diff --git a/AlgoriAgent/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/__pycache__/__init__.cpython-310.pyc
index 6000a78..65ba710 100644
Binary files a/AlgoriAgent/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/projects/algoriAgent/__pycache__/agent_manager.cpython-310.pyc b/AlgoriAgent/projects/algoriAgent/__pycache__/agent_manager.cpython-310.pyc
index ac596df..f76d2bd 100644
Binary files a/AlgoriAgent/projects/algoriAgent/__pycache__/agent_manager.cpython-310.pyc and b/AlgoriAgent/projects/algoriAgent/__pycache__/agent_manager.cpython-310.pyc differ
diff --git a/AlgoriAgent/projects/algoriAgent/agent/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/projects/algoriAgent/agent/__pycache__/__init__.cpython-310.pyc
index 3415652..b7372b7 100644
Binary files a/AlgoriAgent/projects/algoriAgent/agent/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/projects/algoriAgent/agent/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/projects/algoriAgent/agent/__pycache__/algori_agent.cpython-310.pyc b/AlgoriAgent/projects/algoriAgent/agent/__pycache__/algori_agent.cpython-310.pyc
index a5e1a54..145ae3a 100644
Binary files a/AlgoriAgent/projects/algoriAgent/agent/__pycache__/algori_agent.cpython-310.pyc and b/AlgoriAgent/projects/algoriAgent/agent/__pycache__/algori_agent.cpython-310.pyc differ
diff --git a/AlgoriAgent/projects/algoriAgent/agent/__pycache__/flex_service_toolkit.cpython-310.pyc b/AlgoriAgent/projects/algoriAgent/agent/__pycache__/flex_service_toolkit.cpython-310.pyc
index 0104194..d9d8f26 100644
Binary files a/AlgoriAgent/projects/algoriAgent/agent/__pycache__/flex_service_toolkit.cpython-310.pyc and b/AlgoriAgent/projects/algoriAgent/agent/__pycache__/flex_service_toolkit.cpython-310.pyc differ
diff --git a/AlgoriAgent/projects/algoriAgent/agent/algori_agent.py b/AlgoriAgent/projects/algoriAgent/agent/algori_agent.py
index 2575519..4a0f3ea 100644
--- a/AlgoriAgent/projects/algoriAgent/agent/algori_agent.py
+++ b/AlgoriAgent/projects/algoriAgent/agent/algori_agent.py
@@ -1,476 +1,576 @@
-# -*- coding: utf-8 -*-
-"""An agent class that implements the ReAct algorithm. The agent will reason
-and act iteratively to solve problems. More details can be found in the paper
-https://arxiv.org/abs/2210.03629.
-"""
-from typing import Any, Callable, Optional, Union, Sequence
-
-from loguru import logger
-from agent.flex_service_toolkit import FlexServiceToolkit
-from agentscope.exception import ResponseParsingError, FunctionCallError
-from agentscope.agents import AgentBase
-from agentscope.memory.temporary_memory import TemporaryMemory
-from agentscope.message import Msg
-from agentscope.parsers import MarkdownJsonDictParser
-from agentscope.service.service_toolkit import ServiceFunction
-from agentscope.service import (
-    ServiceToolkit,
-    ServiceResponse,
-    ServiceExecStatus,
-)
-import json
-
-
-# 这个INSTRUCTION_PROMPT 交给免费Agent进行使用，不包含评测与得分标准，但是会在没有自己注释的大段代码或复制粘贴代码是，通知学生进入一次理解性测试。
-INSTRUCTION_PROMPT = """## Your Role:
-You are a powerful AI agent to teach student computer algorithm, you are to company a student to learn algorithm or solve problems.
-
-## What You Should Do:
-0. You MUST follow the teaching instructions made by human teacher with several chapters, ONLY focus on one chapter at a time.
-1. Student may ask you some questions, DONT answer the question directly or completely, instead, based on context (like student's code) to answer only one step forward.
-2. Each chapter in teaching instruction will specifically guide what to teach to student, follow the instruction.
-3. System will gather student's infomation sometimes, you neet to analyze student's learning statement from student's history codes, call the necessary function we he is stuck.
-
-## Note:
-1. Fully understand the tool functions and their arguments before using them.
-2. Make sure the types and values of the arguments you provided to the tool functions are correct.
-3. Don't take things for granted. For example, where you are, what's the time now, etc. You can try to use the tool functions to get information.
-4. If the function execution fails, you should analyze the error and try to solve it.
-5. Remember to decide whether the student's question or idea is regarding about current chapter, if not, you should not awnser and ask the student to focus on the chapter.
-
-## Resources:
-1. The tool functions you can use.
-2. The context, including the student's code, and teaching instructions.
-"""  # noqa
-CHAPTER_FINISH = 0
-CHAPTER_FOCUS = 1
-CHAPTER_LATTER = 2
-CHAPTER_FAILED = 3
-
-class  Chapter():
-    def __init__(self, No = 1, focus = 2,title="", chapter = "", chapter_prompt = "", score_prompt = "", append_tools : Callable[..., Any] = None) -> None:
-        self.No = No
-        self.focus = focus
-        self.chapter = chapter
-        self.append_tools = append_tools
-        self.memory = None
-    def GetChapter(self):
-        if self.focus == CHAPTER_FINISH: f = "(Finished Chapter)"
-        elif self.focus == CHAPTER_FOCUS: f = "(Your Now Chapter)"
-        elif self.focus == CHAPTER_LATTER: f = "(Ignore for Latter Chapter)"
-        elif self.focus == CHAPTER_FAILED: f = "(Failed Chapter)"
-        return f"{f} Chapter {self.No}: {self.chapter}"
-    
-    def Finished(self, successful: bool = True):
-        if successful: self.focus = 0
-        else: self.focus = 3
-    
-    def Focus(self):
-        self.focus = 1
-
-def record(content: str, name: str):
-    """Call it to save some important info for other chapters. Like file path, function call example.
-
-    Args:
-        content (`str`): The content to be recorded.
-        name (`str`): The name of the agent to record.
-    Returns:
-        None
-    """
-    ChapterChainAgent.Instance[name].memory.add(Msg(
-        "assistant", content, role="assistant"
-    ))
-    ChapterChainAgent.Instance[name].chapter_memory.add(Msg(
-        "assistant", "I recorded globally: "+content, role="assistant"
-    ))
-    status = ServiceExecStatus.SUCCESS
-    return ServiceResponse(status, None)
-
-def next_chapter(name: str):
-    """Call it to go to next chapter.
-
-    Args:
-        name (`str`): The name of the agent to go to next chapter.
-        
-    Returns:
-        None
-    """
-    try:
-        ChapterChainAgent.Instance[name].next_chapter()
-    except:
-        status = ServiceExecStatus.ERROR
-        return ServiceResponse(status, "Failed to go to next chapter. name '"+name+"' not found in ChapterChainAgent.Instance")
-
-    status = ServiceExecStatus.SUCCESS
-    return ServiceResponse(status, None)
-
-
-
-class ChapterChainAgent(AgentBase):
-    """An agent class to preform chapter chain algorithm. 
-    Project will be split to some chapters chained. 
-    Each chapter will haave a main objective, having its own tools, and holding its own context.
-    """
-    Instance = {}
-    def __init__(
-        self,
-        name: str,
-        model_config_name: str,
-        service_toolkit: FlexServiceToolkit = None,
-        sys_prompt: str = "You're a helpful assistant.",
-        max_iters: int = 10,
-        verbose: bool = True,
-        chapter_chain: Sequence[Chapter] = None,
-        stop_when_one_CHAPTER_failed: bool = True,
-        **kwargs: Any,
-    ) -> None:
-        """
-        Args:
-            name (`str`):
-                The name of the agent.
-            sys_prompt (`str`):
-                The system prompt of the agent.
-            model_config_name (`str`):
-                The name of the model config, which is used to load model from
-                configuration.
-            service_toolkit (`ServiceToolkit`):
-                A `ServiceToolkit` object that contains the tool functions.
-            max_iters (`int`, defaults to `10`):
-                The maximum number of iterations of each chapter.
-            verbose (`bool`, defaults to `True`):
-                Whether to print the detailed information during reasoning and
-                acting steps. If `False`, only the content in speak field will
-                be print out.
-            chapter_chain (`Sequence[str]`):
-                The chapter chain to be performed. If None, the agent will add a 
-                init chapter to construct the chapter chain.
-        """
-        super().__init__(
-            name=name,
-            sys_prompt=sys_prompt,
-            model_config_name=model_config_name,
-        )
-
-        # TODO: To compatible with the old version, which will be deprecated
-        #  soon
-        if "tools" in kwargs:
-            logger.warning(
-                "The argument `tools` will be deprecated soon. "
-                "Please use `service_toolkit` instead. Example refers to "
-                "https://github.com/modelscope/agentscope/blob/main/"
-                "examples/conversation_with_react_agent/code/"
-                "conversation_with_react_agent.py",
-            )
-
-            service_funcs = {}
-            for func, json_schema in kwargs["tools"]:
-                name = json_schema["function"]["name"]
-                service_funcs[name] = ServiceFunction(
-                    name=name,
-                    original_func=func,
-                    processed_func=func,
-                    json_schema=json_schema,
-                )
-
-            if service_toolkit is None:
-                service_toolkit = FlexServiceToolkit()
-                service_toolkit.service_funcs = service_funcs
-            else:
-                service_toolkit.service_funcs.update(service_funcs)
-
-        elif service_toolkit is None:
-            raise ValueError(
-                "The argument `service_toolkit` is required to initialize "
-                "the ReActAgent.",
-            )
-        
-        service_toolkit.add(record,name=name)
-        service_toolkit.add(next_chapter,name=name)
-
-        self.service_toolkit = service_toolkit
-        self.service_toolkit.agent = self
-        self.verbose = verbose
-        self.max_iters = max_iters
-
-        if not sys_prompt.endswith("\n"):
-            sys_prompt = sys_prompt + "\n"
-
-        self.sys_prompt = "\n".join(
-            [
-                # The brief intro of the role and target
-                sys_prompt.format(name=self.name),
-                # The detailed instruction prompt for the agent
-                INSTRUCTION_PROMPT,
-            ],
-        )
-
-        
-        # This memory is used to store all useful memory with the chapter chain switching
-        #self.memory_on_chain = TemporaryMemory()
-        self.memory.add(Msg("system", self.sys_prompt, role="system"))
-        self.chapter_memory = TemporaryMemory()
-
-        # Initialize a parser object to formulate the response from the model
-        self.parser = MarkdownJsonDictParser(
-            content_hint={
-                "thought": "what you thought",
-                "speak": "what you speak",
-                "function": service_toolkit.tools_calling_format,
-            },
-            required_keys=["thought", "speak", "function"],
-            # Only print the speak field when verbose is False
-            keys_to_content=True if self.verbose else "speak",
-        )
-
-
-        self.chapter_chain = chapter_chain
-
-        self.chapter_chain_now = 0
-        self.finish_now_chapter = False
-
-        self.stop_when_one_CHAPTER_failed = stop_when_one_CHAPTER_failed
-
-
-        ChapterChainAgent.Instance[name] = self
-
-    def chapter_chain_to_memory(self):
-        mem = "##The Chapter Chain##\n"
-        for chapter in self.chapter_chain:
-            mem+=chapter.GetChapter()+"\n"
-        mem+="\n"
-        mem+= "Remember your now chapter is "+ self.chapter_chain[self.chapter_chain_now].GetChapter()+"\n"
-        return Msg("system", mem, role="system")
-
-    def _has_unfinished_chapter(self):
-        for chapter in self.chapter_chain:
-            if chapter.focus == CHAPTER_LATTER or chapter.focus == CHAPTER_FOCUS:
-                return True
-        return False
-    def next_chapter(self, successful: bool = True):
-        if self.chapter_chain[self.chapter_chain_now].append_tools is not None:
-            for callable_tools in self.chapter_chain[self.chapter_chain_now].append_tools:
-                self.service_toolkit.Remove(callable_tools)
-        
-        self.chapter_chain[self.chapter_chain_now].Finished(successful)
-
-        for i in range(len(self.chapter_chain)):
-            if self.chapter_chain[i].focus == 2:
-                self.chapter_chain[i].Focus()
-                if self.chapter_chain[i].append_tools is not None:
-                    for callable_tools in self.chapter_chain[i].append_tools:
-                        self.service_toolkit.add(callable_tools)
-                self.chapter_chain_now = i
-                return 
-                
-    def add_chapters_after(self, CHAPTER_No, chapters: Sequence[Chapter]):
-        for i in range(len(self.chapter_chain)):
-            if self.chapter_chain[i].No == CHAPTER_No:
-                for j in range(len(chapters)):
-                    CHAPTER_to_insert = Chapter(No = CHAPTER_No + j + 1, chapter = chapters[j], focus = 2, append_tools=
-                                          self.chapter_chain[self.chapter_chain_now].append_tools)
-                    self.chapter_chain.insert(i+1+j, CHAPTER_to_insert)
-                for j in range(i+len(chapters), len(self.chapter_chain)):
-                    self.chapter_chain[j].No += len(chapters)
-                break
-    def reply(self, x: Optional[Union[Msg, Sequence[Msg]]] = None, user_backboard:str = "") -> Msg:
-            if self.chapter_memory.size() == 0:
-
-                self.chapter_chain[self.chapter_chain_now].memory = self.chapter_memory
-                for msg in self.memory.get_memory():
-                    self.chapter_memory.add(msg)
-                self.chapter_memory.add(self.chapter_chain_to_memory())
-            
-            self.chapter_memory.add(x)
-
-            
-            for _ in range(1):
-                # Step 1: Thought
-                if self.verbose:
-                    self.speak(f" ITER {_+1}, STEP 1: REASONING ".center(70, "#"))
-
-                # Prepare hint to remind model what the response format is
-                # Won't be recorded in memory to save tokens
-                backboard_hint_msg = Msg(
-                    "system",
-                    user_backboard,
-                    role="system",
-                    echo=self.verbose,
-                )
-                tools_hint_msg = Msg(
-                    "system",
-                    self.service_toolkit.tools_instruction,
-                    role="system",
-                    echo=self.verbose,
-                )
-                hint_msg = Msg(
-                    "system",
-                    self.parser.format_instruction+"\n ####",
-                    role="system",
-                    echo=self.verbose,
-                )
-
-                prompt = self.model.format(self.chapter_memory.get_memory(),
-                    backboard_hint_msg, 
-                    # The instruction prompt for tools
-                    tools_hint_msg,
-                    hint_msg)
-                print("*********************************************************************")
-                print(prompt)
-                if self.verbose:
-                    self.speak(f"Prompt".center(70, "#"))
-                    self.speak(str(self.chapter_memory.get_memory()[-1].content))
-                    # print(prompt)
-                
-                try:
-                    res = self.model(
-                        prompt,
-                        parse_func=self.parser.parse,
-                        max_retries=1,
-                    )
-
-                    
-                    if self.verbose:
-                        self.speak(f"Result Parsed".center(70, "#"))
-                        self.speak(str(res.parsed))
-                        print(res.parsed)
-                        
-                    self.chapter_memory.add(
-                        Msg(
-                            self.name,
-                            self.parser.to_memory(res.parsed),
-                            "assistant",
-                        ),
-                    )
-
-                    # Print out the response
-                    msg_returned = Msg(
-                        self.name,
-                        self.parser.to_content(res.parsed),
-                        "assistant",
-                    )
-                    if self.verbose:
-                        self.speak(f"Speak".center(70, "#"))
-                        print(msg_returned)
-                    self.speak(msg_returned)
-                    return msg_returned
-                    
-                    if self.verbose:
-                        self.speak(f"Function".center(70, "#"))
-                        self.speak(str(res.parsed["function"]))
-                        print(res.parsed["function"])
-
-                    arg_function = res.parsed["function"]
-                    if (
-                        isinstance(arg_function, str)
-                        and arg_function in ["[]", ""]
-                        or isinstance(arg_function, list)
-                        and len(arg_function) == 0
-                    ):
-                        # Only the speak field is exposed to users or other agents
-                        continue
-
-                # Only catch the response parsing error and expose runtime
-                # errors to developers for debugging
-                except ResponseParsingError as e:
-                    # Print out raw response from models for developers to debug
-                    response_msg = Msg(self.name, e.raw_response, "assistant")
-                    self.speak(response_msg)
-
-                    # Re-correct by model itself
-                    error_msg = Msg("system", str(e), "system")
-                    self.speak(error_msg)
-
-                    self.chapter_memory.add([response_msg, error_msg])
-
-                    # Skip acting step to re-correct the response
-                    continue
-                
-                # 2. Acting
-                if self.verbose:
-                    self.speak(f" ITER {_+1}, STEP 2: ACTING ".center(70, "#"))
-
-                # Parse, check and execute the tool functions in service toolkit
-                try:
-                    
-                    
-                        
-                    execute_results = self.service_toolkit.parse_and_call_func(
-                        res.parsed["function"],
-                    )
-                    # 在这里补充任务结束行动
-                    if self.finish_now_chapter:
-                        self.finish_now_chapter = False
-                        hint_msg = Msg(
-                            "system",
-                            "You have finished your own chapter."
