dev：修复记忆构建文本名称问题

llm_tool_benchmark_results.json

@@ -0,0 +1,145 @@
+{
+  "测试时间": "2025-04-28 14:12:36",
+  "测试迭代次数": 10,
+  "不使用工具调用": {
+    "平均耗时": 4.596814393997192,
+    "最短耗时": 2.957131862640381,
+    "最长耗时": 10.121938705444336,
+    "标准差": 2.1705468730949593,
+    "所有耗时": [
+      3.18,
+      4.65,
+      10.12,
+      3.5,
+      4.46,
+      4.24,
+      3.23,
+      6.2,
+      2.96,
+      3.42
+    ]
+  },
+  "不使用工具调用_详细响应": [
+    {
+      "内容摘要": "```json\n{\n  \"action\": \"text_reply\",\n  \"content\": \"怎么啦？\",\n  \"emoji_query\": \"友好地询问\"\n}\n```",
+      "推理内容摘要": ""
+    },
+    {
+      "内容摘要": "decide_reply_action(\n    action=\"text_reply\",\n    reasoning=\"千石连续两次叫了我的名字，显然是想引起我的注意或有事要说，作为礼貌应当回应\",\n    emoji_query=\"友善的回应\"\n)",
+      "推理内容摘要": ""
+    },
+    {
+      "内容摘要": "```json\n{\n  \"action\": \"text_reply\",\n  \"content\": \"怎么啦？我在呢~\",\n  \"emoji_query\": \"友好的询问\",\n  \"reasoning\": \"由于对方连续两次提到我的名字，显然是想与我交流，应当及时给予友好回应避免冷场\"\n}\n```",
+      "推理内容摘要": ""
+    },
+    {
+      "内容摘要": "```json\n{\n  \"action\": \"text_reply\",\n  \"content\": \"在呢在呢~怎么啦？\",\n  \"emoji_query\": \"好奇的回应\"\n}\n```",
+      "推理内容摘要": ""
+    },
+    {
+      "内容摘要": "decide_reply_action(\n    action=\"text_reply\",\n    reasoning=\"千石连续两次提到我的名字，显然需要我回应。文字回复更正式且能明确表示我在关注他的信息。\",\n    emoji_query=\"友好的回应\"\n)",
+      "推理内容摘要": ""
+    },
+    {
+      "内容摘要": "```json\n{\n  \"action\": \"text_reply\",\n  \"reasoning\": \"千石连续两次提到我的名字，显然是需要我的回应。作为日常交流，应该给予友善简短的答复。\",\n  \"emoji_query\": \"疑惑的歪头\"\n}\n```",
+      "推理内容摘要": ""
+    },
+    {
+      "内容摘要": "```json\n{\n  \"action\": \"text_reply\",\n  \"content\": \"咋啦千石~\",\n  \"emoji_query\": \"好奇的询问\"\n}\n```",
+      "推理内容摘要": ""
+    },
+    {
+      "内容摘要": "decide_reply_action\n```json\n{\n  \"action\": \"text_reply\",\n  \"content\": \"我在呢～怎么啦？\",\n  \"emoji_query\": \"友好的关心\",\n  \"reasoning\": \"千石连续两次呼唤我的名字，显然是有事情要找我或想引起我的注意。根据回复原则2（有人提到你但未回应），应该用友善的文字进行回应，并附上表达关心的表情符号来延...",
+      "推理内容摘要": ""
+    },
+    {
+      "内容摘要": "```json\n{\n  \"action\": \"text_reply\",\n  \"content\": \"千石怎么啦～\",\n  \"emoji_query\": \"好奇的探询\"\n}\n```",
+      "推理内容摘要": ""
+    },
+    {
+      "内容摘要": "```json\n{\n  \"action\": \"text_reply\",\n  \"content\": \"怎么啦？突然叫我两次\",\n  \"emoji_query\": \"好奇的疑问\"\n}\n```",
+      "推理内容摘要": ""
+    }
+  ],
+  "使用工具调用": {
+    "平均耗时": 8.139546775817871,
+    "最短耗时": 4.9980738162994385,
+    "最长耗时": 18.803313732147217,
+    "标准差": 4.008772720760647,
+    "所有耗时": [
+      5.81,
+      18.8,
+      6.06,
+      8.06,
+      10.07,
+      6.34,
+      7.9,
+      6.66,
+      5.0,
+      6.69
+    ]
+  },
+  "使用工具调用_详细响应": [
+    {
+      "内容摘要": "",
+      "推理内容摘要": "",
+      "工具调用数量": 0,
+      "工具调用详情": []
+    },
+    {
+      "内容摘要": "",
+      "推理内容摘要": "",
+      "工具调用数量": 0,
+      "工具调用详情": []
+    },
+    {
+      "内容摘要": "",
+      "推理内容摘要": "",
+      "工具调用数量": 0,
+      "工具调用详情": []
+    },
+    {
+      "内容摘要": "",
+      "推理内容摘要": "",
+      "工具调用数量": 0,
+      "工具调用详情": []
+    },
+    {
+      "内容摘要": "",
+      "推理内容摘要": "",
+      "工具调用数量": 0,
+      "工具调用详情": []
+    },
+    {
+      "内容摘要": "",
+      "推理内容摘要": "",
+      "工具调用数量": 0,
+      "工具调用详情": []
+    },
+    {
+      "内容摘要": "",
+      "推理内容摘要": "",
+      "工具调用数量": 0,
+      "工具调用详情": []
+    },
+    {
+      "内容摘要": "",
+      "推理内容摘要": "",
+      "工具调用数量": 0,
+      "工具调用详情": []
+    },
+    {
+      "内容摘要": "",
+      "推理内容摘要": "",
+      "工具调用数量": 0,
+      "工具调用详情": []
+    },
+    {
+      "内容摘要": "",
+      "推理内容摘要": "",
+      "工具调用数量": 0,
+      "工具调用详情": []
+    }
+  ],
+  "差异百分比": 77.07
+}

src/heart_flow/mai_state_manager.py

@@ -8,8 +8,8 @@ from src.plugins.moods.moods import MoodManager
 logger = get_logger("mai_state")
 
