息。\n"
|
||
"输出严格 JSON，不要 markdown。\n"
|
||
"{"
|
||
"\"topics\":[\"主题1\"],"
|
||
"\"interaction_style\":\"一句中文\","
|
||
"\"message_pattern\":\"一句中文\","
|
||
"\"response_style_hint\":\"一句中文\","
|
||
"\"habit_signals\":[\"信号1\"],"
|
||
"\"engagement_traits\":[\"特征1\"],"
|
||
"\"reply_taboos\":[\"避坑1\"],"
|
||
"\"temperament_signal\":\"一句中文，描述当天显露的沟通倾向，必须克制\","
|
||
"\"summary_text\":\"一段不超过100字的日摘要\","
|
||
"\"representative_messages\":[\"原话1\",\"原话2\"],"
|
||
"\"confidence\":0.0"
|
||
"}\n"
|
||
"要求：\n"
|
||
"1. topics 最多4个，habit_signals 最多4个，engagement_traits 最多4个，reply_taboos 最多3个。\n"
|
||
"2. temperament_signal 只能写当日可观察到的沟通倾向，不可上升为长期性格判断。\n"
|
||
"3. representative_messages 保留最能代表当天风格的短句，最多3条。\n"
|
||
f"成员: {display_name} ({wxid})\n"
|
||
f"群ID: {chatroom_id}\n"
|
||
f"日期: {digest_date}\n"
|
||
"当日消息:\n" + ("\n".join(lines) or "暂无")
|
||
)
|
||
|
||
@staticmethod
|
||
def build_period_digest_prompt(digest_type: str, chatroom_id: str, wxid: str,
|
||
display_name: str, period_key: str, items: List[Dict]) -> str:
|
||
structured_lines = []
|
||
for item in items:
|
||
structured = item.get("structured", {}) or {}
|
||
payload = {
|
||
"period_key": item.get("period_key"),
|
||
"summary_text": item.get("summary_text", ""),
|
||
"topics": structured.get("topics") or structured.get("stable_topics") or structured.get("long_term_topics") or [],
|
||
"habit_signals": structured.get("habit_signals") or structured.get("habit_patterns") or [],
|
||
"engagement_traits": structured.get("engagement_traits") or structured.get("stable_traits") or [],
|
||
"reply_preferences": structured.get("reply_preferences") or structured.get("long_term_reply_preferences") or [],
|
||
"temperament_signal": structured.get("temperament_signal") or structured.get("temperament_tendency") or "",
|
||
"recent_state": structured.get("recent_state") or [],
|
||
}
|
||
structured_lines.append(json.dumps(payload, ensure_ascii=False))
|
||
|
||
if digest_type == "weekly":
|
||
schema = (
|
||
"{"
|
||
"\"stable_topics\":[\"主题1\"],"
|
||
"\"stable_traits\":[\"特征1\"],"
|
||
"\"habit_patterns\":[\"习惯1\"],"
|
||
"\"reply_preferences\":[\"偏好1\"],"
|
||
"\"recent_state\":[\"状态1\"],"
|
||
"\"temperament_tendency\":\"一句中文\","
|
||
"\"summary_text\":\"一段不超过120字的周摘要\","
|
||
"\"confidence\":0.0"
|
||
"}"
|
||
)
|
||
extra = "请从多个日摘要中提炼本周重复出现的模式，过滤单日噪音。"
|
||
else:
|
||
schema = (
|
||
"{"
|
||
"\"long_term_topics\":[\"主题1\"],"
|
||
"\"stable_traits\":[\"特征1\"],"
|
||
"\"habit_patterns\":[\"习惯1\"],"