-                            "Now generate a reply by summarizing the current "
-                            "situation. Remember this summary will be recorded in global memory. Make sure keep important information.",
-                            role="system",
-                            echo=self.verbose,
-                        )
-
-                        # Generate a reply by summarizing the current situation
-                        prompt = self.model.format(self.chapter_memory.get_memory(), hint_msg)
-                        res = self.model(prompt)
-                        res_msg = Msg(self.name, res.text, "assistant")
-                        self.memory.add(res_msg)
-                        self.speak(res_msg)
-                        self.next_chapter(successful=True)
-                        self.success_own_chapter = True
-                        break
-
-                    # Note: Observing the execution results and generate response
-                    # are finished in the next reasoning step. We just put the
-                    # execution results into memory, and wait for the next loop
-                    # to generate response.
-
-                    # Record execution results into memory as system message
-                    msg_res = Msg("system", execute_results, "system")
-                    self.speak(msg_res)
-                    self.chapter_memory.add(msg_res)
-
-                except FunctionCallError as e:
-                    # Catch the function calling error that can be handled by
-                    # the model
-                    error_msg = Msg("system", str(e), "system")
-                    self.speak(error_msg)
-                    self.chapter_memory.add(error_msg)
-                    continue
-
-                except Exception as e:
-                    # Catch the function calling error that can not be handled
-                    error_msg = Msg("system", str(e), "system")
-                    self.speak(error_msg)
-                    self.chapter_memory.add(error_msg)
-                    continue
-
-
-
-        
-    def GetProcess(self):
-        process = []
-        t = {}
-        t['topic'] = "Global Memory"
-        mems = []
-        for msg in self.memory.get_memory():
-            mems.append(msg.serialize())
-
-        t['memory'] = "$MEMORYMEMORY$".join(mems)
-        t['finish'] = "True"
-        ts = f"{t['topic']}$TOPICTOPIC${t['memory']}$TOPICTOPIC${t['finish']}"
-        process.append(ts)
-        for chapter in self.chapter_chain:
-            t = {}
-            t['topic'] = f'Chapter {chapter.No}'
-            mems = []
-            if chapter.memory is not None:
-                for msg in self.memory.get_memory():
-                    mems.append(msg.serialize())
-            t['memory'] =   "$MEMORYMEMORY$".join(mems)
-            t['finish'] = "True" if chapter.focus == CHAPTER_FINISH else "False"
-            ts = f"{t['topic']}$TOPICTOPIC${t['memory']}$TOPICTOPIC${t['finish']}"
-            process.append(ts)
-        
-        returnString = ""
-        returnString = "$TASKTASK$".join(process)
-        return returnString
+# -*- coding: utf-8 -*-
+"""An agent class that implements the ReAct algorithm. The agent will reason
+and act iteratively to solve problems. More details can be found in the paper
+https://arxiv.org/abs/2210.03629.
+"""
+from typing import Any, Callable, Optional, Union, Sequence
+
+from loguru import logger
+from agent.flex_service_toolkit import FlexServiceToolkit
+from agentscope.exception import ResponseParsingError, FunctionCallError
+from agentscope.agents import AgentBase
+from agentscope.memory.temporary_memory import TemporaryMemory
+from agentscope.message import Msg
+from agentscope.parsers import MarkdownJsonDictParser
+from agentscope.service.service_toolkit import ServiceFunction
+from agentscope.service import (
+    ServiceToolkit,
+    ServiceResponse,
+    ServiceExecStatus,
+)
+import json
+from AlgoriAgent.projects.algoriAgent.agent.score_agent import ScoreAgent
+from AlgoriAgent.projects.algoriAgent.agent.total_score_agent import TotalScoreAgent
+
+import re
+
+
+
+
+# 这个INSTRUCTION_PROMPT 交给免费Agent进行使用，不包含评测与得分标准，但是会在没有自己注释的大段代码或复制粘贴代码是，通知学生进入一次理解性测试。
+INSTRUCTION_PROMPT_EN = """## Your Role:
+You need to help the student learn algorithms, verify the learning process, ensure that the student understands the current chapter's content, and then guide the student to the next chapter.
+
+## What You Should Do:
+0. You MUST follow the teaching instructions made by human teacher with several chapters, ONLY focus on one chapter at a time.
+1. Student may ask you some questions, DONT answer the question directly or completely, instead, based on context (like student's code) to answer only one step forward.
+2. Each chapter in teaching instruction will specifically guide what to teach to student, follow the instruction.
+3. System will gather student's infomation sometimes, you neet to analyze student's learning statement from student's history codes, call the necessary function we he is stuck.
+
+## Note:
+1. Fully understand the tool functions and their arguments before using them.
+2. Make sure the types and values of the arguments you provided to the tool functions are correct.
+3. Don't take things for granted. For example, where you are, what's the time now, etc. You can try to use the tool functions to get information.
+4. If the function execution fails, you should analyze the error and try to solve it.
+5. Remember to decide whether the student's question or idea is regarding about current chapter, if not, you should not awnser and ask the student to focus on the chapter.
+
+## Resources:
+1. The tool functions you can use.
+2. The context, including the student's code, and teaching instructions.
+"""  # noqa
+INSTRUCTION_PROMPT_CN = """## 你的角色：
+你需要帮助学生学习算法，检验学习过程，确保学生理解当前章节的内容后，带领学生进入下一章
+
+## 你需要做的：
+0. 你必须根据老师提供的教学大纲，按照章节顺序，依次学习。
+1. 学生可能会提出一些问题，你无需回答问题，而是根据上下文（比如学生自己的代码）来分析，并给出一步的解决方案。
+2. 教学大纲中会指明每个章节的学习目标，你按照目标学习。
+3. 系统会occasionally gather学生信息，你需要分析学生学习记录，调用必要的函数。
+
+## 注意事项：
+1. 确保使用工具函数时，提供的参数类型和值是正确的。
+2. 不要太相信自己，例如，当前位置，当前时间等，你可以尝试使用工具函数获取信息。
+3. 如果函数执行失败，你需要分析错误并尝试解决。
+4. 确保学生提出的问题或想法是否与当前章节相关，如果不相关，你应答并提醒学生关注当前章节。
+
+## 资源：
+1. 你可以使用的工具函数。
+2. 上下文，包括学生的代码，和教学大纲。
+"""
+
+
+LANGUAGE_HINT_EN = "Use English to reply"
+LANGUAGE_HINT_CN = "使用中文回复"
+
+
+
+CHAPTER_FINISH = 0
+CHAPTER_FOCUS = 1
+CHAPTER_LATTER = 2
+CHAPTER_FAILED = 3
+
+class  Chapter():
+    
+    def __init__(self, course_id, lesson_id, No = 1, focus = 2,title="", chapter = "", chapter_prompt = "", 
+                 score_prompt = "", append_tools : list[Callable[..., Any], dict] = None) -> None:
+        '''
+        No : 序号
+        focus : 0: Finished, 1: Focus, 2: Latter, 3: Failed
+        title : 子章节标题
+        chapter : 教案子章节内容
+        chapter_prompt : 教案的提示
+        score_prompt : 评测标准
+        append_tools : 追加工具函数
+        '''
+        self.No = No
+        self.focus = focus
+        self.chapter = chapter
+        self.chapter_prompt = chapter_prompt
+        self.score_prompt = score_prompt
+        self.append_tools = append_tools
+        self.memory = None
+        self.title =title
+        self.course_id = course_id
+        self.lesson_id = lesson_id 
+
+
+    def GetChapter(self):
+        if self.focus == CHAPTER_FINISH: f = "(Finished Chapter)"
+        elif self.focus == CHAPTER_FOCUS: f = "(Your Now Chapter)"
+        elif self.focus == CHAPTER_LATTER: f = "(Ignore for Latter Chapter)"
+        elif self.focus == CHAPTER_FAILED: f = "(Failed Chapter)"
+        # des = f"{f} Chapter {self.No}: {self.chapter if self.focus else 'Hidden for further study.'}"
+        des = f"{f} Chapter {self.No}: "
+        if self.focus == CHAPTER_FINISH or self.focus == CHAPTER_FAILED: des += 'Hidden due to finished.'
+        if self.focus == CHAPTER_LATTER: des += 'Hidden for further study.'
+        if self.focus == CHAPTER_FOCUS:
+            des+=self.chapter
+            des+='\n'
+            des+='Here are some concrete instructions:\n'
+            des+=self.chapter_prompt+'\n'
+        
+        return des
+    
+    def Finished(self, successful: bool = True):
+        if successful: self.focus = 0
+        else: self.focus = 3
+    
+    def Focus(self):
+        self.focus = 1
+
+def record(content: str, name: str):
+    """Call it to save some important info for other chapters. Like file path, function call example.
+
+    Args:
+        content (`str`): The content to be recorded.
+        name (`str`): The name of the agent to record.
+    Returns:
+        None
+    """
+    ChapterChainAgent.Instance[name].memory.add(Msg(
+        "assistant", content, role="assistant"
+    ))
+    ChapterChainAgent.Instance[name].chapter_memory.add(Msg(
+        "assistant", "I recorded globally: "+content, role="assistant"
+    ))
+    status = ServiceExecStatus.SUCCESS
+    return ServiceResponse(status, None)
+
+def next_chapter(name: str):
+    """Call it to go to next chapter.
+
+    Args:
+        name (`str`): The name of the agent to go to next chapter.
+        
+    Returns:
+        message (`str`): The result of the operation.
+    """
+    res=""
+    try:
+        print(ChapterChainAgent.Instance)
+        print(name in ChapterChainAgent.Instance)
+        res = ChapterChainAgent.Instance[name].next_chapter()
+    except Exception as e:
+        status = ServiceExecStatus.ERROR
+        return ServiceResponse(status, str(e))
+
+    status = ServiceExecStatus.SUCCESS
+    return ServiceResponse(status, res)
+
+from AlgoriAgent.projects.algoriAgent.tools.file_tools import file_tools
+
+class ChapterChainAgent(AgentBase):
+    """An agent class to preform chapter chain algorithm. 
+    Project will be split to some chapters chained. 
+    Each chapter will haave a main objective, having its own tools, and holding its own context.
+    """
+    Instance = {}
+    def __init__(
+        self,
+        name: str,
+        model_config_name: str,
+        service_toolkit: FlexServiceToolkit = None,
+        sys_prompt: str = "You're a helpful assistant.",
+        max_iters: int = 10,
+        verbose: bool = True,
+        chapter_chain: Sequence[Chapter] = None,
+        stop_when_one_CHAPTER_failed: bool = True,
+        root_path: str = ".",
+        **kwargs: Any,
+    ) -> None:
+        """
+        Args:
+            name (`str`):
+                The name of the agent.
+            sys_prompt (`str`):
+                The system prompt of the agent.
+            model_config_name (`str`):
+                The name of the model config, which is used to load model from
+                configuration.
+            service_toolkit (`ServiceToolkit`):
+                A `ServiceToolkit` object that contains the tool functions.
+            max_iters (`int`, defaults to `10`):
+                The maximum number of iterations of each chapter.
+            verbose (`bool`, defaults to `True`):
+                Whether to print the detailed information during reasoning and
+                acting steps. If `False`, only the content in speak field will
+                be print out.
+            chapter_chain (`Sequence[str]`):
+                The chapter chain to be performed. If None, the agent will add a 
+                init chapter to construct the chapter chain.
+        """
+        super().__init__(
+            name=name,
+            sys_prompt=sys_prompt,
+            model_config_name=model_config_name,
+        )
+        self.model_config_name =  model_config_name
+        # TODO: To compatible with the old version, which will be deprecated
+        #  soon
+        self.name=name
+        if "tools" in kwargs:
+            logger.warning(
+                "The argument `tools` will be deprecated soon. "
+                "Please use `service_toolkit` instead. Example refers to "
+                "https://github.com/modelscope/agentscope/blob/main/"
+                "examples/conversation_with_react_agent/code/"
+                "conversation_with_react_agent.py",
+            )
+
+            service_funcs = {}
+            for func, json_schema in kwargs["tools"]:
+                name = json_schema["function"]["name"]
+                service_funcs[name] = ServiceFunction(
+                    name=name,
+                    original_func=func,
+                    processed_func=func,
+                    json_schema=json_schema,
+                )
+
+            if service_toolkit is None:
+                service_toolkit = FlexServiceToolkit()
+                service_toolkit.service_funcs = service_funcs
+            else:
+                service_toolkit.service_funcs.update(service_funcs)
+
+        elif service_toolkit is None:
+            raise ValueError(
+                "The argument `service_toolkit` is required to initialize "
+                "the ReActAgent.",
+            )
+        
+        service_toolkit.add(record,name=name)
+        service_toolkit.add(next_chapter,name=name)
+        for tool in file_tools:
+            service_toolkit.add(tool, root_path=root_path)
+
+        self.service_toolkit = service_toolkit
+        self.service_toolkit.agent = self
+        self.verbose = verbose
+        self.max_iters = max_iters
+
+        if not sys_prompt.endswith("\n"):
+            sys_prompt = sys_prompt + "\n"
+
+        self.sys_prompt = "\n".join(
+            [
+                # The brief intro of the role and target
+                sys_prompt.format(name=self.name),
+                # The detailed instruction prompt for the agent
+                INSTRUCTION_PROMPT_CN,
+            ],
+        )
+
+        
+        # This memory is used to store all useful memory with the chapter chain switching
+        #self.memory_on_chain = TemporaryMemory()
+        self.memory.add(Msg("system", self.sys_prompt, role="system"))
+        self.chapter_memory = TemporaryMemory()
+
+        # Initialize a parser object to formulate the response from the model
+        self.parser = MarkdownJsonDictParser(
+            content_hint={
+                "thought": "what you thought",
+                "speak": "what you speak",
+                "function": service_toolkit.tools_calling_format,
+            },
+            required_keys=["thought", "speak", "function"],
+            # Only print the speak field when verbose is False
+            keys_to_content=True if self.verbose else "speak",
+        )
+        self.language_hint = LANGUAGE_HINT_CN
+
+        self.chapter_chain = chapter_chain
+
+        self.chapter_chain_now = 0
+        self.finish_now_chapter = False
+
+        self.stop_when_one_CHAPTER_failed = stop_when_one_CHAPTER_failed
+
+
+        ChapterChainAgent.Instance[name] = self
+        self.agent_manager = None
+        self.root_path = root_path
+
+    def chapter_chain_to_memory(self):
+        mem = "##The Chapter Chain##\n"
+        for chapter in self.chapter_chain:
+            if (chapter.focus == CHAPTER_FOCUS or chapter.focus == CHAPTER_FINISH):
+                mem+=chapter.GetChapter()+"\n"
+        mem+="\n"
+        # mem+= "Remember your now chapter is "+ self.chapter_chain[self.chapter_chain_now].GetChapter()+"\n"
+        return Msg("system", mem, role="system")
+
+    def _has_unfinished_chapter(self):
+        for chapter in self.chapter_chain:
+            if chapter.focus == CHAPTER_LATTER or chapter.focus == CHAPTER_FOCUS:
+                return True
+        return False
+    def next_chapter(self, successful: bool = True):
+        if self.chapter_chain[self.chapter_chain_now].append_tools is not None or len(self.chapter_chain[self.chapter_chain_now].append_tools)==0:
+            for callable_tools, args in self.chapter_chain[self.chapter_chain_now].append_tools:
+                self.service_toolkit.Remove(callable_tools)
+        
+        self.chapter_chain[self.chapter_chain_now].Finished(successful)
+        try:
+            score_agent = ScoreAgent(name=self.name, model_config_name=self.model_config_name, memory=self.chapter_memory, 
+                                    chapter_score_prompt=self.chapter_chain[self.chapter_chain_now].score_prompt,
+                                    service_toolkit=self.service_toolkit, sys_prompt="", max_iters=self.max_iters, verbose=self.verbose)
+            res = score_agent.reply()
+
+            total_score_agent = TotalScoreAgent(name=self.name, model_config_name=self.model_config_name, input=res['content']['speak'], 
+                                                service_toolkit=self.service_toolkit, sys_prompt="", max_iters=self.max_iters, verbose=self.verbose)
+            score = total_score_agent.reply()
+
+        except Exception as e:
+            print("=*="*50)
+            print(e)
+        self.agent_manager.message_pass(self, "chapter_score", json.dumps({'chapter_id':self.chapter_chain_now+1, 'data':res}))
+        self.agent_manager.save_chapter_memory(self,self.chapter_chain[self.chapter_chain_now].course_id, self.chapter_chain[self.chapter_chain_now].lesson_id,  
+                                               self.chapter_chain[self.chapter_chain_now].title,  self.chapter_memory.export(to_mem = True),
+                                               score, is_rebuttal=False)
+        self.chapter_memory = TemporaryMemory()
+        self.agent_manager.message_pass(self, "next_chapter","")
+        self.agent_manager.system_message("进入下一节")
+
+        for i in range(len(self.chapter_chain)):
+            if self.chapter_chain[i].focus == CHAPTER_LATTER:
+                self.chapter_chain[i].Focus()
+                if self.chapter_chain[i].append_tools is not None and len(self.chapter_chain[i].append_tools) > 0:
+                    for callable_tools, args in self.chapter_chain[i].append_tools:
+                        print("add tools:", callable_tools)
+                        print("args:", args)
+                        args['root_path'] = self.root_path
+                        self.service_toolkit.add(callable_tools, **args)
+                self.chapter_chain_now = i
+                return "成功进入下一章"
+        # 找不到章节
+        return "学生完成了本课所有章节，已经没有下一章了！"
+                
+    def add_chapters_after(self, CHAPTER_No, chapters: Sequence[Chapter]):
+        for i in range(len(self.chapter_chain)):
+            if self.chapter_chain[i].No == CHAPTER_No:
+                for j in range(len(chapters)):
+                    CHAPTER_to_insert = Chapter(No = CHAPTER_No + j + 1, chapter = chapters[j], focus = 2, append_tools=
+                                          self.chapter_chain[self.chapter_chain_now].append_tools)
+                    self.chapter_chain.insert(i+1+j, CHAPTER_to_insert)
+                for j in range(i+len(chapters), len(self.chapter_chain)):
+                    self.chapter_chain[j].No += len(chapters)
+                break
+    def reply(self, x: Optional[Union[Msg, Sequence[Msg]]] = None, user_backboard:str = "") -> Msg:
+            if self.chapter_memory.size() <= 1:
+                self.chapter_chain[self.chapter_chain_now].memory = self.chapter_memory
+                for msg in self.memory.get_memory():
+                    self.chapter_memory.add(msg)
+                self.chapter_memory.add(self.chapter_chain_to_memory())
+            
+            self.chapter_memory.add(x)
+
+            
+            for _ in range(self.max_iters):
+                # Step 1: Thought
+                if self.verbose:
+                    self.speak(f" ITER {_+1}, STEP 1: REASONING ".center(70, "#"))
+
+                # Prepare hint to remind model what the response format is
+                # Won't be recorded in memory to save tokens
+                backboard_hint_msg = Msg(
+                    "system",
+                    user_backboard,
+                    role="system",
+                    echo=self.verbose,
+                )
+                tools_hint_msg = Msg(
+                    "system",
+                    self.service_toolkit.tools_instruction,
+                    role="system",
+                    echo=self.verbose,
+                )
+                hint_msg = Msg(
+                    "system",
+                    self.parser.format_instruction+"\n ####" + "!!"+ self.language_hint +"!!",
+                    role="system",
+                    echo=self.verbose,
+                )
+                prompt = self.model.format(self.chapter_memory.get_memory(),
+                    backboard_hint_msg, 
+                    # The instruction prompt for tools
+                    tools_hint_msg,
+                    hint_msg)
+                if self.verbose:
+                    self.speak(f"Prompt".center(70, "#"))
+                    self.speak(str(prompt))
+                
+                try:
+                    if self.verbose:
+                        self.speak(f"API Trigger".center(70, "#"))
+                    res = self.model(
+                        prompt,
+                        parse_func=self.parser.parse,
+                        max_retries=2,
+                        ######################
+                    )
+                    
+                    if self.verbose:
+                        self.speak(f"Result Parsed".center(70, "#"))
+                        self.speak(str(res.parsed))
+                        print(res.parsed)
+                        
+                    self.chapter_memory.add(
+                        Msg(
+                            self.name,
+                            self.parser.to_memory(res.parsed),
+                            "assistant",
+                        ),
+                    )
+
+                    # Print out the response
+                    msg_returned = Msg(
+                        self.name,
+                        self.parser.to_content(res.parsed),
+                        "assistant",
+                    )
+                    if self.verbose:
+                        self.speak(f"Speak".center(70, "#"))
+                        print(msg_returned)
+                    self.speak(msg_returned)
+                    return msg_returned
+                
+                    
+                    if self.verbose:
+                        self.speak(f"Function".center(70, "#"))
+                        self.speak(str(res.parsed["function"]))
+                        print(res.parsed["function"])
+
+                    arg_function = res.parsed["function"]
+                    if (
+                        isinstance(arg_function, str)
+                        and arg_function in ["[]", ""]
+                        or isinstance(arg_function, list)
+                        and len(arg_function) == 0
+                    ):
+                        # Only the speak field is exposed to users or other agents
+                        continue
+
+                # Only catch the response parsing error and expose runtime
+                # errors to developers for debugging
+                except ResponseParsingError as e:
+                    # Print out raw response from models for developers to debug
+                    response_msg = Msg(self.name, e.raw_response, "assistant")
+                    self.speak(response_msg)
+
+                    # Re-correct by model itself
+                    error_msg = Msg("system", str(e), "system")
+                    self.speak(error_msg)
+
+                    self.chapter_memory.add([response_msg, error_msg])
+
+                
+
+    def function_call(self,  parsed_function) -> Msg:
+                if self.verbose:
+                    self.speak(f"ACTING ".center(70, "#"))
+                    print(str(parsed_function))
+
+                # Parse, check and execute the tool functions in service toolkit
+                try:
+                    execute_results = self.service_toolkit.parse_and_call_func(
+                        parsed_function# res.parsed["function"],
+                    )
+                    # 在这里补充任务结束行动
+                    if self.finish_now_chapter: self.finish()
+                        
+
+                    # Note: Observing the execution results and generate response
+                    # are finished in the next reasoning step. We just put the
+                    # execution results into memory, and wait for the next loop
+                    # to generate response.