 
-# enable_unlimited_hfc_chat = True
+enable_unlimited_hfc_chat = True
-enable_unlimited_hfc_chat = False
+# enable_unlimited_hfc_chat = False
 
 
 class MaiState(enum.Enum):

src/heart_flow/sub_mind.py

@@ -8,7 +8,7 @@ from src.individuality.individuality import Individuality
 import random
 from ..plugins.utils.prompt_builder import Prompt, global_prompt_manager
 from src.do_tool.tool_use import ToolUser
-from src.plugins.utils.json_utils import safe_json_dumps, normalize_llm_response, process_llm_tool_calls
+from src.plugins.utils.json_utils import safe_json_dumps, process_llm_tool_calls
 from src.heart_flow.chat_state_info import ChatStateInfo
 from src.plugins.chat.chat_stream import chat_manager
 from src.plugins.heartFC_chat.heartFC_Cycleinfo import CycleInfo
@@ -20,14 +20,12 @@ logger = get_logger("sub_heartflow")
 def init_prompt():
     prompt = ""
     prompt += "{extra_info}\n"
-    prompt += "{prompt_personality}\n"
+    prompt += "你的名字是{bot_name},{prompt_personality}\n"
     prompt += "{last_loop_prompt}\n"
     prompt += "{cycle_info_block}\n"
     prompt += "现在是{time_now}，你正在上网，和qq群里的网友们聊天，以下是正在进行的聊天内容：\n{chat_observe_info}\n"
     prompt += "\n你现在{mood_info}\n"
-    prompt += (
+    prompt += "请仔细阅读当前群聊内容，分析讨论话题和群成员关系，分析你刚刚发言和别人对你的发言的反应，思考你要不要回复。然后思考你是否需要使用函数工具。"
-        "请仔细阅读当前群聊内容，分析讨论话题和群成员关系，分析你刚刚发言和别人对你的发言的反应，思考你要不要回复。"
-    )
     prompt += "思考并输出你的内心想法\n"
     prompt += "输出要求：\n"
     prompt += "1. 根据聊天内容生成你的想法，{hf_do_next}\n"
@@ -80,8 +78,6 @@ class SubMind:
         # 更新活跃时间
         self.last_active_time = time.time()
 
-        
-        
         # ---------- 1. 准备基础数据 ----------
         # 获取现有想法和情绪状态
         current_thinking_info = self.current_mind
@@ -106,18 +102,7 @@ class SubMind:
         individuality = Individuality.get_instance()
 
         # 构建个性部分
-        prompt_personality = f"你正在扮演名为{individuality.personality.bot_nickname}的人类，你"
+        prompt_personality = individuality.get_prompt(x_person=2, level=2)
-        prompt_personality += individuality.personality.personality_core
-
-        # 随机添加个性侧面
-        if individuality.personality.personality_sides:
-            random_side = random.choice(individuality.personality.personality_sides)
-            prompt_personality += f"，{random_side}"
-
-        # 随机添加身份细节
-        if individuality.identity.identity_detail:
-            random_detail = random.choice(individuality.identity.identity_detail)
-            prompt_personality += f"，{random_detail}"
 
         # 获取当前时间
         time_now = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
@@ -211,7 +196,7 @@ class SubMind:
         prompt = (await global_prompt_manager.get_prompt_async("sub_heartflow_prompt_before")).format(
             extra_info="",  # 可以在这里添加额外信息
             prompt_personality=prompt_personality,
-            bot_name=individuality.personality.bot_nickname,
+            bot_name=individuality.name,
             time_now=time_now,
             chat_observe_info=chat_observe_info,
             mood_info=mood_info,
@@ -228,7 +213,9 @@ class SubMind:
 
         try:
             # 调用LLM生成响应
-            response, _reasoning_content, tool_calls = await self.llm_model.generate_response_tool_async(prompt=prompt, tools=tools)
+            response, _reasoning_content, tool_calls = await self.llm_model.generate_response_tool_async(
+                prompt=prompt, tools=tools
+            )
 
             logger.debug(f"{self.log_prefix} 子心流输出的原始LLM响应: {response}")
 
@@ -246,9 +233,7 @@ class SubMind:
                     tool_calls_str = ", ".join(
                         [call.get("function", {}).get("name", "未知工具") for call in valid_tool_calls]
                     )
-                    logger.info(
+                    logger.info(f"{self.log_prefix} 模型请求调用{len(valid_tool_calls)}个工具: {tool_calls_str}")
-                        f"{self.log_prefix} 模型请求调用{len(valid_tool_calls)}个工具: {tool_calls_str}"
-                    )
 
                     # 收集工具执行结果
                     await self._execute_tool_calls(valid_tool_calls, tool_instance)

src/individuality/identity.py

@@ -1,6 +1,5 @@
 from dataclasses import dataclass
 from typing import List
-import random
 
 
 @dataclass
@@ -86,27 +85,6 @@ class Identity:
         instance.appearance = appearance
         return instance
 
-    def get_prompt(self, x_person, level):
-        """
-        获取身份特征的prompt
-        """
-        if x_person == 2:
-            prompt_identity = "你"
-        elif x_person == 1:
-            prompt_identity = "我"
-        else:
-            prompt_identity = "他"
-
-        if level == 1:
-            identity_detail = self.identity_detail
-            random.shuffle(identity_detail)
-            prompt_identity += identity_detail[0]
-        elif level == 2:
-            for detail in self.identity_detail:
-                prompt_identity += f",{detail}"
-        prompt_identity += "。"
-        return prompt_identity
-
     def to_dict(self) -> dict:
         """将身份特征转换为字典格式"""
         return {

src/individuality/individuality.py

@@ -1,6 +1,7 @@
 from typing import Optional
 from .personality import Personality
 from .identity import Identity
+import random
 
 
 class Individuality:
@@ -8,15 +9,16 @@ class Individuality:
 
     _instance = None
 
-    def __new__(cls, *args, **kwargs):
-        if cls._instance is None:
-            cls._instance = super().__new__(cls)
-        return cls._instance
-
     def __init__(self):
+        if Individuality._instance is not None:
+            raise RuntimeError("Individuality 类是单例，请使用 get_instance() 方法获取实例。")
+
+        # 正常初始化实例属性
         self.personality: Optional[Personality] = None
         self.identity: Optional[Identity] = None
 
+        self.name = ""
+
     @classmethod
     def get_instance(cls) -> "Individuality":
         """获取Individuality单例实例
@@ -25,7 +27,13 @@ class Individuality:
             Individuality: 单例实例
         """
         if cls._instance is None:
-            cls._instance = cls()
+            # 实例不存在，调用 cls() 创建新实例
+            # cls() 会调用 __init__
+            # 因为此时 cls._instance 仍然是 None，__init__ 会正常执行初始化
+            new_instance = cls()
+            # 将新创建的实例赋值给类变量 _instance
+            cls._instance = new_instance
+        # 返回（新创建的或已存在的）单例实例
         return cls._instance
 
     def initialize(
@@ -63,6 +71,8 @@ class Individuality:
             identity_detail=identity_detail, height=height, weight=weight, age=age, gender=gender, appearance=appearance
         )
 