|
||
"\"long_term_reply_preferences\":[\"偏好1\"],"
|
||
"\"phase_state\":[\"状态1\"],"
|
||
"\"temperament_tendency\":\"一句中文\","
|
||
"\"summary_text\":\"一段不超过140字的月摘要\","
|
||
"\"confidence\":0.0"
|
||
"}"
|
||
)
|
||
extra = "请从多个周摘要中提炼阶段性稳定特征，只有反复出现的模式才能进入长期层。"
|
||
|
||
return (
|
||
f"你是微信群后台的成员{digest_type}摘要生成器。\n"
|
||
f"{extra}\n"
|
||
"不可做心理诊断、负面评价、隐私猜测。输出严格 JSON，不要 markdown。\n"
|
||
f"{schema}\n"
|
||
"要求：\n"
|
||
"1. 所有列表字段最多5项，必须中性克制。\n"
|
||
"2. 只有多个下级摘要反复出现的特征，才允许写进 stable_traits / habit_patterns / long_term_reply_preferences。\n"
|
||
"3. recent_state / phase_state 只描述当前阶段状态，不要冒充长期人格。\n"
|
||
f"成员: {display_name} ({wxid})\n"
|
||
f"群ID: {chatroom_id}\n"
|
||
f"周期: {period_key}\n"
|
||
"下级摘要:\n" + ("\n".join(structured_lines) or "暂无")
|
||
)
|
||
|
||
@staticmethod
|
||
def build_final_context_prompt(chatroom_id: str, wxid: str, display_name: str,
|
||
monthly_digests: List[Dict], weekly_digests: List[Dict],
|
||
daily_digests: List[Dict]) -> str:
|
||
monthly_lines = [json.dumps(item.get("structured", {}), ensure_ascii=False) for item in monthly_digests[:6]]
|
||
weekly_lines = [json.dumps(item.get("structured", {}), ensure_ascii=False) for item in weekly_digests[:4]]
|
||
daily_lines = [json.dumps(item.get("structured", {}), ensure_ascii=False) for item in daily_digests[:6]]
|
||
|
||
return (
|
||
"你是微信群后台的最终成员交互画像整理器。\n"
|
||
"请结合月级、周级、日级摘要，输出一个既有长期层又有近期层的后台交互画像。\n"
|
||
"不要做敏感推断、心理诊断、隐私猜测。输出严格 JSON，不要 markdown。\n"
|
||
"{"
|
||
"\"activity_level\":\"高活跃|中活跃|低活跃|观察中\","
|
||
"\"message_pattern\":\"一句中文\","
|
||
"\"interaction_style\":\"一句中文\","
|
||
"\"response_style_hint\":\"一句中文\","
|
||
"\"topics_of_interest\":[\"主题1\"],"
|
||
"\"recent_focus\":[\"近期主题1\"],"
|
||
"\"stable_traits\":[\"长期特征1\"],"
|
||
"\"habit_patterns\":[\"习惯1\"],"
|
||
"\"long_term_reply_preferences\":[\"偏好1\"],"
|
||
"\"recent_state\":[\"近期状态1\"],"
|
||
"\"temperament_tendency\":\"一句中文\","
|
||
"\"summary_text\":\"一段不超过150字的后台摘要\","
|
||
"\"confidence\":0.0,"
|
||
"\"engagement_traits\":[\"特征1\"],"
|
||
"\"reply_taboos\":[\"避坑1\"]"
|
||
"}\n"
|
||
"要求：\n"
|
||
"1. stable_traits、habit_patterns、long_term_reply_preferences 只从月级和多次重复证据中提取。\n"
|
||
"2. recent_focus、recent_state 更依赖最近周级和日级。\n"
|
||
"3. summary_text 要像后台备注，不要明显暴露在给用户做画像。\n"
|
||
f"成员: {display_name} ({wxid})\n"
|
||
f"群ID: {chatroom_id}\n"
|
||
"月级摘要:\n" + ("\n".join(monthly_lines) or "暂无")
|
||
+ "\n周级摘要:\n" + ("\n".join(weekly_lines) or "暂无")
|
||
+ "\n日级摘要:\n" + ("\n".join(daily_lines) or "暂无")
|
||
)
|