+
+                    # Record execution results into memory as system message
+                    msg_res = Msg("system", execute_results, "system")
+                    self.speak(msg_res)
+                    self.chapter_memory.add(msg_res)
+                    pattern = r"\[RESULT\]: (.*)"
+                    match = re.search(pattern, msg_res['content'])
+                    result_content='正则匹配失败'
+                    if match:
+                        result_content = match.group(1)
+                    res = self.reply(Msg(name='system', content="命令执行结果如上，请总结并引导学生进行下一步操作", role="system"))
+
+                    self.agent_manager.system_message(parsed_function['name'] + "执行" + ("成功" if "[STATUS]: SUCCESS" in str(msg_res['content']) else "失败") + "，结果为：" + result_content)
+                    self.agent_manager.message_pass(self, "message", res.content['speak'])
+
+                except FunctionCallError as e:
+                    # Catch the function calling error that can be handled by
+                    # the model
+                    error_msg = Msg("system", str(e), "system")
+                    self.speak(error_msg)
+                    self.chapter_memory.add(error_msg)
+
+                except Exception as e:
+                    # Catch the function calling error that can not be handled
+                    error_msg = Msg("system", str(e), "system")
+                    self.speak(error_msg)
+                    self.chapter_memory.add(error_msg)
+
+    def finish(self):
+        self.finish_now_chapter = False
+        hint_msg = Msg(
+            "system",
+            "You have finished your now chapter."
+            "Now generate a reply by summarizing the current "
+            "situation. Remember this summary will be recorded in global memory. Make sure keep important information.",
+            role="system",
+            echo=self.verbose,
+        )
+
+        # Generate a reply by summarizing the current situation
+        prompt = self.model.format(self.chapter_memory.get_memory(), hint_msg)
+        res = self.model(prompt)
+        res_msg = Msg(self.name, res.text, "assistant")
+        self.memory.add(res_msg)
+        self.speak(res_msg)
+        self.next_chapter(successful=True)
+        self.success_own_chapter = True
+
+    def change_language(self, language):
+        if(language == "zh"):self.language_hint = LANGUAGE_HINT_CN
+        if(language == "en"):self.language_hint = LANGUAGE_HINT_EN
+
+    def GetProcess(self):
+        process = []
+        t = {}
+        t['topic'] = "Global Memory"
+        mems = []
+        for msg in self.memory.get_memory():
+            mems.append(msg.serialize())
+
+        t['memory'] = "$MEMORYMEMORY$".join(mems)
+        t['finish'] = "True"
+        ts = f"{t['topic']}$TOPICTOPIC${t['memory']}$TOPICTOPIC${t['finish']}"
+        process.append(ts)
+        for chapter in self.chapter_chain:
+            t = {}
+            t['topic'] = f'Chapter {chapter.No}'
+            mems = []
+            if chapter.memory is not None:
+                for msg in self.memory.get_memory():
+                    mems.append(msg.serialize())
+            t['memory'] =   "$MEMORYMEMORY$".join(mems)
+            t['finish'] = "True" if chapter.focus == CHAPTER_FINISH else "False"
+            ts = f"{t['topic']}$TOPICTOPIC${t['memory']}$TOPICTOPIC${t['finish']}"
+            process.append(ts)
+        
+        returnString = ""
+        returnString = "$TASKTASK$".join(process)
+        return returnString
diff --git a/AlgoriAgent/projects/algoriAgent/agent/score_agent.py b/AlgoriAgent/projects/algoriAgent/agent/score_agent.py
new file mode 100644
index 0000000..67997bb
--- /dev/null
+++ b/AlgoriAgent/projects/algoriAgent/agent/score_agent.py
@@ -0,0 +1,180 @@
+
+from typing import Any, Callable, Optional, Union, Sequence
+
+from loguru import logger
+from agent.flex_service_toolkit import FlexServiceToolkit
+from agentscope.exception import ResponseParsingError, FunctionCallError
+from agentscope.agents import AgentBase
+from agentscope.memory.temporary_memory import TemporaryMemory
+from agentscope.message import Msg
+from agentscope.parsers import MarkdownJsonDictParser
+from agentscope.service.service_toolkit import ServiceFunction
+from agentscope.service import (
+    ServiceToolkit,
+    ServiceResponse,
+    ServiceExecStatus,
+)
+import json
+
+
+INSTRUCTION_SCORE_PROMPT_CN = """
+你是一名经验丰富的算法课程评分专家，你的任务是：基于学生对一个章节的学习日志、对话流，按照指定的评分标准进行评分。
+但是不同于普通老师，为了提高得分的区分度，你需要作为一名非常挑剔的专家，根据评分标准仔细挑剔学生的学习日志，并给出评分报告。
+
+重要的挑剔点：
+1. 存在错误的概念理解
+2. 存在矛盾的概念理解
+3. 存在不清晰、不连贯的概念表达
+4. 其他你认为可以挑剔的地方
+
+下面是学生的整个学习日志和对话流：
+"""
+
+CHAPTER_HINT_CN ="""
+上面就是学生的整个学习日志和对话流，请进行下面的各条章节评分标准进行挑剔与评分。
+"""
+
+RESPONSE_HINT_CN = """
+你的返回必须包含：
+对于每条章节评分标准，报告学生的得分以及挑剔的结果。
+例如：
+```
+{'评分标准1': {'得分': 8, '挑剔报告': '答案正确，但对于xxx概念表述有误'}
+}
+```
+"""
+
+INSTRUCTION_SCORE_PROMPT_EN = """
+You are a experienced algorithm course scoring expert, your task is to: Based on the students' learning logs and dialogues, score them according to the specified scoring standard.
+But unlike a normal teacher, to improve the distinction of the score, you need to be a very critical expert, carefully pick out the students' learning logs according to the scoring standard and give a score report.
+
+The important points to pick out:
+1. There are errors in the concept understanding
+2. There are contradictions in the concept understanding
+3. There are unclear, incoherent concept expressions
+4. Other places you think can be picked out
+Below is the students' entire learning log and dialogue:
+"""
+
+CHAPTER_HINT_EN = """
+Above is the students' entire learning log and dialogue, please pick out according to the following chapter scoring standard and score them.
+"""
+
+RESPONSE_HINT_EN="""
+Your response must contain:
+For each chapter scoring standard, report the students' score and pick out the results.
+For example:
+```
+{'Score standard 1': {'Score': 8, 'Pick out report': 'The answer is correct, but for the xxx concept expression is wrong'}
+}
+```
+"""
+
+
+class ScoreAgent(AgentBase):
+    """
+    ScoreAgent is a agent that can score students' learning logs.
+    """
+    def __init__(self, name: str, model_config_name: str, memory: TemporaryMemory,
+                 chapter_score_prompt: str,
+                 service_toolkit: ServiceToolkit, 
+                 sys_prompt: str = "You're a helpful assistant.",
+                 max_iters: int = 5,
+                 verbose: bool = True,
+                   **kwargs):
+        self.service_toolkit = service_toolkit
+        super().__init__(
+            name=name,
+            sys_prompt=sys_prompt,
+            model_config_name=model_config_name,
+        )
+        self.max_iters = max_iters
+        self.verbose = verbose
+        if not sys_prompt.endswith("\n"):
+            sys_prompt = sys_prompt + "\n"
+
+        self.sys_prompt = "\n".join(
+            [
+                # The brief intro of the role and target
+                sys_prompt.format(name=self.name),
+                # The detailed instruction prompt for the agent
+                INSTRUCTION_SCORE_PROMPT_CN,
+            ],
+        )
+        self.memory = TemporaryMemory()
+        self.memory.add(Msg("system", self.sys_prompt, role="system"))
+        
+        self.memory.add(memory.get_memory())
+
+        self.memory.add(Msg("system", CHAPTER_HINT_CN, role="system"))
+
+        self.memory.add(Msg("system", chapter_score_prompt, role="system"))
+
+        self.memory.add(Msg("system", RESPONSE_HINT_CN, role="system"))
+
+        self.parser = MarkdownJsonDictParser(
+            content_hint={
+                "thought": "what you thought",
+                "speak": "actual response in correct format mentioned above",
+            },
+            required_keys=["thought", "speak"],
+            # Only print the speak field when verbose is False
+            keys_to_content=True if self.verbose else "speak",
+        )
+
+    def reply(self, x = None, user_backboard = ""):
+        for _ in range(self.max_iters):
+            if self.verbose:
+                self.speak(f" ITER {_+1} ".center(70, "#"))
+            try:
+                
+                hint_msg = Msg(
+                    "system",
+                    self.parser.format_instruction+"\n",
+                    role="system",
+                    echo=self.verbose,
+                )
+                prompt = self.model.format(self.memory.get_memory(),hint_msg
+                    )
+                if self.verbose:
+                        self.speak(f"API Trigger".center(70, "#"))
+                        self.speak(str(prompt))
+                res = self.model(
+                    prompt,
+                    parse_func=self.parser.parse,
+                    max_retries=2,
+                    ######################
+                )
+
+                if self.verbose:
+                    self.speak(f"Result Parsed".center(70, "#"))
+                    self.speak(str(res.parsed))
+                    print(res.parsed)
+                self.memory.add(
+                        Msg(
+                            self.name,
+                            self.parser.to_memory(res.parsed),
+                            "assistant",
+                        ),
+                    )
+                msg_returned = Msg(
+                        self.name,
+                        self.parser.to_content(res.parsed),
+                        "assistant",
+                    )
+                return msg_returned
+            except ResponseParsingError as e:
+                    # Print out raw response from models for developers to debug
+                    response_msg = Msg(self.name, e.raw_response, "assistant")
+                    self.speak(response_msg)
+                    # Re-correct by model itself
+                    error_msg = Msg("system", str(e), "system")
+                    self.speak(error_msg)
+                    self.memory.add([response_msg, error_msg])
+            except Exception as e:
+                self.speak(f"Error: {e}".center(70, "#"))
+                self.speak(f"Retrying".center(70, "#"))
+                continue
+
+        return Msg(self.name, "Error: Max iterations reached", "assistant")
+
diff --git a/AlgoriAgent/projects/algoriAgent/agent/total_score_agent.py b/AlgoriAgent/projects/algoriAgent/agent/total_score_agent.py
new file mode 100644
index 0000000..26ce77c
--- /dev/null
+++ b/AlgoriAgent/projects/algoriAgent/agent/total_score_agent.py
@@ -0,0 +1,152 @@
+
+from typing import Any, Callable, Optional, Union, Sequence
+
+from loguru import logger
+from agent.flex_service_toolkit import FlexServiceToolkit
+from agentscope.exception import ResponseParsingError, FunctionCallError
+from agentscope.agents import AgentBase
+from agentscope.memory.temporary_memory import TemporaryMemory
+from agentscope.message import Msg
+from agentscope.parsers import MarkdownJsonDictParser
+from agentscope.service.service_toolkit import ServiceFunction
+from agentscope.service import (
+    ServiceToolkit,
+    ServiceResponse,
+    ServiceExecStatus,
+)
+import json
+
+
+INSTRUCTION_TOTAL_SCORE_PROMPT_CN = """
+你是一名精通文件处理的专家，你的本次任务非常简单，给你一段由其他老师给出的评分评价，你需要从中提取每一条标准下的得分，并计算总分，返回总分即可。
+"""
+
+
+RESPONSE_TOTAL_SCORE_HINT_CN = """
+你的返回必须包含且仅包含一个整数，表示得分。
+例如：
+输入：
+```
+{'评分标准1': {'得分': 8, '挑剔报告': '答案正确，但对于xxx概念表述有误'},
+ '评分标准2': {'得分': 3, '挑剔报告': '答案错误，对于xxx概念理解完全错误，导致回答有误'}
+}
+```
+输出：
+11
+下面是真正的输入：
+"""
+
+INSTRUCTION_TOTAL_SCORE_PROMPT_EN = """
+You are an expert in file processing. Your task is very simple. You will be given a segment of evaluation comments provided by another teacher. You need to extract the scores for each criterion and calculate the total score, returning only the total score.
+"""
+
+
+RESPONSE_TOTAL_SCORE_HINT_EN="""
+Your return must contain and only contain an integer representing the score. For example: 
+Input:
+```
+{'Criterion 1': {'Score': 8, 'Critical Report': 'The answer is correct, but there is a mistake in the explanation of the xxx concept'},
+ 'Criterion 2': {'Score': 3, 'Critical Report': 'The answer is incorrect, with a complete misunderstanding of the xxx concept, leading to an incorrect response'}
+}
+```
+Output:
+```
+{'output': 11}
+```
+Below is the actual input:
+"""
+
+
+class TotalScoreAgent(AgentBase):
+    """
+    TotalScoreAgent is a agent that can gather students' all score.