+        self.name = bot_nickname
+
     def to_dict(self) -> dict:
         """将个体特征转换为字典格式"""
         return {
@@ -80,16 +90,148 @@ class Individuality:
             instance.identity = Identity.from_dict(data["identity"])
         return instance
 
-    def get_prompt(self, type, x_person, level):
+    def get_personality_prompt(self, level: int, x_person: int = 2) -> str:
         """
-        获取个体特征的prompt
+        获取人格特征的prompt
+
+        Args:
+            level (int): 详细程度 (1: 核心, 2: 核心+随机侧面, 3: 核心+所有侧面)
+            x_person (int, optional): 人称代词 (0: 无人称, 1: 我, 2: 你). 默认为 2.
+
+        Returns:
+            str: 生成的人格prompt字符串
         """
-        if type == "personality":
+        if x_person not in [0, 1, 2]:
-            return self.personality.get_prompt(x_person, level)
+            return "无效的人称代词，请使用 0 (无人称), 1 (我) 或 2 (你)。"
-        elif type == "identity":
+        if not self.personality:
-            return self.identity.get_prompt(x_person, level)
+            return "人格特征尚未初始化。"
+
+        if x_person == 2:
+            p_pronoun = "你"
+            prompt_personality = f"{p_pronoun}{self.personality.personality_core}"
+        elif x_person == 1:
+            p_pronoun = "我"
+            prompt_personality = f"{p_pronoun}{self.personality.personality_core}"
+        else:  # x_person == 0
+            p_pronoun = ""  # 无人称
+            # 对于无人称，直接描述核心特征
+            prompt_personality = f"{self.personality.personality_core}"
+
+        # 根据level添加人格侧面
+        if level >= 2 and self.personality.personality_sides:
+            personality_sides = list(self.personality.personality_sides)
+            random.shuffle(personality_sides)
+            if level == 2:
+                prompt_personality += f"，有时也会{personality_sides[0]}"
+            elif level == 3:
+                sides_str = "、".join(personality_sides)
+                prompt_personality += f"，有时也会{sides_str}"
+        prompt_personality += "。"
+        return prompt_personality
+
+    def get_identity_prompt(self, level: int, x_person: int = 2) -> str:
+        """
+        获取身份特征的prompt
+
+        Args:
+            level (int): 详细程度 (1: 随机细节, 2: 所有细节+外貌年龄性别, 3: 同2)
+            x_person (int, optional): 人称代词 (0: 无人称, 1: 我, 2: 你). 默认为 2.
+
+        Returns:
+            str: 生成的身份prompt字符串
+        """
+        if x_person not in [0, 1, 2]:
+            return "无效的人称代词，请使用 0 (无人称), 1 (我) 或 2 (你)。"
+        if not self.identity:
+            return "身份特征尚未初始化。"
+
+        if x_person == 2:
+            i_pronoun = "你"
+        elif x_person == 1:
+            i_pronoun = "我"
+        else:  # x_person == 0
+            i_pronoun = ""  # 无人称
+
+        identity_parts = []
+
+        # 根据level添加身份细节
+        if level >= 1 and self.identity.identity_detail:
+            identity_detail = list(self.identity.identity_detail)
+            random.shuffle(identity_detail)
+            if level == 1:
+                identity_parts.append(f"身份是{identity_detail[0]}")
+            elif level >= 2:
+                details_str = "、".join(identity_detail)
+                identity_parts.append(f"身份是{details_str}")
+
+        # 根据level添加其他身份信息
+        if level >= 3:
+            if self.identity.appearance:
+                identity_parts.append(f"{self.identity.appearance}")
+            if self.identity.age > 0:
+                identity_parts.append(f"年龄大约{self.identity.age}岁")
+            if self.identity.gender:
+                identity_parts.append(f"性别是{self.identity.gender}")
+
+        if identity_parts:
+            details_str = "，".join(identity_parts)
+            if x_person in [1, 2]:
+                return f"{i_pronoun}，{details_str}。"
+            else:  # x_person == 0
+                # 无人称时，直接返回细节，不加代词和开头的逗号
+                return f"{details_str}。"
         else:
-            return ""
+            if x_person in [1, 2]:
+                return f"{i_pronoun}的身份信息不完整。"
+            else:  # x_person == 0
+                return "身份信息不完整。"
+
+    def get_prompt(self, level: int, x_person: int = 2) -> str:
+        """
+        获取合并的个体特征prompt
+
+        Args:
+            level (int): 详细程度 (1: 核心/随机细节, 2: 核心+侧面/细节+其他, 3: 全部)
+            x_person (int, optional): 人称代词 (0: 无人称, 1: 我, 2: 你). 默认为 2.
+
+        Returns:
+            str: 生成的合并prompt字符串
+        """
+        if x_person not in [0, 1, 2]:
+            return "无效的人称代词，请使用 0 (无人称), 1 (我) 或 2 (你)。"
+
+        if not self.personality or not self.identity:
+            return "个体特征尚未完全初始化。"
+
+        # 调用新的独立方法
+        prompt_personality = self.get_personality_prompt(level, x_person)
+        prompt_identity = self.get_identity_prompt(level, x_person)
+
+        # 移除可能存在的错误信息，只合并有效的 prompt
+        valid_prompts = []
+        if "尚未初始化" not in prompt_personality and "无效的人称" not in prompt_personality:
+            valid_prompts.append(prompt_personality)
+        if (
+            "尚未初始化" not in prompt_identity
+            and "无效的人称" not in prompt_identity
+            and "信息不完整" not in prompt_identity
+        ):
+            # 从身份 prompt 中移除代词和句号，以便更好地合并
+            identity_content = prompt_identity
+            if x_person == 2 and identity_content.startswith("你，"):
+                identity_content = identity_content[2:]
+            elif x_person == 1 and identity_content.startswith("我，"):
+                identity_content = identity_content[2:]
+            # 对于 x_person == 0，身份提示不带前缀，无需移除
+
+            if identity_content.endswith("。"):
+                identity_content = identity_content[:-1]
+            valid_prompts.append(identity_content)
+
+        # --- 合并 Prompt ---
+        final_prompt = " ".join(valid_prompts)
+
+        return final_prompt.strip()
 
     def get_traits(self, factor):
         """

src/individuality/personality.py

@@ -2,7 +2,6 @@ from dataclasses import dataclass
 from typing import Dict, List
 import json
 from pathlib import Path
-import random
 
 
 @dataclass
@@ -119,28 +118,3 @@ class Personality:
         for key, value in data.items():
             setattr(instance, key, value)
         return instance
-
-    def get_prompt(self, x_person, level):
-        # 开始构建prompt
-        if x_person == 2:
-            prompt_personality = "你"
-        elif x_person == 1:
-            prompt_personality = "我"
-        else:
-            prompt_personality = "他"
-        # person
-
-        prompt_personality += self.personality_core
-
-        if level == 2:
-            personality_sides = self.personality_sides
-            random.shuffle(personality_sides)
-            prompt_personality += f",{personality_sides[0]}"
-        elif level == 3:
-            personality_sides = self.personality_sides
-            for side in personality_sides:
-                prompt_personality += f",{side}"
-
-        prompt_personality += "。"
-
-        return prompt_personality

src/plugins/heartFC_chat/heartFC_chat.py

@@ -1,7 +1,6 @@
 import asyncio
 import time
 import traceback
-import random  # <-- 添加导入
 from typing import List, Optional, Dict, Any, Deque, Callable, Coroutine
 from collections import deque
 from src.plugins.chat.message import MessageRecv, BaseMessageInfo, MessageThinking, MessageSending
@@ -14,17 +13,20 @@ from src.plugins.models.utils_model import LLMRequest
 from src.config.config import global_config
 from src.plugins.chat.utils_image import image_path_to_base64  # Local import needed after move
 from src.plugins.utils.timer_calculator import Timer  # <--- Import Timer
-from src.plugins.heartFC_chat.heartFC_generator import HeartFCGenerator
 from src.do_tool.tool_use import ToolUser
 from src.plugins.emoji_system.emoji_manager import emoji_manager
 from src.plugins.utils.json_utils import process_llm_tool_calls, extract_tool_call_arguments
 from src.heart_flow.sub_mind import SubMind
 from src.heart_flow.observation import Observation
-from src.plugins.heartFC_chat.heartflow_prompt_builder import global_prompt_manager
+from src.plugins.heartFC_chat.heartflow_prompt_builder import global_prompt_manager, prompt_builder
 import contextlib
 from src.plugins.utils.chat_message_builder import num_new_messages_since
 from src.plugins.heartFC_chat.heartFC_Cycleinfo import CycleInfo
 from .heartFC_sender import HeartFCSender
+from src.plugins.chat.utils import process_llm_response
+from src.plugins.respon_info_catcher.info_catcher import info_catcher_manager
+from src.plugins.moods.moods import MoodManager
+from src.individuality.individuality import Individuality
 
 
 INITIAL_DURATION = 60.0
@@ -181,12 +183,18 @@ class HeartFChatting:
         self.action_manager = ActionManager()
 
         # 初始化状态控制
-        self._initialized = False  # 是否已初始化标志
+        self._initialized = False
-        self._processing_lock = asyncio.Lock()  # 处理锁(确保单次Plan-Replier-Sender周期)
+        self._processing_lock = asyncio.Lock()
 
-        # 依赖注入存储
+        # --- 移除 gpt_instance, 直接初始化 LLM 模型 ---
-        self.gpt_instance = HeartFCGenerator()  # 文本回复生成器
+        # self.gpt_instance = HeartFCGenerator() # <-- 移除
-        self.tool_user = ToolUser()  # 工具使用实例
+        self.model_normal = LLMRequest(  # <-- 新增 LLM 初始化
+            model=global_config.llm_normal,
+            temperature=global_config.llm_normal["temp"],
+            max_tokens=256,
+            request_type="response_heartflow",
+        )
+        self.tool_user = ToolUser()
         self.heart_fc_sender = HeartFCSender()
 
         # LLM规划器配置
@@ -401,7 +409,6 @@ class HeartFChatting:
             with Timer("决策", cycle_timers):
                 planner_result = await self._planner(current_mind, cycle_timers)
 
-            
             # 效果不太好，还没处理replan导致观察时间点改变的问题
 
             # action = planner_result.get("action", "error")
@@ -1058,9 +1065,13 @@ class HeartFChatting:
                 # 如果最近的活动循环不是文本回复，或者没有活动循环
                 cycle_info_block = "\n【近期回复历史】\n(最近没有连续文本回复)\n"
 
+            individuality = Individuality.get_instance()
+            prompt_personality = individuality.get_prompt(x_person=2, level=2)
+
             # 获取提示词模板并填充数据
             prompt = (await global_prompt_manager.get_prompt_async("planner_prompt")).format(
                 bot_name=global_config.BOT_NICKNAME,
+                prompt_personality=prompt_personality,
                 structured_info_block=structured_info_block,
                 chat_content_block=chat_content_block,
                 current_mind_block=current_mind_block,
@@ -1083,27 +1094,66 @@ class HeartFChatting:
         thinking_id: str,
     ) -> Optional[List[str]]:
         """
-        回复器 (Replier): 核心逻辑用于生成回复。
+        回复器 (Replier): 核心逻辑，负责生成回复文本。
+        (已整合原 HeartFCGenerator 的功能)
         """
-        response_set: Optional[List[str]] = None
         try:
-            response_set = await self.gpt_instance.generate_response(
+            # 1. 获取情绪影响因子并调整模型温度
-                structured_info=self.sub_mind.structured_info,
+            arousal_multiplier = MoodManager.get_instance().get_arousal_multiplier()
-                current_mind_info=self.sub_mind.current_mind,
+            current_temp = global_config.llm_normal["temp"] * arousal_multiplier
+            self.model_normal.temperature = current_temp  # 动态调整温度
+
+            # 2. 获取信息捕捉器
+            info_catcher = info_catcher_manager.get_info_catcher(thinking_id)
+
+            # 3. 构建 Prompt
+            with Timer("构建Prompt", {}):  # 内部计时器，可选保留
+                prompt = await prompt_builder.build_prompt(
+                    build_mode="focus",
                     reason=reason,
-                message=anchor_message,  # Pass anchor_message positionally (matches 'message' parameter)
+                    current_mind_info=self.sub_mind.current_mind,
-                thinking_id=thinking_id,  # Pass thinking_id positionally
+                    structured_info=self.sub_mind.structured_info,
+                    message_txt="",  # 似乎是固定的空字符串
+                    sender_name="",  # 似乎是固定的空字符串
+                    chat_stream=anchor_message.chat_stream,
                 )
 