+    """
+    def __init__(self, name: str, model_config_name: str, input: str,
+                 service_toolkit: ServiceToolkit, 
+                 sys_prompt: str = "You're a helpful assistant.",
+                 max_iters: int = 5,
+                 verbose: bool = True,
+                   **kwargs):
+        self.service_toolkit = service_toolkit
+        super().__init__(
+            name=name,
+            sys_prompt=sys_prompt,
+            model_config_name=model_config_name,
+        )
+        self.max_iters = max_iters
+        self.verbose = verbose
+        if not sys_prompt.endswith("\n"):
+            sys_prompt = sys_prompt + "\n"
+
+        self.sys_prompt = "\n".join(
+            [
+                # The brief intro of the role and target
+                sys_prompt.format(name=self.name),
+                # The detailed instruction prompt for the agent
+                INSTRUCTION_TOTAL_SCORE_PROMPT_CN,
+                RESPONSE_TOTAL_SCORE_HINT_CN
+            ],
+        )
+        self.memory = TemporaryMemory()
+        self.memory.add(Msg("system", self.sys_prompt, role="system"))
+        
+        self.memory.add(Msg("User", input, role="system"))
+        self.parser = MarkdownJsonDictParser(
+            content_hint={
+                "output": "total score",
+            },
+            required_keys=["output"],
+            # Only print the speak field when verbose is False
+            keys_to_content=True if self.verbose else "speak",
+        )
+
+    def reply(self, x = None, user_backboard = ""):
+        for _ in range(self.max_iters):
+            if self.verbose:
+                self.speak(f" ITER {_+1} ".center(70, "#"))
+            try:
+                
+                hint_msg = Msg(
+                    "system",
+                    self.parser.format_instruction+"\n",
+                    role="system",
+                    echo=self.verbose,
+                )
+                prompt = self.model.format(self.memory.get_memory(),hint_msg
+                    )
+                if self.verbose:
+                        self.speak(f"API Trigger".center(70, "#"))
+                        self.speak(str(prompt))
+                res = self.model(
+                    prompt,
+                    parse_func=self.parser.parse,
+                    max_retries=2,
+                    ######################
+                )
+
+                print('====-======-=====')
+
+                return res.parsed['output']
+
+
+                if self.verbose:
+                    self.speak(f"Result Parsed".center(70, "#"))
+                    self.speak(str(res))
+                    print(res)
+                    
+                return res
+            except ResponseParsingError as e:
+                    # Print out raw response from models for developers to debug
+                    response_msg = Msg(self.name, e.raw_response, "assistant")
+                    self.speak(response_msg)
+                    # Re-correct by model itself
+                    error_msg = Msg("system", str(e), "system")
+                    self.speak(error_msg)
+                    self.memory.add([response_msg, error_msg])
+            except Exception as e:
+                self.speak(f"Error: {e}".center(70, "#"))
+                self.speak(f"Retrying".center(70, "#"))
+                continue
+
+        return Msg(self.name, "Error: Max iterations reached", "assistant")
+
diff --git a/AlgoriAgent/projects/algoriAgent/agent_manager.py b/AlgoriAgent/projects/algoriAgent/agent_manager.py
index 128e80b..a2f35bd 100644
--- a/AlgoriAgent/projects/algoriAgent/agent_manager.py
+++ b/AlgoriAgent/projects/algoriAgent/agent_manager.py
@@ -1,235 +1,329 @@
-import sys  
-import os  
-current_directory = os.path.dirname(os.path.abspath(__file__))    
-sys.path.append(current_directory)  
-
-import agentscope
-from agentscope.message import Msg
-import os
-import asyncio
-import threading
-import json
-
-from agent.flex_service_toolkit import *
-from AlgoriAgent.projects.algoriAgent.agent.algori_agent import *
-
-from agentscope.service import (
-    ServiceToolkit,
-    ServiceResponse,
-    ServiceExecStatus,
-)
-import configparser
-
-config = configparser.ConfigParser()
-config.read('config.ini')
-openai_api_key = config['Global']['api_key']
-
-
-
-OPENAI_CFG_DICT = {
-    "config_name": "openai_cfg",    # 此配置的名称，必须保证唯一
-    "model_type": "openai_chat",         # 模型类型
-    "model_name": "gpt-4o-mini",  # 模型名称
-    #"model_name": "gpt-4",  # 模型名称
-    #"model_name": "llama3",
-    
-    "api_key": openai_api_key,      # OpenAI API key. 如果没有设置，将使用环境变量中的 OPENAI_API_KEY
-
-    "client_args": {
-        "base_url": config['Global']['base_chat_url']
-    },
-
-}
-
-import uuid
-from AlgoriAgent.projects.algoriAgent.tools.judge_tools import judge
-EMPTY_CHAPTER_CHAIN = [ Chapter(1, CHAPTER_FOCUS, "本章是未打开某个具体章节时的默认章节。", "处于本章节时不会有任何章节跳转。请作为一名经验丰富的算法教师，回答用户的问题。") ]
-
-def tool_name_to_tool(tool_name_list):
-    tools = []
-    for tool_name in tool_name_list:
-        if tool_name == "":
-            continue
-        if tool_name == "judge":
-            tools.append(judge)
-            
-    return tools
-
-class AgentManager:
-    def __init__(self):
-
-        agentscope.init(model_configs=[OPENAI_CFG_DICT])#, studio_url="http://0.0.0.0:5000")
-        self.agents = {}
-
-    def new_agent(self, markdown:str, markdown_prompt:str, score_prompt:str, id = None):
-        '''
-        markdown: 教案的markdown文件内容
-        markdown_prompt: 教案的prompt的markdown文件内容
-        score_prompt: 评分的prompt的markdown文件内容
-        根据3个教案的prompt，生成一个agent，返回agent的id与agent
-        '''
-        markdown_list = markdown.split("\n")
-        markdown_prompt_list = markdown_prompt.split("\n")
-        score_prompt_list = score_prompt.split("\n")
-        # 获取 H1 标题
-        title = ""
-        for line in markdown_list:
-            if line.startswith("# "):
-                title = line[2:]
-                break
-        
-        # 对于 H3 标题 构造每一个 Chapter
-        # 首先找到所有的 H3 标题 
-        chapter_dict = {}
-        chapter_sequence = []
-        for line in markdown_list:
-            if line.startswith("### "):
-                chapter_name = line[4:]
-                chapter_dict[chapter_name] = {}
-                chapter_sequence.append(chapter_name)
-        # 将 H3 标题 和 其对应的内容 构造成一个 Chapter
-
-        h3_name = ""
-        content = ""
-        for i in range(len(markdown_list)):
-            if markdown_list[i].startswith("### "): 
-                if(h3_name != ""):
-                    chapter_dict[h3_name]["markdown"] = content
-                h3_name = markdown_list[i][4:]
-                content = ""
-                continue
-            if h3_name != "":
-                content += markdown_list[i]+"\n"
-        if(h3_name != ""):
-            chapter_dict[h3_name]["markdown"] = content
-        
-
-        h3_name = ""
-        content = ""
-        require_tools = []
-        for i in range(len(markdown_prompt_list)):
-            if markdown_prompt_list[i].startswith("### "): 
-                if(h3_name != ""):
-                    chapter_dict[h3_name]["markdown_prompt"] = content
-                    chapter_dict[h3_name]["require_tools"] = tool_name_to_tool(require_tools)
-                h3_name = markdown_prompt_list[i][4:]
-                content = ""
-                continue
-
-            if h3_name != "":
-                if markdown_prompt_list[i].startswith("_require_tools"):
-                    require_tools.append(markdown_prompt_list[i].split("=")[1].strip())
-                    continue
-                content += markdown_prompt_list[i]+"\n"
-        if(h3_name != ""):
-            chapter_dict[h3_name]["markdown_prompt"] = content
-            chapter_dict[h3_name]["require_tools"] = tool_name_to_tool(require_tools)
-            
-        
-        h3_name = ""
-        content = ""
-        for i in range(len(score_prompt_list)):
-            if score_prompt_list[i].startswith("### "): 
-                if(h3_name != ""):
-                    chapter_dict[h3_name]["score_prompt"] = content
-                h3_name = score_prompt_list[i][4:]
-                content = ""
-                continue
-
-            if h3_name != "":
-                content += score_prompt_list[i]+"\n"
-        if(h3_name != ""):
-            chapter_dict[h3_name]["score_prompt"] = content
-
-
-        chapter_chain = []
-        No = 1
-        print (chapter_dict)
-        for chapter_name in chapter_sequence:
-            chapter_chain.append(Chapter(No, CHAPTER_LATTER, title, chapter_dict[chapter_name]["markdown"], chapter_dict[chapter_name]["markdown_prompt"], chapter_dict[chapter_name]["score_prompt"],chapter_dict[chapter_name]["require_tools"]))
-            No+=1
-        chapter_chain[0].Focus()
-
-
-        # Prepare the tools for the agent
-        service_toolkit = FlexServiceToolkit()
-        # for tool_function in unity_function_list:
-        #     service_toolkit.add(tool_function)
-        agent = ChapterChainAgent(
-                    name="assistant",
-                    model_config_name="openai_cfg",
-                    verbose=True,
-                    service_toolkit=service_toolkit,
-                    max_iters=5,
-                    chapter_chain=chapter_chain
-                )
-        self.agents[id] = agent
-        return id, agent
-        
-
-
-    def new_agent_with_chain(self, chapter_chain = None, id = None):
-        '''
-
-        '''
-        if chapter_chain is None:
-            chapter_chain = EMPTY_CHAPTER_CHAIN
-        
-        # Prepare the tools for the agent
-        service_toolkit = FlexServiceToolkit()
-        # for tool_function in unity_function_list:
-        #     service_toolkit.add(tool_function)
-        agent = ChapterChainAgent(
-                    name="assistant",
-                    model_config_name="openai_cfg",
-                    verbose=True,
-                    service_toolkit=service_toolkit,
-                    max_iters=5,
-                    chapter_chain=EMPTY_CHAPTER_CHAIN
-                )
-        if id is None: id = uuid.uuid4()
-        self.agents[id] = agent
-        return id, agent
-
-
-    def get_agent(self, id=None):
-        '''
-            如果在get agent之前没有new agent，并用id进行访问，就返回一个章节链为空的agent
-        '''
-        if id:
-            if  id in self.agents:
-                return self.agents[id]
-            self.agents[id] = self.new_agent(id)[1]
-        else:
-            id = str(uuid.uuid4())
-            self.agents[id] = self.new_agent()[1]
-        return id, self.agents[id]
-        
-
-    def invoke(self,id, query, user_backboard):
-        msg = Msg("user", query, role="user")
-        return self.agents[id](msg, user_backboard = user_backboard)
-    
-    
-import time
-if __name__ == '__main__':
-    # tool_demo = ToolDemo("Lava")
-    # response = tool_demo.invoke("I want such a Gurouce, when use it, it will be thrown and fly towards the mouse position for a short time, and stop. And will damage Enemy who step on it. I call it Lava Gurouce.")
-    # print(response)
-    # Start the WebSocket server in a separate thread
-    manager = AgentManager()
-    id, agent = manager.new_agent()
-
-    # Main thread logic
-    try:
-        # Your code that might be interrupted
-        while True:
-            input_data = input("用户：")
-            print(manager.invoke(id, input_data))
-            
-    except KeyboardInterrupt:
-        print("Process interrupted by user.")
-        # You can add any cleanup code here if needed
-    finally:
-        print("Exiting program.")
-        # Code to run before the program exits
+import sys  
+import os  
+current_directory = os.path.dirname(os.path.abspath(__file__))    
+sys.path.append(current_directory)  
+from flask_socketio import SocketIO, join_room, emit, Namespace
+import agentscope
+from agentscope.message import Msg
+import os
+import asyncio
+import threading
+import json
+
+from agent.flex_service_toolkit import *
+from AlgoriAgent.projects.algoriAgent.agent.algori_agent import *
+
+from agentscope.service import (
+    ServiceToolkit,
+    ServiceResponse,
+    ServiceExecStatus,
+)
+import configparser
+
+config = configparser.ConfigParser()
+config.read('config.ini')
+openai_api_key = config['Global']['api_key']
+
+
+
+OPENAI_CFG_DICT = {
+    "config_name": "openai_cfg",    # 此配置的名称，必须保证唯一
+    "model_type": "openai_chat",         # 模型类型
+    "model_name": "gpt-4o-mini",  # 模型名称
+    #"model_name": "gpt-4",  # 模型名称
+    #"model_name": "llama3",
+    "api_key": openai_api_key,      # OpenAI API key. 如果没有设置，将使用环境变量中的 OPENAI_API_KEY
+
+    "client_args": {
+        "base_url": config['Global']['base_chat_url']
+    },
+
+}
+
+import uuid
+from AlgoriAgent.projects.algoriAgent.tools.judge_tools import judge
+EMPTY_CHAPTER_CHAIN = [ Chapter(1, CHAPTER_FOCUS, "本章是未打开某个具体章节时的默认章节。", "处于本章节时不会有任何章节跳转。请作为一名经验丰富的算法教师，回答用户的问题。") ]
+
+def tool_name_to_tool(tool_name_list):
+    tools = []
+    for tool_name in tool_name_list:
+        if tool_name == "":
+            continue
+        if tool_name == "judge":
+            tools.append(judge)
+            
+    return tools
+
+def tool_name_to_tool_with_args(tool_name_list, tool_args_list) -> list[tuple[Callable, dict]]:
+    tools = []
+    for tool_name, tool_args in zip(tool_name_list, tool_args_list):
+        if tool_name == "":
+            continue
+        if tool_name == "judge":
+            tools.append((judge, {"course_id": tool_args[0], "lesson_id": tool_args[1], "name": tool_args[2]}))
+            
+    return tools
+class AgentManager:
+    def __init__(self,max_iter = 5, app=None, socketio=None):
+
+        agentscope.init(model_configs=[OPENAI_CFG_DICT])#, studio_url="http://0.0.0.0:5000")
+        self.agents = {}
+        self.max_iter = max_iter
+        self.agent_to_id = {}
+        self.app = app
+        self.socketio = socketio
+    def new_agent(self, course_id, lesson_id, markdown:str, markdown_prompt:str, score_prompt:str, id = "Defult Assistant", root_path="."):
+        '''
+        markdown: 教案的markdown文件内容
+        markdown_prompt: 教案的prompt的markdown文件内容
+        score_prompt: 评分的prompt的markdown文件内容
+        根据3个教案的prompt，生成一个agent，返回agent的id与agent
+        '''
+        markdown_list = markdown.split("\n")
+        markdown_prompt_list = markdown_prompt.split("\n")
+        score_prompt_list = score_prompt.split("\n")
+        # 获取 H1 标题
+        title = ""
+        for line in markdown_list:
+            if line.startswith("# "):
+                title = line[2:]
+                break
+        
+        # 对于 H3 标题 构造每一个 Chapter
+        # 首先找到所有的 H3 标题 
+        chapter_dict = {}
+        chapter_sequence = []
+        for line in markdown_list:
+            if line.startswith("### "):
+                chapter_name = line[4:]
+                chapter_dict[chapter_name] = {}
+                chapter_sequence.append(chapter_name)
+        # 将 H3 标题 和 其对应的内容 构造成一个 Chapter
+
+        h3_name = ""
+        content = ""
+        for i in range(len(markdown_list)):
+            if markdown_list[i].startswith("### "): 
+                if(h3_name != ""):
+                    chapter_dict[h3_name]["markdown"] = content
+                h3_name = markdown_list[i][4:]
+                content = ""
+                continue
+            if h3_name != "":
+                content += markdown_list[i]+"\n"
+        if(h3_name != ""):
+            chapter_dict[h3_name]["markdown"] = content
+        
+
+        h3_name = ""
+        content = ""
+        require_tools = []
+        require_tools_args = []
+        for i in range(len(markdown_prompt_list)):
+            if markdown_prompt_list[i].startswith("### "): 
+                if(h3_name != ""):
+                    chapter_dict[h3_name]["markdown_prompt"] = content
+                    chapter_dict[h3_name]["require_tools"] = tool_name_to_tool_with_args(require_tools, require_tools_args)
+                h3_name = markdown_prompt_list[i][4:]
+                content = ""
+                require_tools=[]
+                require_tools_args = []
+                continue
+
+            if h3_name != "":
+                if markdown_prompt_list[i].startswith("_require_tools"):
+                    require_tools.append(markdown_prompt_list[i].split("=")[1].strip().split(",")[0])