-            if not response_set:
+            # 4. 调用 LLM 生成回复
-                logger.warning(f"{self.log_prefix}[Replier-{thinking_id}] LLM生成了一个空回复集。")
+            content = None
+            reasoning_content = None
+            model_name = "unknown_model"
+            try:
+                with Timer("LLM生成", {}):  # 内部计时器，可选保留
+                    content, reasoning_content, model_name = await self.model_normal.generate_response(prompt)
+                logger.info(f"{self.log_prefix}[Replier-{thinking_id}]\\nPrompt:\\n{prompt}\\n生成回复: {content}\\n")
+                # 捕捉 LLM 输出信息
+                info_catcher.catch_after_llm_generated(
+                    prompt=prompt, response=content, reasoning_content=reasoning_content, model_name=model_name
+                )
+
+            except Exception as llm_e:
+                # 精简报错信息
+                logger.error(f"{self.log_prefix}[Replier-{thinking_id}] LLM 生成失败: {llm_e}")
+                return None  # LLM 调用失败则无法生成回复
+
+            # 5. 处理 LLM 响应
+            if not content:
+                logger.warning(f"{self.log_prefix}[Replier-{thinking_id}] LLM 生成了空内容。")
                 return None
 
-            return response_set
+            with Timer("处理响应", {}):  # 内部计时器，可选保留
+                processed_response = process_llm_response(content)
+
+            if not processed_response:
+                logger.warning(f"{self.log_prefix}[Replier-{thinking_id}] 处理后的回复为空。")
+                return None
+
+            return processed_response
 
         except Exception as e:
-            logger.error(f"{self.log_prefix}[Replier-{thinking_id}] Unexpected error in replier_work: {e}")
+            # 更通用的错误处理，精简信息
-            logger.error(traceback.format_exc())
+            logger.error(f"{self.log_prefix}[Replier-{thinking_id}] 回复生成意外失败: {e}")
+            # logger.error(traceback.format_exc()) # 可以取消注释这行以在调试时查看完整堆栈
             return None
 
     # --- Methods moved from HeartFCController start ---

src/plugins/heartFC_chat/heartFC_generator.py

@@ -1,107 +0,0 @@
-from typing import List, Optional
-
-
-from ..models.utils_model import LLMRequest
-from ...config.config import global_config
-from ..chat.message import MessageRecv
-from .heartflow_prompt_builder import prompt_builder
-from ..chat.utils import process_llm_response
-from src.common.logger_manager import get_logger
-from src.plugins.respon_info_catcher.info_catcher import info_catcher_manager
-from ..utils.timer_calculator import Timer
-
-from src.plugins.moods.moods import MoodManager
-
-
-logger = get_logger("llm")
-
-
-class HeartFCGenerator:
-    def __init__(self):
-        self.model_normal = LLMRequest(
-            model=global_config.llm_normal,
-            temperature=global_config.llm_normal["temp"],
-            max_tokens=256,
-            request_type="response_heartflow",
-        )
-
-        self.model_sum = LLMRequest(
-            model=global_config.llm_summary_by_topic, temperature=0.6, max_tokens=2000, request_type="relation"
-        )
-        self.current_model_type = "r1"  # 默认使用 R1
-        self.current_model_name = "unknown model"
-
-    async def generate_response(
-        self,
-        structured_info: str,
-        current_mind_info: str,
-        reason: str,
-        message: MessageRecv,
-        thinking_id: str,
-    ) -> Optional[List[str]]:
-        """根据当前模型类型选择对应的生成函数"""
-
-        arousal_multiplier = MoodManager.get_instance().get_arousal_multiplier()
-
-        current_model = self.model_normal
-        current_model.temperature = global_config.llm_normal["temp"] * arousal_multiplier  # 激活度越高，温度越高
-        model_response = await self._generate_response_with_model(
-            structured_info, current_mind_info, reason, message, current_model, thinking_id
-        )
-
-        if model_response:
-            model_processed_response = await self._process_response(model_response)
-
-            return model_processed_response
-        else:
-            logger.info(f"{self.current_model_type}思考，失败")
-            return None
-
-    async def _generate_response_with_model(
-        self,
-        structured_info: str,
-        current_mind_info: str,
-        reason: str,
-        message: MessageRecv,
-        model: LLMRequest,
-        thinking_id: str,
-    ) -> str:
-        info_catcher = info_catcher_manager.get_info_catcher(thinking_id)
-
-        with Timer() as _build_prompt:
-            prompt = await prompt_builder.build_prompt(
-                build_mode="focus",
-                reason=reason,
-                current_mind_info=current_mind_info,
-                structured_info=structured_info,
-                message_txt="",
-                sender_name="",
-                chat_stream=message.chat_stream,
-            )
-        # logger.info(f"构建prompt时间: {t_build_prompt.human_readable}")
-
-        try:
-            content, reasoning_content, self.current_model_name = await model.generate_response(prompt)
-
-            logger.info(f"\nprompt:{prompt}\n生成回复{content}\n")
-
-            info_catcher.catch_after_llm_generated(
-                prompt=prompt, response=content, reasoning_content=reasoning_content, model_name=self.current_model_name
-            )
-
-        except Exception:
-            logger.exception("生成回复时出错")
-            return None
-
-        return content
-
-    async def _process_response(self, content: str) -> List[str]:
-        """处理响应内容，返回处理后的内容和情感标签"""
-        if not content:
-            return None
-
-        processed_response = process_llm_response(content)
-
-        # print(f"得到了处理后的llm返回{processed_response}")
-
-        return processed_response

src/plugins/heartFC_chat/heartflow_prompt_builder.py

@@ -25,7 +25,7 @@ def init_prompt():
 {chat_target}
 {chat_talking_prompt}
 现在你想要在群里发言或者回复。\n
-你需要扮演一位网名叫{bot_name}的人进行回复，这个人的特点是："{prompt_personality} {prompt_identity}"。
+你需要扮演一位网名叫{bot_name}的人进行回复，这个人的特点是："{prompt_personality}"。
 你正在{chat_target_2},现在请你读读之前的聊天记录，然后给出日常且口语化的回复，平淡一些，你可以参考贴吧，知乎或者微博的回复风格。
 看到以上聊天记录，你刚刚在想：
 
@@ -49,7 +49,7 @@ def init_prompt():
 
     # Planner提示词 - 优化版
     Prompt(
-        """你的名字是{bot_name}，一名大二女生，正在QQ群聊中。需要基于以下信息决定如何参与对话：
+        """你的名字是{bot_name},{prompt_personality}，你现在正在一个群聊中。需要基于以下信息决定如何参与对话：
 {structured_info_block}
 {chat_content_block}
 你的内心想法：
@@ -164,9 +164,7 @@ class PromptBuilder:
 
     async def _build_prompt_focus(self, reason, current_mind_info, structured_info, chat_stream) -> tuple[str, str]:
         individuality = Individuality.get_instance()
-        prompt_personality = individuality.get_prompt(type="personality", x_person=2, level=1)
+        prompt_personality = individuality.get_prompt(x_person=0, level=2)
-        prompt_identity = individuality.get_prompt(type="identity", x_person=2, level=1)
-
         # 日程构建
         # schedule_prompt = f'''你现在正在做的事情是：{bot_schedule.get_current_num_task(num = 1,time_info = False)}'''
 
@@ -214,7 +212,6 @@ class PromptBuilder:
             chat_talking_prompt=chat_talking_prompt,
             bot_name=global_config.BOT_NICKNAME,
             prompt_personality=prompt_personality,
-            prompt_identity=prompt_identity,
             chat_target_2=await global_prompt_manager.get_prompt_async("chat_target_group2")
             if chat_in_group
             else await global_prompt_manager.get_prompt_async("chat_target_private2"),
@@ -230,21 +227,8 @@ class PromptBuilder:
         return prompt
 
     async def _build_prompt_normal(self, chat_stream, message_txt: str, sender_name: str = "某人") -> tuple[str, str]:
-        # 开始构建prompt
-        prompt_personality = "你"
-        # person
         individuality = Individuality.get_instance()
-
+        prompt_personality = individuality.get_prompt(x_person=2, level=2)
-        personality_core = individuality.personality.personality_core
-        prompt_personality += personality_core
-
-        personality_sides = individuality.personality.personality_sides
-        random.shuffle(personality_sides)
-        prompt_personality += f",{personality_sides[0]}"
-
-        identity_detail = individuality.identity.identity_detail
-        random.shuffle(identity_detail)
-        prompt_personality += f",{identity_detail[0]}"
 
         # 关系
         who_chat_in_group = [

src/plugins/memory_system/Hippocampus.py

@@ -14,51 +14,14 @@ from ...common.database import db
 from ...plugins.models.utils_model import LLMRequest
 from src.common.logger_manager import get_logger
 from src.plugins.memory_system.sample_distribution import MemoryBuildScheduler  # 分布生成器
+from ..utils.chat_message_builder import (
+    get_raw_msg_by_timestamp,
+    build_readable_messages,
+)  # 导入 build_readable_messages
+from ..chat.utils import translate_timestamp_to_human_readable
 from .memory_config import MemoryConfig
 
 
-def get_closest_chat_from_db(length: int, timestamp: str):
-    # print(f"获取最接近指定时间戳的聊天记录，长度: {length}, 时间戳: {timestamp}")
-    # print(f"当前时间: {timestamp},转换后时间: {time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(timestamp))}")
-    chat_records = []
-    closest_record = db.messages.find_one({"time": {"$lte": timestamp}}, sort=[("time", -1)])
-    # print(f"最接近的记录: {closest_record}")
-    if closest_record:
-        closest_time = closest_record["time"]
-        chat_id = closest_record["chat_id"]  # 获取chat_id
-        # 获取该时间戳之后的length条消息，保持相同的chat_id
-        chat_records = list(
-            db.messages.find(
-                {
-                    "time": {"$gt": closest_time},
-                    "chat_id": chat_id,  # 添加chat_id过滤
-                }
-            )
-            .sort("time", 1)
-            .limit(length)
-        )
-        # print(f"获取到的记录: {chat_records}")
-        length = len(chat_records)
-        # print(f"获取到的记录长度: {length}")
-        # 转换记录格式
-        formatted_records = []
-        for record in chat_records:
-            # 兼容行为，前向兼容老数据
-            formatted_records.append(
-                {
-                    "_id": record["_id"],
-                    "time": record["time"],
-                    "chat_id": record["chat_id"],
-                    "detailed_plain_text": record.get("detailed_plain_text", ""),  # 添加文本内容
-                    "memorized_times": record.get("memorized_times", 0),  # 添加记忆次数
-                }
-            )
-
-        return formatted_records
-
-    return []
-
-
 def calculate_information_content(text):
     """计算文本的信息量（熵）"""
     char_count = Counter(text)
@@ -263,16 +226,17 @@ class Hippocampus:
     @staticmethod
     def find_topic_llm(text, topic_num):
         prompt = (
-            f"这是一段文字：{text}。请你从这段话中总结出最多{topic_num}个关键的概念，可以是名词，动词，或者特定人物，帮我列出来，"
+            f"这是一段文字：\n{text}\n\n请你从这段话中总结出最多{topic_num}个关键的概念，可以是名词，动词，或者特定人物，帮我列出来，"
             f"将主题用逗号隔开，并加上<>,例如<主题1>,<主题2>......尽可能精简。只需要列举最多{topic_num}个话题就好，不要有序号，不要告诉我其他内容。"
             f"如果确定找不出主题或者没有明显主题，返回<none>。"
         )
         return prompt
 
     @staticmethod
-    def topic_what(text, topic, time_info):
+    def topic_what(text, topic):
+        # 不再需要 time_info 参数
         prompt = (
-            f'这是一段文字，{time_info}：{text}。我想让你基于这段文字来概括"{topic}"这个概念，帮我总结成一句自然的话，'
+            f'这是一段文字：\n{text}\n\n我想让你基于这段文字来概括"{topic}"这个概念，帮我总结成一句自然的话，'
             f"可以包含时间和人物，以及具体的观点。只输出这句话就好"
         )
         return prompt
@@ -845,9 +809,12 @@ class EntorhinalCortex:
         )
 
         timestamps = sample_scheduler.get_timestamp_array()
-        logger.info(f"回忆往事: {[time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(ts)) for ts in timestamps]}")
+        # 使用 translate_timestamp_to_human_readable 并指定 mode="normal"
+        readable_timestamps = [translate_timestamp_to_human_readable(ts, mode="normal") for ts in timestamps]
+        logger.info(f"回忆往事: {readable_timestamps}")
         chat_samples = []
         for timestamp in timestamps:
+            # 调用修改后的 random_get_msg_snippet
             messages = self.random_get_msg_snippet(
                 timestamp, self.config.build_memory_sample_length, max_memorized_time_per_msg
             )
@@ -862,22 +829,45 @@ class EntorhinalCortex:
 