+                    require_tools_args.append((markdown_prompt_list[i].split("=")[1].strip().split(",")[1:]))
+                    continue
+                content += markdown_prompt_list[i]+"\n"
+        if(h3_name != ""):
+            chapter_dict[h3_name]["markdown_prompt"] = content
+            chapter_dict[h3_name]["require_tools"] = tool_name_to_tool_with_args(require_tools, require_tools_args)
+            
+        
+        h3_name = ""
+        content = ""
+        for i in range(len(score_prompt_list)):
+            if score_prompt_list[i].startswith("### "): 
+                if(h3_name != ""):
+                    chapter_dict[h3_name]["score_prompt"] = content
+                h3_name = score_prompt_list[i][4:]
+                content = ""
+                continue
+
+            if h3_name != "":
+                content += score_prompt_list[i]+"\n"
+        if(h3_name != ""):
+            chapter_dict[h3_name]["score_prompt"] = content
+
+
+        chapter_chain = []
+        No = 1
+        print (chapter_dict)
+        for chapter_name in chapter_sequence:
+            chapter_chain.append(Chapter(course_id, lesson_id, No, CHAPTER_LATTER, chapter_name, chapter_dict[chapter_name]["markdown"], chapter_dict[chapter_name]["markdown_prompt"], chapter_dict[chapter_name]["score_prompt"],chapter_dict[chapter_name]["require_tools"]))
+            No+=1
+        chapter_chain[0].Focus()
+
+
+        # Prepare the tools for the agent
+        service_toolkit = FlexServiceToolkit()
+        # for tool_function in unity_function_list:
+        #     service_toolkit.add(tool_function)
+        agent = ChapterChainAgent(
+                    name=id,
+                    model_config_name="openai_cfg",
+                    verbose=True,
+                    service_toolkit=service_toolkit,
+                    max_iters=5,
+                    chapter_chain=chapter_chain,
+                    root_path = root_path,
+                )
+        self.agents[id] = agent
+        self.agent_to_id[agent] = id
+        agent.agent_manager = self
+        return id, agent
+        
+    def message_pass(self, agent, messagetype, message):
+        id = self.agent_to_id[agent]
+        print("message_pass", id, messagetype, message)
+        with self.app.app_context(): 
+            try:
+                self.socketio.emit(messagetype, message, room=id, namespace='/agent')
+            except Exception as e:
+                print("message_pass",e)
+
+    def system_message(self, message):
+        with self.app.app_context(): 
+            try:
+                self.socketio.emit('system_message', message, room=id, namespace='/agent')
+            except Exception as e:
+                print("ERROR: system_message",e)
+
+
+    def save_chapter_memory(self, agent,course_id, lesson_id, subchapter_title, mem_list, score, is_rebuttal):
+        id = self.agent_to_id[agent]
+        self.app.my_function.save_chapter_memory(id, course_id, lesson_id, subchapter_title, mem_list, score, is_rebuttal)
+
+
+    def new_agent_with_chain(self, chapter_chain = None, id = None):
+        if chapter_chain is None:
+            chapter_chain = EMPTY_CHAPTER_CHAIN
+        
+        # Prepare the tools for the agent
+        service_toolkit = FlexServiceToolkit()
+        # for tool_function in unity_function_list:
+        #     service_toolkit.add(tool_function)
+        agent = ChapterChainAgent(
+                    name="assistant",
+                    model_config_name="openai_cfg",
+                    verbose=True,
+                    service_toolkit=service_toolkit,
+                    max_iters=5,
+                    chapter_chain=EMPTY_CHAPTER_CHAIN
+                )
+        if id is None: id = uuid.uuid4()
+        self.agents[id] = agent
+        return id, agent
+
+
+    def get_agent(self, id=None):
+        '''
+            如果在get agent之前没有new agent，并用id进行访问，就返回一个章节链为空的agent
+        '''
+        if id:
+            if  id in self.agents:
+                return self.agents[id]
+            self.agents[id] = self.new_agent(id)[1]
+        else:
+            id = str(uuid.uuid4())
+            self.agents[id] = self.new_agent()[1]
+        return id, self.agents[id]
+        
+
+    def invoke(self,id, query, user_backboard, reset=False):
+        msg = Msg("user", query, role="user")
+        agent = self.agents[id]
+        res  = agent(msg, user_backboard=user_backboard)
+        return res
+        # 暂时取消生成器的写法
+        # 假设 agent 是一个生成器
+        # if reset or not hasattr(agent, '_generator'):
+        #     agent._generator = agent(msg, user_backboard=user_backboard)
+        
+        # try:
+        #     response = next(agent._generator)
+        #     yield response
+        # except StopIteration:
+        #     return
+    
+    def function_call(self, id, arg_function):
+        agent = self.agents[id]
+        return agent.function_call(arg_function)
+
+    def change_language(self, id, language):
+        agent = self.agents[id]
+        agent.change_language(language)
+
+    def sample_judge(self, id, bb):
+        agent = self.agents[id]
+        
+        with self.app.app_context():
+            try:
+                    print(bb.get_active_file_reletive_path())
+                    assert bb is not None and bb.active_file_path != "" , "no path specified"
+                    if bb.get_active_file_reletive_path().endswith(".py"):
+                        self.socketio.emit('terminal', {'data': f"python {bb.get_active_file_reletive_path()}"}, room=id, namespace='/vscode')
+                        self.socketio.emit('system_message', "代码试运行尚在建设，推荐直接通过命令行执行！", room=id, namespace='/agent')
+                    else:
+                        self.socketio.emit('system_message', "代码试运行尚仅支持python文件", room=id, namespace='/agent')
+            except Exception as e:
+                print(e)
+                if (str(e) == "no path specified"):
+                    self.socketio.emit("system_message", "未加载IDE文件同步，请尝试重新打开IDE文件或刷新页面，并确保插件已连接", room=id, namespace='/agent')
+
+    def judge(self, id, bb):
+        agent = self.agents[id]
+        with self.app.app_context():
+            try:
+                if (agent.chapter_chain[agent.chapter_chain_now].append_tools is not None and len(agent.chapter_chain[agent.chapter_chain_now].append_tools)):
+                    for tool, tool_kwargs in agent.chapter_chain[agent.chapter_chain_now].append_tools:
+                        if tool.__name__ == "judge":
+                                print(bb.get_active_file_reletive_path())
+                                assert bb is not None and bb.active_file_path != "" , "no path specified"
+                                if bb.get_active_file_reletive_path().endswith(".py"):
+                                    # self.socketio.emit('terminal', {'data': f"python {bb.get_active_file_reletive_path()}"}, room=id, namespace='/vscode')
+                                    self.socketio.emit('system_message', "代码评测中，请稍后", room=id, namespace='/agent')
+                                    agent.function_call({'name':'judge','arguments':{'filepath': bb.get_active_file_reletive_path()}})
+                                else:
+                                    self.socketio.emit('system_message', "代码评测尚仅支持python文件", room=id, namespace='/agent')
+
+                        else:
+                            with self.app.app_context():
+                                    self.socketio.emit("system_message", "本章学习无需评测", room=id, namespace='/agent')
+                else:
+                    with self.app.app_context():
+                            self.socketio.emit("system_message", "本章学习无需评测", room=id, namespace='/agent')
+            except Exception as e:
+                if (str(e) == "no path specified"):
+                    self.socketio.emit("system_message", "未加载IDE文件同步，请尝试重新打开IDE文件或刷新页面，并确保插件已连接", room=id, namespace='/agent')
+
+
+import time
+if __name__ == '__main__':
+    manager = AgentManager()
+    id, agent = manager.new_agent()
+
+    # Main thread logic
+    try:
+        # Your code that might be interrupted
+        while True:
+            input_data = input("用户：")
+            print(manager.invoke(id, input_data))
+            
+    except KeyboardInterrupt:
+        print("Process interrupted by user.")
+        # You can add any cleanup code here if needed
+    finally:
+        print("Exiting program.")
+        # Code to run before the program exits
diff --git a/AlgoriAgent/projects/algoriAgent/tools/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/projects/algoriAgent/tools/__pycache__/__init__.cpython-310.pyc
index 36bf43a..bdca170 100644
Binary files a/AlgoriAgent/projects/algoriAgent/tools/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/projects/algoriAgent/tools/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/projects/algoriAgent/tools/__pycache__/judge_tools.cpython-310.pyc b/AlgoriAgent/projects/algoriAgent/tools/__pycache__/judge_tools.cpython-310.pyc
index 2cb7089..c90bb77 100644
Binary files a/AlgoriAgent/projects/algoriAgent/tools/__pycache__/judge_tools.cpython-310.pyc and b/AlgoriAgent/projects/algoriAgent/tools/__pycache__/judge_tools.cpython-310.pyc differ
diff --git a/AlgoriAgent/projects/algoriAgent/tools/file_tools.py b/AlgoriAgent/projects/algoriAgent/tools/file_tools.py
new file mode 100644
index 0000000..1bd232e
--- /dev/null
+++ b/AlgoriAgent/projects/algoriAgent/tools/file_tools.py
@@ -0,0 +1,54 @@
+from agentscope.service.service_toolkit import ServiceFunction
+from agentscope.service import (
+    ServiceToolkit,
+    ServiceResponse,
+    ServiceExecStatus,
+)
+import os
+def read_file(filepath:str, root_path:str):
+    """Read the file and return the content. Remember use 'judge' function if you want to judge file, instead of this function.
+
+    Args:
+        filepath (`str`): The path to the file.
+        root_path (`str`): The root path of now student's workspace.
+
+    Returns:
+        content (`str`): The content of the file.
+    """
+    # 从books/code_tests/name/中获取所有.in文件
+
+    filepath = os.path.join('..', root_path, filepath)
+    try:
+        with open(filepath, 'r') as f:
+            content = f.read()
+    except Exception as e:
+        status = ServiceExecStatus.ERROR
+        content = str(e)
+        return ServiceResponse(status, (content))
+    
+    status = ServiceExecStatus.SUCCESS
+    return ServiceResponse(status, ('```\n'+content+'```\n'))
+
+
+def write_file(filepath:str, content:str, root_path:str):
+    """Write content to the file.
+
+    Args:
+        filepath (`str`): The path to the file.
+        content (`str`): The content to be written to the file.
+        root_path (`str`): The root path of now student's workspace.
+
+    Returns:
+        None.
+    """
+    filepath = os.path.join('..', root_path, filepath)
+    try:
+        with open(filepath, 'w') as f:
+            f.write(content)
+    except Exception as e:
+        status = ServiceExecStatus.ERROR
+        return ServiceResponse(status, None)
+    status = ServiceExecStatus.SUCCESS
+    return ServiceResponse(status, None)
+        
+file_tools = [read_file, write_file]
\ No newline at end of file
diff --git a/AlgoriAgent/projects/algoriAgent/tools/judge_tools.py b/AlgoriAgent/projects/algoriAgent/tools/judge_tools.py
index 5b7ce5b..c434540 100644
--- a/AlgoriAgent/projects/algoriAgent/tools/judge_tools.py
+++ b/AlgoriAgent/projects/algoriAgent/tools/judge_tools.py
@@ -1,64 +1,78 @@
-from agentscope.service.service_toolkit import ServiceFunction
-from agentscope.service import (
-    ServiceToolkit,
-    ServiceResponse,
-    ServiceExecStatus,
-)
-import os
-def judge(filepath:str, name: str):
-    """Call this method to judge the student's solution after they have completed the problem.
-
-    Args:
-        filepath (`str`): The path to the file containing the student's solution.
-        name (`str`): The name of the problem.
-
-    Returns:
-        Report `tuple(int,str)`: The report of the student's solution.
-    """
-    # 从books/code_tests/name/中获取所有.in文件
-
-    directory = 'books/code_tests/'+name
-    # 如果目录不存在，报错
-    if not os.path.exists(directory):
-        status = ServiceExecStatus.FAILED
-        return ServiceResponse(status, f"The name {name} is not found.")
-    files = os.listdir(directory)
-    in_files = [file for file in files if file.endswith('.in')]
-    save_dir = os.path.dirname(filepath)
-    Report = ""
-    passed_num=0
-    failed_num=0
-    for file in in_files:
-        os.remove(os.path.join(save_dir, f"_{name}.out"))
-        cmd = "cat " + os.path.join(directory, file) + " | " + "python " + filepath +" > " + os.path.join(save_dir, f"_{name}.out")
-        os.system(cmd)
-        # 比较结果
-        re = ""
-        if compare_file(os.path.join(save_dir, f"_{name}.out"), os.path.join(directory, file.replace(".in", ".out"))):
-            re = "Passed {file}"
-            passed_num += 1
-        else:
-            re = "Failed {file}"
-            failed_num += 1
-        Report = Report + re + "\n"
-    os.remove(os.path.join(save_dir, f"_{name}.out"))
-    score = int (100 * passed_num / (passed_num + failed_num))
-    
-
-
-    status = ServiceExecStatus.SUCCESS
-    return ServiceResponse(status, (Report, score))
-
-
-def compare_file(path1:str, path2:str):
-    """
-        Compare two files and return true if they are the same.
-    """
-    with open(path1, 'r') as f1, open(path2, 'r') as f2:
-        # 逐行比较
-        for line1, line2 in zip(f1, f2):
-            if line1.strip() != line2.strip():
-                return False
-    return True
-        
+from agentscope.service.service_toolkit import ServiceFunction
+from agentscope.service import (
+    ServiceToolkit,
+    ServiceResponse,
+    ServiceExecStatus,
+)
+import os
+def judge(filepath:str, name: str, course_id: str, lesson_id: str, root_path:str):
+    """Call this method to judge the student's solution after they have completed the problem.
+
+    Args:
+        filepath (`str`): The path to the file containing the student's solution.
+        name (`str`): The name of the problem.
+        course_id (`str`): The id of the course.
+        lesson_id (`str`): The id of the lesson.
+        root_path (`str`): The root path of the user's workspace project.
+
+    Returns:
+        Report `tuple(int,str)`: The report of the student's solution.
+    """
+    # 从books/code_tests/name/中获取所有.in文件
+
+    directory = 'books/code_tests/'+course_id+'/'+name
+    # 如果目录不存在，报错
+    if not os.path.exists(directory):
+        status = ServiceExecStatus.FAILED
+        return ServiceResponse(status, f"The name {name} is not found.")
+    files = os.listdir(directory)
+    in_files = [file for file in files if file.endswith('.in')]
+    save_dir = root_path
+    Report = ""
+    passed_num=0
+    failed_num=0
+    erroroutput = ""
+    exceptoutput=""
+    for file in in_files:
+        file_path = os.path.join(save_dir, f"_{name}.out")
+        if os.path.exists(file_path):
+            os.remove(file_path)
+        with open(file_path, 'w') as ff:
+            ff.write("") 
+        
+        cmd = "cat " + os.path.join(directory, file) + " | " + "python " + os.path.join(root_path, filepath)  +" > " + os.path.join(save_dir, f"_{name}.out")
+        os.system(cmd)
+        # 比较结果
+        re = ""
+        if compare_file(os.path.join(save_dir, f"_{name}.out"), os.path.join(directory, file.replace(".in", ".out"))):
+            re = f"Passed {file}"
+            passed_num += 1
+        else:
+            re = f"Failed {file}"
+            failed_num += 1
+            with open(os.path.join(directory, file.replace(".in", ".out")),'r') as f: exceptoutput = f.read()
+            with open(os.path.join(save_dir, f"_{name}.out"),'r') as f: erroroutput = f.read()
+        Report = Report + re + "\n"
+    os.remove(os.path.join(save_dir, f"_{name}.out"))
+    score = int (100 * passed_num / (passed_num + failed_num))
+    
+
+
+    status = ServiceExecStatus.SUCCESS
+    return ServiceResponse(status, (score, f"评测得分为 {score}, 在 {passed_num + failed_num} 个测试用例中通过了 {passed_num}个。最后一个错误样例期望输出：'{exceptoutput}'，但你的输出：'{erroroutput}"))
+
+
+def compare_file(path1:str, path2:str):
+    """
+        Compare two files and return true if they are the same.