     @staticmethod
     def random_get_msg_snippet(target_timestamp: float, chat_size: int, max_memorized_time_per_msg: int) -> list:
-        """从数据库中随机获取指定时间戳附近的消息片段"""
+        """从数据库中随机获取指定时间戳附近的消息片段 (使用 chat_message_builder)"""
         try_count = 0
+        time_window_seconds = random.randint(300, 1800)  # 随机时间窗口，5到30分钟
+
         while try_count < 3:
-            messages = get_closest_chat_from_db(length=chat_size, timestamp=target_timestamp)
+            # 定义时间范围：从目标时间戳开始，向后推移 time_window_seconds
-            if messages:
+            timestamp_start = target_timestamp
-                for message in messages:
+            timestamp_end = target_timestamp + time_window_seconds
-                    if message["memorized_times"] >= max_memorized_time_per_msg:
+
-                        messages = None
+            # 使用 chat_message_builder 的函数获取消息
-                        break
+            # limit_mode='earliest' 获取这个时间窗口内最早的 chat_size 条消息
-                if messages:
+            messages = get_raw_msg_by_timestamp(
-                    for message in messages:
+                timestamp_start=timestamp_start, timestamp_end=timestamp_end, limit=chat_size, limit_mode="earliest"
-                        db.messages.update_one(
-                            {"_id": message["_id"]}, {"$set": {"memorized_times": message["memorized_times"] + 1}}
             )
-                    return messages
+
+            if messages:
+                # 检查获取到的所有消息是否都未达到最大记忆次数
+                all_valid = True
+                for message in messages:
+                    if message.get("memorized_times", 0) >= max_memorized_time_per_msg:
+                        all_valid = False
+                        break
+
+                # 如果所有消息都有效
+                if all_valid:
+                    # 更新数据库中的记忆次数
+                    for message in messages:
+                        # 确保在更新前获取最新的 memorized_times，以防万一
+                        current_memorized_times = message.get("memorized_times", 0)
+                        db.messages.update_one(
+                            {"_id": message["_id"]}, {"$set": {"memorized_times": current_memorized_times + 1}}
+                        )
+                    return messages  # 直接返回原始的消息列表
+
+            # 如果获取失败或消息无效，增加尝试次数
             try_count += 1
+            target_timestamp -= 120  # 如果第一次尝试失败，稍微向前调整时间戳再试
+
+        # 三次尝试都失败，返回 None
         return None
 
     async def sync_memory_to_db(self):
@@ -1113,86 +1103,70 @@ class ParahippocampalGyrus:
         """压缩和总结消息内容，生成记忆主题和摘要。
 
         Args:
-            messages (list): 消息列表，每个消息是一个字典，包含以下字段：
+            messages (list): 消息列表，每个消息是一个字典，包含数据库消息结构。
-                - time: float, 消息的时间戳
-                - detailed_plain_text: str, 消息的详细文本内容
             compress_rate (float, optional): 压缩率，用于控制生成的主题数量。默认为0.1。
 
         Returns:
             tuple: (compressed_memory, similar_topics_dict)
                 - compressed_memory: set, 压缩后的记忆集合，每个元素是一个元组 (topic, summary)
-                    - topic: str, 记忆主题
+                - similar_topics_dict: dict, 相似主题字典
-                    - summary: str, 主题的摘要描述
-                - similar_topics_dict: dict, 相似主题字典，key为主题，value为相似主题列表
-                    每个相似主题是一个元组 (similar_topic, similarity)
-                    - similar_topic: str, 相似的主题
-                    - similarity: float, 相似度分数（0-1之间）
 
         Process:
-            1. 合并消息文本并生成时间信息
+            1. 使用 build_readable_messages 生成包含时间、人物信息的格式化文本。
-            2. 使用LLM提取关键主题
+            2. 使用LLM提取关键主题。
-            3. 过滤掉包含禁用关键词的主题
+            3. 过滤掉包含禁用关键词的主题。
-            4. 为每个主题生成摘要
+            4. 为每个主题生成摘要。
-            5. 查找与现有记忆中的相似主题
+            5. 查找与现有记忆中的相似主题。
         """
         if not messages:
             return set(), {}
 
-        # 合并消息文本，同时保留时间信息
+        # 1. 使用 build_readable_messages 生成格式化文本
-        input_text = ""
+        # build_readable_messages 只返回一个字符串，不需要解包
-        time_info = ""
+        input_text = await build_readable_messages(
-        # 计算最早和最晚时间
+            messages,
-        earliest_time = min(msg["time"] for msg in messages)
+            merge_messages=True,  # 合并连续消息
-        latest_time = max(msg["time"] for msg in messages)
+            timestamp_mode="normal",  # 使用 'YYYY-MM-DD HH:MM:SS' 格式
+            replace_bot_name=False,  # 保留原始用户名
+        )
 
-        earliest_dt = datetime.datetime.fromtimestamp(earliest_time)
+        # 如果生成的可读文本为空（例如所有消息都无效），则直接返回
-        latest_dt = datetime.datetime.fromtimestamp(latest_time)
+        if not input_text:
+            logger.warning("无法从提供的消息生成可读文本，跳过记忆压缩。")
+            return set(), {}
 
-        # 如果是同一年
+        logger.debug(f"用于压缩的格式化文本:\n{input_text}")
-        if earliest_dt.year == latest_dt.year:
-            earliest_str = earliest_dt.strftime("%m-%d %H:%M:%S")
-            latest_str = latest_dt.strftime("%m-%d %H:%M:%S")
-            time_info += f"是在{earliest_dt.year}年，{earliest_str} 到 {latest_str} 的对话:\n"
-        else:
-            earliest_str = earliest_dt.strftime("%Y-%m-%d %H:%M:%S")
-            latest_str = latest_dt.strftime("%Y-%m-%d %H:%M:%S")
-            time_info += f"是从 {earliest_str} 到 {latest_str} 的对话:\n"
-
-        for msg in messages:
-            input_text += f"{msg['detailed_plain_text']}\n"
-
-        logger.debug(input_text)
 
+        # 2. 使用LLM提取关键主题
         topic_num = self.hippocampus.calculate_topic_num(input_text, compress_rate)
         topics_response = await self.hippocampus.llm_topic_judge.generate_response(
             self.hippocampus.find_topic_llm(input_text, topic_num)
         )
 
-        # 使用正则表达式提取<>中的内容
+        # 提取<>中的内容
         topics = re.findall(r"<([^>]+)>", topics_response[0])
 
-        # 如果没有找到<>包裹的内容，返回['none']
         if not topics:
             topics = ["none"]
         else:
-            # 处理提取出的话题
             topics = [
                 topic.strip()
                 for topic in ",".join(topics).replace("，", ",").replace("、", ",").replace(" ", ",").split(",")
                 if topic.strip()
             ]
 