+    """
+    with open(path1, 'r') as f1, open(path2, 'r') as f2:
+        # 逐行比较
+        print('-------------------COMPARE-------------------------')
+        for line1, line2 in zip(f1, f2):
+            print(line1.strip(), "||",line2.strip())
+            if line1.strip() != line2.strip():
+                return False
+    return True
+        
 judge_tools = [judge]
\ No newline at end of file
diff --git a/AlgoriAgent/src/agentscope/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/__pycache__/__init__.cpython-310.pyc
index 92a3bf9..40f996a 100644
Binary files a/AlgoriAgent/src/agentscope/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/__pycache__/_init.cpython-310.pyc b/AlgoriAgent/src/agentscope/__pycache__/_init.cpython-310.pyc
index 1db3ab8..80b4292 100644
Binary files a/AlgoriAgent/src/agentscope/__pycache__/_init.cpython-310.pyc and b/AlgoriAgent/src/agentscope/__pycache__/_init.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/__pycache__/_runtime.cpython-310.pyc b/AlgoriAgent/src/agentscope/__pycache__/_runtime.cpython-310.pyc
index d69bbf2..0795868 100644
Binary files a/AlgoriAgent/src/agentscope/__pycache__/_runtime.cpython-310.pyc and b/AlgoriAgent/src/agentscope/__pycache__/_runtime.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/__pycache__/_version.cpython-310.pyc b/AlgoriAgent/src/agentscope/__pycache__/_version.cpython-310.pyc
index 06773cc..413879d 100644
Binary files a/AlgoriAgent/src/agentscope/__pycache__/_version.cpython-310.pyc and b/AlgoriAgent/src/agentscope/__pycache__/_version.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/__pycache__/constants.cpython-310.pyc b/AlgoriAgent/src/agentscope/__pycache__/constants.cpython-310.pyc
index 3d462ca..6c56bec 100644
Binary files a/AlgoriAgent/src/agentscope/__pycache__/constants.cpython-310.pyc and b/AlgoriAgent/src/agentscope/__pycache__/constants.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/__pycache__/exception.cpython-310.pyc b/AlgoriAgent/src/agentscope/__pycache__/exception.cpython-310.pyc
index f32c39c..63925d5 100644
Binary files a/AlgoriAgent/src/agentscope/__pycache__/exception.cpython-310.pyc and b/AlgoriAgent/src/agentscope/__pycache__/exception.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/__pycache__/file_manager.cpython-310.pyc b/AlgoriAgent/src/agentscope/__pycache__/file_manager.cpython-310.pyc
index a6699ea..8ad297a 100644
Binary files a/AlgoriAgent/src/agentscope/__pycache__/file_manager.cpython-310.pyc and b/AlgoriAgent/src/agentscope/__pycache__/file_manager.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/__pycache__/logging.cpython-310.pyc b/AlgoriAgent/src/agentscope/__pycache__/logging.cpython-310.pyc
index 9c2ee62..c007dff 100644
Binary files a/AlgoriAgent/src/agentscope/__pycache__/logging.cpython-310.pyc and b/AlgoriAgent/src/agentscope/__pycache__/logging.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/__pycache__/message.cpython-310.pyc b/AlgoriAgent/src/agentscope/__pycache__/message.cpython-310.pyc
index b26e810..1871fb5 100644
Binary files a/AlgoriAgent/src/agentscope/__pycache__/message.cpython-310.pyc and b/AlgoriAgent/src/agentscope/__pycache__/message.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/__pycache__/msghub.cpython-310.pyc b/AlgoriAgent/src/agentscope/__pycache__/msghub.cpython-310.pyc
index b8379f6..375997f 100644
Binary files a/AlgoriAgent/src/agentscope/__pycache__/msghub.cpython-310.pyc and b/AlgoriAgent/src/agentscope/__pycache__/msghub.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/agents/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/agents/__pycache__/__init__.cpython-310.pyc
index 70a678c..b372256 100644
Binary files a/AlgoriAgent/src/agentscope/agents/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/agents/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/agents/__pycache__/agent.cpython-310.pyc b/AlgoriAgent/src/agentscope/agents/__pycache__/agent.cpython-310.pyc
index d044bd5..b2de94e 100644
Binary files a/AlgoriAgent/src/agentscope/agents/__pycache__/agent.cpython-310.pyc and b/AlgoriAgent/src/agentscope/agents/__pycache__/agent.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/agents/__pycache__/dialog_agent.cpython-310.pyc b/AlgoriAgent/src/agentscope/agents/__pycache__/dialog_agent.cpython-310.pyc
index 87a5af9..d9bd920 100644
Binary files a/AlgoriAgent/src/agentscope/agents/__pycache__/dialog_agent.cpython-310.pyc and b/AlgoriAgent/src/agentscope/agents/__pycache__/dialog_agent.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/agents/__pycache__/dict_dialog_agent.cpython-310.pyc b/AlgoriAgent/src/agentscope/agents/__pycache__/dict_dialog_agent.cpython-310.pyc
index 1e592ac..6ad5b3b 100644
Binary files a/AlgoriAgent/src/agentscope/agents/__pycache__/dict_dialog_agent.cpython-310.pyc and b/AlgoriAgent/src/agentscope/agents/__pycache__/dict_dialog_agent.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/agents/__pycache__/operator.cpython-310.pyc b/AlgoriAgent/src/agentscope/agents/__pycache__/operator.cpython-310.pyc
index d077370..873f5f3 100644
Binary files a/AlgoriAgent/src/agentscope/agents/__pycache__/operator.cpython-310.pyc and b/AlgoriAgent/src/agentscope/agents/__pycache__/operator.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/agents/__pycache__/rag_agent.cpython-310.pyc b/AlgoriAgent/src/agentscope/agents/__pycache__/rag_agent.cpython-310.pyc
index 3cf0dfc..5fabaaf 100644
Binary files a/AlgoriAgent/src/agentscope/agents/__pycache__/rag_agent.cpython-310.pyc and b/AlgoriAgent/src/agentscope/agents/__pycache__/rag_agent.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/agents/__pycache__/react_agent.cpython-310.pyc b/AlgoriAgent/src/agentscope/agents/__pycache__/react_agent.cpython-310.pyc
index dff44f6..8823671 100644
Binary files a/AlgoriAgent/src/agentscope/agents/__pycache__/react_agent.cpython-310.pyc and b/AlgoriAgent/src/agentscope/agents/__pycache__/react_agent.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/agents/__pycache__/rpc_agent.cpython-310.pyc b/AlgoriAgent/src/agentscope/agents/__pycache__/rpc_agent.cpython-310.pyc
index 9c9b69a..17a3287 100644
Binary files a/AlgoriAgent/src/agentscope/agents/__pycache__/rpc_agent.cpython-310.pyc and b/AlgoriAgent/src/agentscope/agents/__pycache__/rpc_agent.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/agents/__pycache__/text_to_image_agent.cpython-310.pyc b/AlgoriAgent/src/agentscope/agents/__pycache__/text_to_image_agent.cpython-310.pyc
index edec4cd..35efa0a 100644
Binary files a/AlgoriAgent/src/agentscope/agents/__pycache__/text_to_image_agent.cpython-310.pyc and b/AlgoriAgent/src/agentscope/agents/__pycache__/text_to_image_agent.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/agents/__pycache__/user_agent.cpython-310.pyc b/AlgoriAgent/src/agentscope/agents/__pycache__/user_agent.cpython-310.pyc
index dbf9653..0352247 100644
Binary files a/AlgoriAgent/src/agentscope/agents/__pycache__/user_agent.cpython-310.pyc and b/AlgoriAgent/src/agentscope/agents/__pycache__/user_agent.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/memory/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/memory/__pycache__/__init__.cpython-310.pyc
index dbc3eb9..f6d8505 100644
Binary files a/AlgoriAgent/src/agentscope/memory/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/memory/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/memory/__pycache__/memory.cpython-310.pyc b/AlgoriAgent/src/agentscope/memory/__pycache__/memory.cpython-310.pyc
index 6021c86..df244a2 100644
Binary files a/AlgoriAgent/src/agentscope/memory/__pycache__/memory.cpython-310.pyc and b/AlgoriAgent/src/agentscope/memory/__pycache__/memory.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/memory/__pycache__/temporary_memory.cpython-310.pyc b/AlgoriAgent/src/agentscope/memory/__pycache__/temporary_memory.cpython-310.pyc
index 892b46b..02f360b 100644
Binary files a/AlgoriAgent/src/agentscope/memory/__pycache__/temporary_memory.cpython-310.pyc and b/AlgoriAgent/src/agentscope/memory/__pycache__/temporary_memory.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/models/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/models/__pycache__/__init__.cpython-310.pyc
index f084595..8f715aa 100644
Binary files a/AlgoriAgent/src/agentscope/models/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/models/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/models/__pycache__/config.cpython-310.pyc b/AlgoriAgent/src/agentscope/models/__pycache__/config.cpython-310.pyc
index 317c2de..be0fc05 100644
Binary files a/AlgoriAgent/src/agentscope/models/__pycache__/config.cpython-310.pyc and b/AlgoriAgent/src/agentscope/models/__pycache__/config.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/models/__pycache__/dashscope_model.cpython-310.pyc b/AlgoriAgent/src/agentscope/models/__pycache__/dashscope_model.cpython-310.pyc
index d96a15c..a823dfc 100644
Binary files a/AlgoriAgent/src/agentscope/models/__pycache__/dashscope_model.cpython-310.pyc and b/AlgoriAgent/src/agentscope/models/__pycache__/dashscope_model.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/models/__pycache__/gemini_model.cpython-310.pyc b/AlgoriAgent/src/agentscope/models/__pycache__/gemini_model.cpython-310.pyc
index 6f8884b..39a3040 100644
Binary files a/AlgoriAgent/src/agentscope/models/__pycache__/gemini_model.cpython-310.pyc and b/AlgoriAgent/src/agentscope/models/__pycache__/gemini_model.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/models/__pycache__/litellm_model.cpython-310.pyc b/AlgoriAgent/src/agentscope/models/__pycache__/litellm_model.cpython-310.pyc
index 34e7155..ac51971 100644
Binary files a/AlgoriAgent/src/agentscope/models/__pycache__/litellm_model.cpython-310.pyc and b/AlgoriAgent/src/agentscope/models/__pycache__/litellm_model.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/models/__pycache__/model.cpython-310.pyc b/AlgoriAgent/src/agentscope/models/__pycache__/model.cpython-310.pyc
index f5dc171..d17ee41 100644
Binary files a/AlgoriAgent/src/agentscope/models/__pycache__/model.cpython-310.pyc and b/AlgoriAgent/src/agentscope/models/__pycache__/model.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/models/__pycache__/ollama_model.cpython-310.pyc b/AlgoriAgent/src/agentscope/models/__pycache__/ollama_model.cpython-310.pyc
index cc86f6b..6f7ea7a 100644
Binary files a/AlgoriAgent/src/agentscope/models/__pycache__/ollama_model.cpython-310.pyc and b/AlgoriAgent/src/agentscope/models/__pycache__/ollama_model.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/models/__pycache__/openai_model.cpython-310.pyc b/AlgoriAgent/src/agentscope/models/__pycache__/openai_model.cpython-310.pyc
index 1a116d4..b9e2f94 100644
Binary files a/AlgoriAgent/src/agentscope/models/__pycache__/openai_model.cpython-310.pyc and b/AlgoriAgent/src/agentscope/models/__pycache__/openai_model.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/models/__pycache__/post_model.cpython-310.pyc b/AlgoriAgent/src/agentscope/models/__pycache__/post_model.cpython-310.pyc
index d3c2c86..2c6c4f1 100644
Binary files a/AlgoriAgent/src/agentscope/models/__pycache__/post_model.cpython-310.pyc and b/AlgoriAgent/src/agentscope/models/__pycache__/post_model.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/models/__pycache__/response.cpython-310.pyc b/AlgoriAgent/src/agentscope/models/__pycache__/response.cpython-310.pyc
index 00b10fc..579c346 100644
Binary files a/AlgoriAgent/src/agentscope/models/__pycache__/response.cpython-310.pyc and b/AlgoriAgent/src/agentscope/models/__pycache__/response.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/models/__pycache__/zhipu_model.cpython-310.pyc b/AlgoriAgent/src/agentscope/models/__pycache__/zhipu_model.cpython-310.pyc
index b92c18e..bf695a7 100644
Binary files a/AlgoriAgent/src/agentscope/models/__pycache__/zhipu_model.cpython-310.pyc and b/AlgoriAgent/src/agentscope/models/__pycache__/zhipu_model.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/parsers/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/parsers/__pycache__/__init__.cpython-310.pyc
index fb58c2c..7e7365b 100644
Binary files a/AlgoriAgent/src/agentscope/parsers/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/parsers/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/parsers/__pycache__/code_block_parser.cpython-310.pyc b/AlgoriAgent/src/agentscope/parsers/__pycache__/code_block_parser.cpython-310.pyc
index 3dc7d15..eea1722 100644
Binary files a/AlgoriAgent/src/agentscope/parsers/__pycache__/code_block_parser.cpython-310.pyc and b/AlgoriAgent/src/agentscope/parsers/__pycache__/code_block_parser.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/parsers/__pycache__/json_object_parser.cpython-310.pyc b/AlgoriAgent/src/agentscope/parsers/__pycache__/json_object_parser.cpython-310.pyc
index 103c93e..e33a08a 100644
Binary files a/AlgoriAgent/src/agentscope/parsers/__pycache__/json_object_parser.cpython-310.pyc and b/AlgoriAgent/src/agentscope/parsers/__pycache__/json_object_parser.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/parsers/__pycache__/parser_base.cpython-310.pyc b/AlgoriAgent/src/agentscope/parsers/__pycache__/parser_base.cpython-310.pyc
index 82465d0..37ddcdb 100644
Binary files a/AlgoriAgent/src/agentscope/parsers/__pycache__/parser_base.cpython-310.pyc and b/AlgoriAgent/src/agentscope/parsers/__pycache__/parser_base.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/parsers/__pycache__/tagged_content_parser.cpython-310.pyc b/AlgoriAgent/src/agentscope/parsers/__pycache__/tagged_content_parser.cpython-310.pyc
index 984987e..3f4909b 100644
Binary files a/AlgoriAgent/src/agentscope/parsers/__pycache__/tagged_content_parser.cpython-310.pyc and b/AlgoriAgent/src/agentscope/parsers/__pycache__/tagged_content_parser.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/pipelines/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/pipelines/__pycache__/__init__.cpython-310.pyc
index aa2ce8c..6f22c33 100644
Binary files a/AlgoriAgent/src/agentscope/pipelines/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/pipelines/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/pipelines/__pycache__/functional.cpython-310.pyc b/AlgoriAgent/src/agentscope/pipelines/__pycache__/functional.cpython-310.pyc
index 3fbd44f..c93d697 100644
Binary files a/AlgoriAgent/src/agentscope/pipelines/__pycache__/functional.cpython-310.pyc and b/AlgoriAgent/src/agentscope/pipelines/__pycache__/functional.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/pipelines/__pycache__/pipeline.cpython-310.pyc b/AlgoriAgent/src/agentscope/pipelines/__pycache__/pipeline.cpython-310.pyc
index 1f23b72..c44fae9 100644
Binary files a/AlgoriAgent/src/agentscope/pipelines/__pycache__/pipeline.cpython-310.pyc and b/AlgoriAgent/src/agentscope/pipelines/__pycache__/pipeline.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/prompt/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/prompt/__pycache__/__init__.cpython-310.pyc
index 58e63a7..4c4fbb4 100644
Binary files a/AlgoriAgent/src/agentscope/prompt/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/prompt/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_comparer.cpython-310.pyc b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_comparer.cpython-310.pyc
index 4aaba5b..a560539 100644
Binary files a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_comparer.cpython-310.pyc and b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_comparer.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_engine.cpython-310.pyc b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_engine.cpython-310.pyc
index 48c176d..f7e1abb 100644
Binary files a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_engine.cpython-310.pyc and b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_engine.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_base.cpython-310.pyc b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_base.cpython-310.pyc
index d8d1a35..a8ed295 100644
Binary files a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_base.cpython-310.pyc and b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_base.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_en.cpython-310.pyc b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_en.cpython-310.pyc
index 590f2a6..3632cd9 100644
Binary files a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_en.cpython-310.pyc and b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_en.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_zh.cpython-310.pyc b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_zh.cpython-310.pyc
index bd64985..0f34626 100644
Binary files a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_zh.cpython-310.pyc and b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_generator_zh.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_optimizer.cpython-310.pyc b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_optimizer.cpython-310.pyc
index 0395ce6..178d371 100644
Binary files a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_optimizer.cpython-310.pyc and b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_optimizer.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_utils.cpython-310.pyc b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_utils.cpython-310.pyc
index 21b27e3..d4c9f27 100644
Binary files a/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_utils.cpython-310.pyc and b/AlgoriAgent/src/agentscope/prompt/__pycache__/_prompt_utils.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/rag/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/rag/__pycache__/__init__.cpython-310.pyc
index 7d4481b..1db4bd3 100644
Binary files a/AlgoriAgent/src/agentscope/rag/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/rag/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/rag/__pycache__/knowledge.cpython-310.pyc b/AlgoriAgent/src/agentscope/rag/__pycache__/knowledge.cpython-310.pyc
index e2bd247..78b45ff 100644
Binary files a/AlgoriAgent/src/agentscope/rag/__pycache__/knowledge.cpython-310.pyc and b/AlgoriAgent/src/agentscope/rag/__pycache__/knowledge.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/rag/__pycache__/knowledge_bank.cpython-310.pyc b/AlgoriAgent/src/agentscope/rag/__pycache__/knowledge_bank.cpython-310.pyc
index 285707e..f6d6162 100644
Binary files a/AlgoriAgent/src/agentscope/rag/__pycache__/knowledge_bank.cpython-310.pyc and b/AlgoriAgent/src/agentscope/rag/__pycache__/knowledge_bank.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/rag/__pycache__/llama_index_knowledge.cpython-310.pyc b/AlgoriAgent/src/agentscope/rag/__pycache__/llama_index_knowledge.cpython-310.pyc
index 7f85c37..29458f2 100644
Binary files a/AlgoriAgent/src/agentscope/rag/__pycache__/llama_index_knowledge.cpython-310.pyc and b/AlgoriAgent/src/agentscope/rag/__pycache__/llama_index_knowledge.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/rpc/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/rpc/__pycache__/__init__.cpython-310.pyc
index dd6478e..a8b69c0 100644
Binary files a/AlgoriAgent/src/agentscope/rpc/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/rpc/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_client.cpython-310.pyc b/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_client.cpython-310.pyc
index a3ad980..80eca09 100644
Binary files a/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_client.cpython-310.pyc and b/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_client.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_pb2.cpython-310.pyc b/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_pb2.cpython-310.pyc
index 5fef677..ff95202 100644
Binary files a/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_pb2.cpython-310.pyc and b/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_pb2.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_pb2_grpc.cpython-310.pyc b/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_pb2_grpc.cpython-310.pyc
index 35ccdf1..5eefddc 100644
Binary files a/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_pb2_grpc.cpython-310.pyc and b/AlgoriAgent/src/agentscope/rpc/__pycache__/rpc_agent_pb2_grpc.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/server/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/server/__pycache__/__init__.cpython-310.pyc
index 699d07a..13a9ff9 100644
Binary files a/AlgoriAgent/src/agentscope/server/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/server/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/server/__pycache__/launcher.cpython-310.pyc b/AlgoriAgent/src/agentscope/server/__pycache__/launcher.cpython-310.pyc
index 562999c..c81a470 100644
Binary files a/AlgoriAgent/src/agentscope/server/__pycache__/launcher.cpython-310.pyc and b/AlgoriAgent/src/agentscope/server/__pycache__/launcher.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/server/__pycache__/servicer.cpython-310.pyc b/AlgoriAgent/src/agentscope/server/__pycache__/servicer.cpython-310.pyc
index 61a343c..be43460 100644
Binary files a/AlgoriAgent/src/agentscope/server/__pycache__/servicer.cpython-310.pyc and b/AlgoriAgent/src/agentscope/server/__pycache__/servicer.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/__pycache__/__init__.cpython-310.pyc
index 5e4fcf1..b520b32 100644
Binary files a/AlgoriAgent/src/agentscope/service/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/__pycache__/service_response.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/__pycache__/service_response.cpython-310.pyc
index 10faaee..ad786ec 100644
Binary files a/AlgoriAgent/src/agentscope/service/__pycache__/service_response.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/__pycache__/service_response.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/__pycache__/service_status.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/__pycache__/service_status.cpython-310.pyc
index e0c8fac..6f11438 100644
Binary files a/AlgoriAgent/src/agentscope/service/__pycache__/service_status.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/__pycache__/service_status.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/__pycache__/service_toolkit.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/__pycache__/service_toolkit.cpython-310.pyc
index c3807f3..c8c371d 100644
Binary files a/AlgoriAgent/src/agentscope/service/__pycache__/service_toolkit.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/__pycache__/service_toolkit.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/__init__.cpython-310.pyc
index 23734c9..5af3fce 100644
Binary files a/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/exec_python.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/exec_python.cpython-310.pyc
index aeaad9b..881ab68 100644
Binary files a/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/exec_python.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/exec_python.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/exec_shell.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/exec_shell.cpython-310.pyc
index 01e8585..1ff3dfd 100644
Binary files a/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/exec_shell.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/execute_code/__pycache__/exec_shell.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/file/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/file/__pycache__/__init__.cpython-310.pyc
index 5c01c4c..d57333d 100644
Binary files a/AlgoriAgent/src/agentscope/service/file/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/file/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/file/__pycache__/common.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/file/__pycache__/common.cpython-310.pyc
index ea84156..169b58d 100644
Binary files a/AlgoriAgent/src/agentscope/service/file/__pycache__/common.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/file/__pycache__/common.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/file/__pycache__/json.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/file/__pycache__/json.cpython-310.pyc
index ea5afa6..493c103 100644
Binary files a/AlgoriAgent/src/agentscope/service/file/__pycache__/json.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/file/__pycache__/json.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/file/__pycache__/text.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/file/__pycache__/text.cpython-310.pyc
index e2b3c54..7278043 100644
Binary files a/AlgoriAgent/src/agentscope/service/file/__pycache__/text.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/file/__pycache__/text.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/__init__.cpython-310.pyc
index b1ae235..e234712 100644
Binary files a/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/dashscope_services.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/dashscope_services.cpython-310.pyc
index d4cba67..73e7d98 100644
Binary files a/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/dashscope_services.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/dashscope_services.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/openai_services.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/openai_services.cpython-310.pyc
index 9f02834..950604f 100644
Binary files a/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/openai_services.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/multi_modality/__pycache__/openai_services.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/__init__.cpython-310.pyc
index 06d3974..5cbde55 100644
Binary files a/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/retrieval_from_list.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/retrieval_from_list.cpython-310.pyc
index ad061c7..629ff64 100644
Binary files a/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/retrieval_from_list.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/retrieval_from_list.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/similarity.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/similarity.cpython-310.pyc
index a2728d8..af49e8a 100644
Binary files a/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/similarity.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/retrieval/__pycache__/similarity.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/__init__.cpython-310.pyc
index 4e90583..1bd3de8 100644
Binary files a/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/mongodb.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/mongodb.cpython-310.pyc
index 8ee0f14..b871810 100644
Binary files a/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/mongodb.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/mongodb.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/mysql.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/mysql.cpython-310.pyc
index 1edf15d..5c53b77 100644
Binary files a/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/mysql.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/mysql.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/sqlite.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/sqlite.cpython-310.pyc
index e8611b7..d383fa2 100644
Binary files a/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/sqlite.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/sql_query/__pycache__/sqlite.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/text_processing/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/text_processing/__pycache__/__init__.cpython-310.pyc
index 1643d73..16cfbb9 100644
Binary files a/AlgoriAgent/src/agentscope/service/text_processing/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/text_processing/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/text_processing/__pycache__/summarization.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/text_processing/__pycache__/summarization.cpython-310.pyc
index 043ba7c..1646d7a 100644
Binary files a/AlgoriAgent/src/agentscope/service/text_processing/__pycache__/summarization.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/text_processing/__pycache__/summarization.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/web/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/web/__pycache__/__init__.cpython-310.pyc
index de27a57..c8b7c90 100644
Binary files a/AlgoriAgent/src/agentscope/service/web/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/web/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/web/__pycache__/arxiv.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/web/__pycache__/arxiv.cpython-310.pyc
index 9bc5c29..62660ac 100644
Binary files a/AlgoriAgent/src/agentscope/service/web/__pycache__/arxiv.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/web/__pycache__/arxiv.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/web/__pycache__/dblp.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/web/__pycache__/dblp.cpython-310.pyc
index dd842f0..cbab202 100644
Binary files a/AlgoriAgent/src/agentscope/service/web/__pycache__/dblp.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/web/__pycache__/dblp.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/web/__pycache__/download.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/web/__pycache__/download.cpython-310.pyc
index 76df274..905d353 100644
Binary files a/AlgoriAgent/src/agentscope/service/web/__pycache__/download.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/web/__pycache__/download.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/web/__pycache__/search.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/web/__pycache__/search.cpython-310.pyc
index 518a9c8..eeae27f 100644
Binary files a/AlgoriAgent/src/agentscope/service/web/__pycache__/search.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/web/__pycache__/search.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/service/web/__pycache__/web_digest.cpython-310.pyc b/AlgoriAgent/src/agentscope/service/web/__pycache__/web_digest.cpython-310.pyc
index 1bf30bf..2253c05 100644
Binary files a/AlgoriAgent/src/agentscope/service/web/__pycache__/web_digest.cpython-310.pyc and b/AlgoriAgent/src/agentscope/service/web/__pycache__/web_digest.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/studio/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/studio/__pycache__/__init__.cpython-310.pyc
index 0ad271a..c54571c 100644
Binary files a/AlgoriAgent/src/agentscope/studio/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/studio/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/studio/__pycache__/_app.cpython-310.pyc b/AlgoriAgent/src/agentscope/studio/__pycache__/_app.cpython-310.pyc
index 6f8d1c5..9f62786 100644
Binary files a/AlgoriAgent/src/agentscope/studio/__pycache__/_app.cpython-310.pyc and b/AlgoriAgent/src/agentscope/studio/__pycache__/_app.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/studio/__pycache__/_client.cpython-310.pyc b/AlgoriAgent/src/agentscope/studio/__pycache__/_client.cpython-310.pyc
index 27f7564..794aff7 100644
Binary files a/AlgoriAgent/src/agentscope/studio/__pycache__/_client.cpython-310.pyc and b/AlgoriAgent/src/agentscope/studio/__pycache__/_client.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/utils/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/utils/__pycache__/__init__.cpython-310.pyc
index 60c0b41..a09358d 100644
Binary files a/AlgoriAgent/src/agentscope/utils/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/utils/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/utils/__pycache__/common.cpython-310.pyc b/AlgoriAgent/src/agentscope/utils/__pycache__/common.cpython-310.pyc
index 088aa7c..d138979 100644
Binary files a/AlgoriAgent/src/agentscope/utils/__pycache__/common.cpython-310.pyc and b/AlgoriAgent/src/agentscope/utils/__pycache__/common.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/utils/__pycache__/monitor.cpython-310.pyc b/AlgoriAgent/src/agentscope/utils/__pycache__/monitor.cpython-310.pyc
index c400b93..6cea868 100644
Binary files a/AlgoriAgent/src/agentscope/utils/__pycache__/monitor.cpython-310.pyc and b/AlgoriAgent/src/agentscope/utils/__pycache__/monitor.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/utils/__pycache__/token_utils.cpython-310.pyc b/AlgoriAgent/src/agentscope/utils/__pycache__/token_utils.cpython-310.pyc
index 686b04c..670e645 100644
Binary files a/AlgoriAgent/src/agentscope/utils/__pycache__/token_utils.cpython-310.pyc and b/AlgoriAgent/src/agentscope/utils/__pycache__/token_utils.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/utils/__pycache__/tools.cpython-310.pyc b/AlgoriAgent/src/agentscope/utils/__pycache__/tools.cpython-310.pyc
index 0300872..ffdddb7 100644
Binary files a/AlgoriAgent/src/agentscope/utils/__pycache__/tools.cpython-310.pyc and b/AlgoriAgent/src/agentscope/utils/__pycache__/tools.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/web/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/web/__pycache__/__init__.cpython-310.pyc
index d69998c..57e5535 100644
Binary files a/AlgoriAgent/src/agentscope/web/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/web/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/web/gradio/__pycache__/__init__.cpython-310.pyc b/AlgoriAgent/src/agentscope/web/gradio/__pycache__/__init__.cpython-310.pyc
index 2db016a..9d35ab5 100644
Binary files a/AlgoriAgent/src/agentscope/web/gradio/__pycache__/__init__.cpython-310.pyc and b/AlgoriAgent/src/agentscope/web/gradio/__pycache__/__init__.cpython-310.pyc differ
diff --git a/AlgoriAgent/src/agentscope/web/gradio/__pycache__/utils.cpython-310.pyc b/AlgoriAgent/src/agentscope/web/gradio/__pycache__/utils.cpython-310.pyc
index 87aefe1..21ceeb3 100644
Binary files a/AlgoriAgent/src/agentscope/web/gradio/__pycache__/utils.cpython-310.pyc and b/AlgoriAgent/src/agentscope/web/gradio/__pycache__/utils.cpython-310.pyc differ
diff --git a/Html/__pycache__/backboardManager.cpython-310.pyc b/Html/__pycache__/backboardManager.cpython-310.pyc
index b26aeca..638913c 100644
Binary files a/Html/__pycache__/backboardManager.cpython-310.pyc and b/Html/__pycache__/backboardManager.cpython-310.pyc differ
diff --git a/Html/app.py b/Html/app.py
index 4cb1ab0..529e251 100644
--- a/Html/app.py
+++ b/Html/app.py
@@ -58,11 +58,13 @@ class VSCodeNamespace(Namespace):
         dataconfig = data['config']
         user_id = dataconfig.get('user_id')
         path = dataconfig.get('path')
+        course_id = dataconfig.get('course_id')
+        lesson_id = dataconfig.get('chapter_id')
         print(f"User {user_id} connected with path: {path}")
         useruuid = username2uuid[user_id]
         join_room(useruuid, namespace='/vscode')
         if backboard_manager.get_backboard(useruuid) == None:
-            backboard_manager.add_backboard(useruuid, path)
+            backboard_manager.add_backboard(useruuid,user_id, course_id, lesson_id, path)
 
 
     def on_message(self, data):
@@ -98,7 +100,20 @@ def realtime_response(config, realtime_action):
         assert type(file_path) == str
         bb.pasted_file_path = file_path
         bb.pasted_content = realtime_action['content']