-        # 过滤掉包含禁用关键词的topic
+        # 3. 过滤掉包含禁用关键词的topic
         filtered_topics = [
             topic for topic in topics if not any(keyword in topic for keyword in self.config.memory_ban_words)
         ]
 
         logger.debug(f"过滤后话题: {filtered_topics}")
 
-        # 创建所有话题的请求任务
+        # 4. 创建所有话题的摘要生成任务
         tasks = []
         for topic in filtered_topics:
-            topic_what_prompt = self.hippocampus.topic_what(input_text, topic, time_info)
+            # 调用修改后的 topic_what，不再需要 time_info
+            topic_what_prompt = self.hippocampus.topic_what(input_text, topic)
             try:
                 task = self.hippocampus.llm_summary_by_topic.generate_response_async(topic_what_prompt)
                 tasks.append((topic.strip(), task))

src/plugins/models/utils_model.py

@@ -750,7 +750,6 @@ class LLMRequest:
             "tools": tools,
         }
 
-        
         response = await self._execute_request(endpoint="/chat/completions", payload=data, prompt=prompt)
         logger.debug(f"向模型 {self.model_name} 发送工具调用请求，包含 {len(tools)} 个工具，返回结果: {response}")
         # 检查响应是否包含工具调用

src/plugins/person_info/person_info.py

@@ -180,10 +180,10 @@ class PersonInfoManager:
         existing_names = ""
         while current_try < max_retries:
             individuality = Individuality.get_instance()
-            prompt_personality = individuality.get_prompt(type="personality", x_person=2, level=1)
+            prompt_personality = individuality.get_prompt(x_person=2, level=1)
             bot_name = individuality.personality.bot_nickname
 
-            qv_name_prompt = f"你是{bot_name}，你{prompt_personality}"
+            qv_name_prompt = f"你是{bot_name}，{prompt_personality}"
             qv_name_prompt += f"现在你想给一个用户取一个昵称，用户是的qq昵称是{user_nickname}，"
             qv_name_prompt += f"用户的qq群昵称名是{user_cardname}，"
             if user_avatar:

src/plugins/utils/json_utils.py

@@ -1,6 +1,6 @@
 import json
 import logging
-from typing import Any, Dict, TypeVar, List, Union, Callable, Tuple
+from typing import Any, Dict, TypeVar, List, Union, Tuple
 
 # 定义类型变量用于泛型类型提示
 T = TypeVar("T")
@@ -70,7 +70,6 @@ def extract_tool_call_arguments(tool_call: Dict[str, Any], default_value: Dict[s
         return default_result
 
 
-
 def safe_json_dumps(obj: Any, default_value: str = "{}", ensure_ascii: bool = False, pretty: bool = False) -> str:
     """
     安全地将Python对象序列化为JSON字符串
@@ -95,8 +94,6 @@ def safe_json_dumps(obj: Any, default_value: str = "{}", ensure_ascii: bool = Fa
         return default_value
 
 
-
-
 def normalize_llm_response(response: Any, log_prefix: str = "") -> Tuple[bool, List[Any], str]:
     """
     标准化LLM响应格式，将各种格式（如元组）转换为统一的列表格式
@@ -140,7 +137,9 @@ def normalize_llm_response(response: Any, log_prefix: str = "") -> Tuple[bool, L
     return True, response, ""
 
 
-def process_llm_tool_calls(tool_calls: List[Dict[str, Any]], log_prefix: str = "") -> Tuple[bool, List[Dict[str, Any]], str]:
+def process_llm_tool_calls(
+    tool_calls: List[Dict[str, Any]], log_prefix: str = ""
+) -> Tuple[bool, List[Dict[str, Any]], str]:
     """
     处理并验证LLM响应中的工具调用列表
 
@@ -165,7 +164,9 @@ def process_llm_tool_calls(tool_calls: List[Dict[str, Any]], log_prefix: str = "
 
         # 检查基本结构
         if tool_call.get("type") != "function":
-            logger.warning(f"{log_prefix}工具调用[{i}]不是function类型: type={tool_call.get('type', '未定义')}, 内容: {tool_call}")
+            logger.warning(
+                f"{log_prefix}工具调用[{i}]不是function类型: type={tool_call.get('type', '未定义')}, 内容: {tool_call}"
+            )
             continue
 
         if "function" not in tool_call or not isinstance(tool_call.get("function"), dict):
@@ -176,7 +177,9 @@ def process_llm_tool_calls(tool_calls: List[Dict[str, Any]], log_prefix: str = "
         if "name" not in func_details or not isinstance(func_details.get("name"), str):
             logger.warning(f"{log_prefix}工具调用[{i}]的'function'字段缺少'name'或类型不正确: {func_details}")
             continue
-        if "arguments" not in func_details or not isinstance(func_details.get("arguments"), str): # 参数是字符串形式的JSON
+        if "arguments" not in func_details or not isinstance(
+            func_details.get("arguments"), str
+        ):  # 参数是字符串形式的JSON
             logger.warning(f"{log_prefix}工具调用[{i}]的'function'字段缺少'arguments'或类型不正确: {func_details}")
             continue
 
@@ -185,7 +188,9 @@ def process_llm_tool_calls(tool_calls: List[Dict[str, Any]], log_prefix: str = "
         try:
             json.loads(args_str)  # 尝试解析，但不存储结果
         except json.JSONDecodeError as e:
-            logger.warning(f"{log_prefix}工具调用[{i}]的'arguments'不是有效的JSON字符串: {e}, 内容: {args_str[:100]}...")
+            logger.warning(
+                f"{log_prefix}工具调用[{i}]的'arguments'不是有效的JSON字符串: {e}, 内容: {args_str[:100]}..."
+            )
             continue
         except Exception as e:
             logger.warning(f"{log_prefix}解析工具调用[{i}]的'arguments'时发生意外错误: {e}, 内容: {args_str[:100]}...")

template/bot_config_template.toml

@@ -48,12 +48,10 @@ personality_sides = [
 identity_detail = [
     "身份特点",
     "身份特点",
-]# 条数任意，不能为0, 该选项还在调试中，可能未完全生效
+]# 条数任意，不能为0, 该选项还在调试中
 #外貌特征
-height = 170 # 身高 单位厘米 该选项还在调试中，暂时未生效
+age = 20 # 年龄 单位岁 
-weight = 50 # 体重 单位千克 该选项还在调试中，暂时未生效
+gender = "男" # 性别 
-age = 20 # 年龄 单位岁 该选项还在调试中，暂时未生效
-gender = "男" # 性别 该选项还在调试中，暂时未生效
 appearance = "用几句话描述外貌特征" # 外貌特征 该选项还在调试中，暂时未生效
 
 [schedule]