+        bb.active_file_path = file_path
+        with open(f"{file_path}") as f:
+            bb.active_file_content = f.read()
         
+    
+    if realtime_action['type'] == 'fileEdit':
+        file_path = realtime_action['filePath']
+        assert type(file_path) == str
+        bb.active_file_path = file_path
+        with open(f"{file_path}") as f:
+            bb.active_file_content = f.read()
+
+
+
 
     print("config"+str(config))
     print("realtime_action"+str(realtime_action))
@@ -131,10 +146,11 @@ class AgentNamespace(Namespace):
             markdown = fmd.read()
             markdown_prompts = fmdp.read()
             score_prompts = fsp.read()
-            user_uuid, agent = agent_manager.new_agent(course_id, chapter_id, markdown, markdown_prompts, score_prompts, id=user_uuid, root_path=f'{user}/{course_id}/{chapter_id}')
+            user_uuid, agent = agent_manager.new_agent(course_id, chapter_id, markdown, markdown_prompts, score_prompts, id=user_uuid, 
+                                                       root_path=f'../../study/{user}/{course_id}/{chapter_id}')
             print(user_uuid)
             join_room(user_uuid, namespace='/agent')  # 将该用户加入以user_id为名的room
-        backboard_manager.add_backboard(user_uuid, chapter_id)
+        backboard_manager.add_backboard(user_uuid, user, course_id, chapter_id, root_path=f'../../study/{user}/{course_id}/{chapter_id}')
 
     def on_language(self, language):
         id = session.get('user_id')
@@ -150,23 +166,23 @@ class AgentNamespace(Namespace):
             res = agent_manager.invoke(id, data['data'], backboard_manager.get_backboard(id).get_info_prompt())
             print('=*='*20)
             print(res.content)
-            reply = f"AI: {res.content['speak']}"
+            reply = f"{res.content['speak']}"
             with app.app_context(): 
                 emit('message',reply, room=id, namespace='/agent')
                 emit('request_function',res.content['function'], room=id, namespace='/agent')
-            # for res in agent_manager.invoke(id, data, backboard_manager.get_backboard(id).get_info_prompt(), reset=True):
-            #     print('=*='*20)
-            #     print(res.content)
-            #     if ('speak' in res.content):
-            #         reply = f"AI: {res.content['speak']}"
-            #         with app.app_context(): emit('message',reply, room=id, namespace='/agent')
-            #     if ('function' in res.content):
-            #         functions = res.content['function']
-            #         if functions is None: break
-            #         if len(functions)==0:break
+
         if data['type'] == 'function':
             agent_manager.function_call(id, data['data'])
 
+    def on_initiative(self,data):
+        print("User active function call")
+        user_id = session['user_id']
+        if data['name'] == 'sample_judge':
+            agent_manager.sample_judge(user_id, backboard_manager.get_backboard(user_id))
+        if data['name'] == 'judge':
+            agent_manager.judge(user_id, backboard_manager.get_backboard(user_id))
+
+
 
     def on_disconnect(self,data):
         print("VSCode client disconnected")
@@ -358,7 +374,7 @@ def dashboard():
     for course_id in user_id2UserClass[user_id].select_course:
         course_brief_info = course_list.get_course_brief_info(course_id, load_course_from_json(course_id, course_data_dir = COURSE_DATA_DIR))
         user_course_data.append(course_brief_info)
-    return render_template('dashboard.html',user_data = json.dumps(user_id2UserClass[user_id].__json__()), user_course_data = json.dumps(user_course_data))
+    return render_template('dashboard.html',user_data = user_id2UserClass[user_id].to_json_without_dialog(), user_course_data = json.dumps(user_course_data))
 
 '''
 Course
diff --git a/Html/backboardManager.py b/Html/backboardManager.py
index 7ebfe2d..1777f8d 100644
--- a/Html/backboardManager.py
+++ b/Html/backboardManager.py
@@ -2,27 +2,15 @@
 
 
 import time
-
-
-class BackBoardManager:
-    def __init__(self):
-        self.backboards = {} # user_id: Backboard
-    
-    def add_backboard(self, user_id, folder):
-        self.backboards[user_id] = Backboard(user_id, folder)
-
-    def get_backboard(self, user_id):
-        if user_id not in self.backboards:
-            return None
-        return self.backboards[user_id]
-
-
-
+import os
 
 class Backboard:
-    def __init__(self, user_id, folder):
+    def __init__(self, user_id, username, course_id, lesson_id, root_path):
         self.user_id = user_id
-        self.folder = folder
+        self.username = username
+        self.course_id = course_id
+        self.lesson_id = lesson_id
+        self.root_path = root_path
         self.create_time = time.time()
         self.enter_chapter_time = time.time()
         self.history = []
@@ -48,12 +36,12 @@ class Backboard:
         f"- User's total study time is {self.get_deltatime_mmss(self.create_time, time.time())}{endl}"
         f"- User's current chapter study time is {self.get_deltatime_mmss(self.enter_chapter_time, time.time())}{endl}"
         f"- Activated file path: {self.active_file_path}{endl}"
-        f"- Activated file content (current editing 10 lines): {endl}"
         f"```{endl}"
         f"{self.active_file_content[-10:]}{endl}"
         f"```{endl}"
         f"- Last five action:{five_history}{endl}"
         f"- File tree: {self.file_tree}{endl}")
+        # f"- Activated file content (current editing 10 lines): {endl}"
     def add_history(self, realtime_action):
         if(realtime_action['type']=='config'):return
         if len(self.history)!=0:
@@ -72,3 +60,21 @@ class Backboard:
         return f"{hours:02d}:{minutes:02d}:{seconds:02d}"
         
 
+    def get_active_file_reletive_path(self):
+        relative_path = os.path.relpath(self.active_file_path, self.root_path)
+        return relative_path
+    
+
+
+class BackBoardManager:
+    def __init__(self):
+        self.backboards = {} # user_id: Backboard
+    
+    def add_backboard(self, user_id, username, course_id, lesson_id, root_path):
+        self.backboards[user_id] = Backboard(user_id,username, course_id, lesson_id, root_path)
+
+    def get_backboard(self, user_id) -> Backboard:
+        return self.backboards[user_id]
+
+
+
diff --git a/Html/db/data/course/CS101.json b/Html/db/data/course/CS101.json
index 8135c4a..5c7ba5e 100644
--- a/Html/db/data/course/CS101.json
+++ b/Html/db/data/course/CS101.json
@@ -2,7 +2,7 @@
     "course_id": "CS101",
     "course_name": "Computer Science 101",
     "course_auther": "John Doe",
-    "course_img_path": "/static/image/algorithm/book_cover.png",
+    "course_img_path": "/static/image/CS101/CS101.png",
     "description":"这个老师很懒，没有写介绍",
     "lessons": [
         {
@@ -11,7 +11,7 @@
             "markdown": "This is the introduction lesson.",
             "markdown_prompt": "What is Computer Science?",
             "score_prompt": "Please provide an overview of CS.",
-            "lesson_img_path": "/static/image/algorithm/book_cover.png"
+            "lesson_img_path": "/static/image/CS101/CS101.png"
         },
         {
             "lesson_id": "L002",
@@ -19,7 +19,7 @@
             "markdown": "In this lesson, we cover Data Structures.",
             "markdown_prompt": "What is an array?",
             "score_prompt": "Explain how linked lists work.",
-            "lesson_img_path": "/static/image/algorithm/book_cover.png"
+            "lesson_img_path": "/static/image/CS101/CS101.png"
         }
     ]
 }
diff --git a/Html/db/data/course/course_list.json b/Html/db/data/course/course_list.json
index 9f695a2..5988d99 100644
--- a/Html/db/data/course/course_list.json
+++ b/Html/db/data/course/course_list.json
@@ -3,7 +3,7 @@
         "course_id": "CS101",
         "course_name": "Computer Science 101",
         "course_description":"这个老师很懒，没有写介绍",
-        "course_img_path": "/static/image/algorithm/book_cover.png",
+        "course_img_path": "/static/image/CS101/CS101.png",
         "course_create_date": "2023-01-01",
         "course_update_data": "2023-01-01"},
 
@@ -19,7 +19,7 @@
         "course_id": "data-structure",
         "course_name": "数据结构",
         "course_description":"这个老师很懒，没有写介绍",
-        "course_img_path": "/static/image/algorithm/book_cover.png",
+        "course_img_path": "/static/image/data-structure/book_cover-data-structure.png",
         "course_create_date": "2023-01-01",
         "course_update_data": "2023-01-01"}
 }
\ No newline at end of file
diff --git a/Html/db/data/course/data-structure.json b/Html/db/data/course/data-structure.json
index ce96ac9..b1326b8 100644
--- a/Html/db/data/course/data-structure.json
+++ b/Html/db/data/course/data-structure.json
@@ -2,7 +2,7 @@
     "course_id": "data-structure",
     "course_name": "data-structure",
     "course_auther": "Chao Peng",
-    "course_img_path": "/static/image/algorithm/book_cover.png",
+    "course_img_path": "/static/image/data-structure/book_cover-data-structure.png",
     "description":"这个老师很懒，没有写介绍",
     "lessons": [
         {
@@ -11,7 +11,7 @@
             "markdown": "This is the introduction lesson.",
             "markdown_prompt": "What is Data structure?",
             "score_prompt": "Please provide an overview of CS.",
-            "lesson_img_path": "/static/image/algorithm/book_cover.png"
+            "lesson_img_path": "/static/image/data-structure/book_cover-data-structure.png"
         },
         {
             "lesson_id": "L002",
@@ -19,7 +19,7 @@
             "markdown": "In this lesson, we cover Data Structures.",
             "markdown_prompt": "What is an array?",
             "score_prompt": "Explain how linked lists work.",
-            "lesson_img_path": "/static/image/algorithm/book_cover.png"
+            "lesson_img_path": "/static/image/data-structure/book_cover-data-structure.png"
         }
     ]
 }
diff --git a/Html/db/data/user/cake.json b/Html/db/data/user/cake.json
index 67e05bf..354c128 100644
--- a/Html/db/data/user/cake.json
+++ b/Html/db/data/user/cake.json
@@ -15,17 +15,17 @@
                             "title": "\u5f15\u5165",
                             "dialog": [
                                 {
-                                    "id": "0b26417b9a4a41c3a86a63e96ac279f7",
-                                    "timestamp": "2025-01-01 01:40:36",
+                                    "id": "c22798c82f664930aa74a06834e4310a",
+                                    "timestamp": "2025-01-01 18:47:30",
                                     "name": "system",
-                                    "content": "You're a helpful assistant.\n\n## \u4f60\u7684\u89d2\u8272\uff1a\n\u4f60\u9700\u8981\u5e2e\u52a9\u5b66\u751f\u5b66\u4e60\u7b97\u6cd5\uff0c\u5b8c\u6210\u7279\u5b9a\u7684\u7f16\u7a0b\u4efb\u52a1\u3002\n\n## \u4f60\u9700\u8981\u505a\u7684\uff1a\n0. \u4f60\u5fc5\u987b\u6839\u636e\u8001\u5e08\u63d0\u4f9b\u7684\u6559\u5b66\u5927\u7eb2\uff0c\u6309\u7167\u7ae0\u8282\u987a\u5e8f\uff0c\u4f9d\u6b21\u5b66\u4e60\u3002\n1. \u5b66\u751f\u53ef\u80fd\u4f1a\u63d0\u51fa\u4e00\u4e9b\u95ee\u9898\uff0c\u4f60\u65e0\u9700\u56de\u7b54\u95ee\u9898\uff0c\u800c\u662f\u6839\u636e\u4e0a\u4e0b\u6587\uff08\u6bd4\u5982\u5b66\u751f\u81ea\u5df1\u7684\u4ee3\u7801\uff09\u6765\u5206\u6790\uff0c\u5e76\u7ed9\u51fa\u4e00\u6b65\u7684\u89e3\u51b3\u65b9\u6848\u3002\n2. \u6559\u5b66\u5927\u7eb2\u4e2d\u4f1a\u6307\u660e\u6bcf\u4e2a\u7ae0\u8282\u7684\u5b66\u4e60\u76ee\u6807\uff0c\u4f60\u6309\u7167\u76ee\u6807\u5b66\u4e60\u3002\n3. \u7cfb\u7edf\u4f1aoccasionally gather\u5b66\u751f\u4fe1\u606f\uff0c\u4f60\u9700\u8981\u5206\u6790\u5b66\u751f\u5b66\u4e60\u8bb0\u5f55\uff0c\u8c03\u7528\u5fc5\u8981\u7684\u51fd\u6570\u3002\n\n## \u6ce8\u610f\u4e8b\u9879\uff1a\n1. \u786e\u4fdd\u4f7f\u7528\u5de5\u5177\u51fd\u6570\u65f6\uff0c\u63d0\u4f9b\u7684\u53c2\u6570\u7c7b\u578b\u548c\u503c\u662f\u6b63\u786e\u7684\u3002\n2. \u4e0d\u8981\u592a\u76f8\u4fe1\u81ea\u5df1\uff0c\u4f8b\u5982\uff0c\u5f53\u524d\u4f4d\u7f6e\uff0c\u5f53\u524d\u65f6\u95f4\u7b49\uff0c\u4f60\u53ef\u4ee5\u5c1d\u8bd5\u4f7f\u7528\u5de5\u5177\u51fd\u6570\u83b7\u53d6\u4fe1\u606f\u3002\n3. \u5982\u679c\u51fd\u6570\u6267\u884c\u5931\u8d25\uff0c\u4f60\u9700\u8981\u5206\u6790\u9519\u8bef\u5e76\u5c1d\u8bd5\u89e3\u51b3\u3002\n4. \u786e\u4fdd\u5b66\u751f\u63d0\u51fa\u7684\u95ee\u9898\u6216\u60f3\u6cd5\u662f\u5426\u4e0e\u5f53\u524d\u7ae0\u8282\u76f8\u5173\uff0c\u5982\u679c\u4e0d\u76f8\u5173\uff0c\u4f60\u5e94\u7b54\u5e76\u63d0\u9192\u5b66\u751f\u5173\u6ce8\u5f53\u524d\u7ae0\u8282\u3002\n\n## \u8d44\u6e90\uff1a\n1. \u4f60\u53ef\u4ee5\u4f7f\u7528\u7684\u5de5\u5177\u51fd\u6570\u3002\n2. \u4e0a\u4e0b\u6587\uff0c\u5305\u62ec\u5b66\u751f\u7684\u4ee3\u7801\uff0c\u548c\u6559\u5b66\u5927\u7eb2\u3002\n",
+                                    "content": "You're a helpful assistant.\n\n## \u4f60\u7684\u89d2\u8272\uff1a\n\u4f60\u9700\u8981\u5e2e\u52a9\u5b66\u751f\u5b66\u4e60\u7b97\u6cd5\uff0c\u68c0\u9a8c\u5b66\u4e60\u8fc7\u7a0b\uff0c\u786e\u4fdd\u5b66\u751f\u7406\u89e3\u5f53\u524d\u7ae0\u8282\u7684\u5185\u5bb9\u540e\uff0c\u5e26\u9886\u5b66\u751f\u8fdb\u5165\u4e0b\u4e00\u7ae0\n\n## \u4f60\u9700\u8981\u505a\u7684\uff1a\n0. \u4f60\u5fc5\u987b\u6839\u636e\u8001\u5e08\u63d0\u4f9b\u7684\u6559\u5b66\u5927\u7eb2\uff0c\u6309\u7167\u7ae0\u8282\u987a\u5e8f\uff0c\u4f9d\u6b21\u5b66\u4e60\u3002\n1. \u5b66\u751f\u53ef\u80fd\u4f1a\u63d0\u51fa\u4e00\u4e9b\u95ee\u9898\uff0c\u4f60\u65e0\u9700\u56de\u7b54\u95ee\u9898\uff0c\u800c\u662f\u6839\u636e\u4e0a\u4e0b\u6587\uff08\u6bd4\u5982\u5b66\u751f\u81ea\u5df1\u7684\u4ee3\u7801\uff09\u6765\u5206\u6790\uff0c\u5e76\u7ed9\u51fa\u4e00\u6b65\u7684\u89e3\u51b3\u65b9\u6848\u3002\n2. \u6559\u5b66\u5927\u7eb2\u4e2d\u4f1a\u6307\u660e\u6bcf\u4e2a\u7ae0\u8282\u7684\u5b66\u4e60\u76ee\u6807\uff0c\u4f60\u6309\u7167\u76ee\u6807\u5b66\u4e60\u3002\n3. \u7cfb\u7edf\u4f1aoccasionally gather\u5b66\u751f\u4fe1\u606f\uff0c\u4f60\u9700\u8981\u5206\u6790\u5b66\u751f\u5b66\u4e60\u8bb0\u5f55\uff0c\u8c03\u7528\u5fc5\u8981\u7684\u51fd\u6570\u3002\n\n## \u6ce8\u610f\u4e8b\u9879\uff1a\n1. \u786e\u4fdd\u4f7f\u7528\u5de5\u5177\u51fd\u6570\u65f6\uff0c\u63d0\u4f9b\u7684\u53c2\u6570\u7c7b\u578b\u548c\u503c\u662f\u6b63\u786e\u7684\u3002\n2. \u4e0d\u8981\u592a\u76f8\u4fe1\u81ea\u5df1\uff0c\u4f8b\u5982\uff0c\u5f53\u524d\u4f4d\u7f6e\uff0c\u5f53\u524d\u65f6\u95f4\u7b49\uff0c\u4f60\u53ef\u4ee5\u5c1d\u8bd5\u4f7f\u7528\u5de5\u5177\u51fd\u6570\u83b7\u53d6\u4fe1\u606f\u3002\n3. \u5982\u679c\u51fd\u6570\u6267\u884c\u5931\u8d25\uff0c\u4f60\u9700\u8981\u5206\u6790\u9519\u8bef\u5e76\u5c1d\u8bd5\u89e3\u51b3\u3002\n4. \u786e\u4fdd\u5b66\u751f\u63d0\u51fa\u7684\u95ee\u9898\u6216\u60f3\u6cd5\u662f\u5426\u4e0e\u5f53\u524d\u7ae0\u8282\u76f8\u5173\uff0c\u5982\u679c\u4e0d\u76f8\u5173\uff0c\u4f60\u5e94\u7b54\u5e76\u63d0\u9192\u5b66\u751f\u5173\u6ce8\u5f53\u524d\u7ae0\u8282\u3002\n\n## \u8d44\u6e90\uff1a\n1. \u4f60\u53ef\u4ee5\u4f7f\u7528\u7684\u5de5\u5177\u51fd\u6570\u3002\n2. \u4e0a\u4e0b\u6587\uff0c\u5305\u62ec\u5b66\u751f\u7684\u4ee3\u7801\uff0c\u548c\u6559\u5b66\u5927\u7eb2\u3002\n",
                                     "role": "system",
                                     "url": null,
                                     "metadata": null
                                 },
                                 {
-                                    "id": "461938c9265547c9bd0187ece94d659d",
-                                    "timestamp": "2025-01-01 01:40:43",
+                                    "id": "b5f28da4edf74912878f3b0092f8db2f",
+                                    "timestamp": "2025-01-01 18:47:42",
                                     "name": "system",
                                     "content": "##The Chapter Chain##\n(Your Now Chapter) Chapter 1: \n\u4e8c\u5206\u662f\u4e00\u4e2a\u5f88\u7b80\u5355\u57fa\u7840\uff0c\u4f46\u5f88\u91cd\u8981\u7684\u77e5\u8bc6\u70b9\uff0c\u4e3a\u4ee5\u540e\u8bb8\u591a\u9ad8\u7ea7\u7684\u6570\u636e\u7ed3\u6784\u4e0e\u7b97\u6cd5\u94fa\u57ab\u3002\n\n\u4e0b\u9762\u662f\u4e00\u4e2a\u7528\u4e8c\u5206\u7684\u7b80\u5355\u573a\u666f\uff1a\n\n\u5047\u8bbe\u5c0f\u660e\u4ece0\u52301000\u4e4b\u95f4\u9009\u62e9\u4e86\u4e00\u4e2a\u6570\u5b57\u4f46\u4e0d\u544a\u8bc9\u4f60\uff0c\u4f60\u53ef\u4ee5\u4e0d\u65ad\u731c\u6d4b\u8fd9\u4e2a\u6570\uff0c\u6bcf\u6b21\u731c\u6d4b\u5c0f\u660e\u4f1a\u544a\u77e5\u4f60\u7684\u731c\u6d4b\u5f97\u8fc7\u5927\u8fd8\u662f\u8fc7\u5c0f\uff0c\u95ee\u6700\u591a\u51e0\u6b21\u5c31\u4e00\u5b9a\u80fd\u731c\u4e2d\uff1f\n\n\u7b54\u6848\u662f\u5229\u7528\u4e8c\u5206\u67e5\u627e\u7684\u539f\u7406\uff0c\u731c\u6d4b11\u6b21\u5373\u53ef\u3002\n\n1. \u5bf9\u4e8e0\u52301000\u7684\u7b54\u6848\u5907\u9009\u533a\uff0c\u731c\u6d4b\u4e2d\u4f4d\u6570500\uff0c\u5047\u8bbe\u8fc7\u5c0f\uff0c\n2. \u5219\u5bf9\u4e8e501\u52301000\u7684\u7b54\u6848\u5907\u9009\u533a\uff0c\u731c\u6d4b750\uff0c\u5047\u8bbe\u8fc7\u5927\n3. \u5219\u5bf9\u4e8e501\u5230749\u7684\u7b54\u6848\u5907\u9009\u533a\uff0c\u731c\u6d4b625\uff0c\u5047\u8bbe\u8fc7\u5c0f\uff0c\n4. \u5219\u5bf9\u4e8e626\u5230749\u533a\u95f4......\n5. \uff08688-749\uff09\n6. \uff08718-749\uff09\n7. \uff08734-749\uff09\n8. \uff08742-749\uff09\n9. \uff08746-749\uff09\n10. \uff08748-749\uff09\n11. \uff08749-749\uff09\n\n\u5728\u6700\u5dee\u7684\u60c5\u51b5\u4e0b\uff0c\u7b2c11\u6b21\u7684\u7b54\u6848\u5907\u9009\u533a\u5c31\u4e00\u5b9a\u957f\u5ea6\u4e3a1\u4e86\uff0c\u4e5f\u5c31\u662f\u5fc5\u7136\u662f\u7b54\u6848\u3002\n\n\u56e0\u6b64\u5982\u679c\u5e8f\u5217\u662f\u6709\u5e8f\u7684\uff0c\u5c31\u53ef\u4ee5\u901a\u8fc7\u4e8c\u5206\u67e5\u627e\u5feb\u901f\u5b9a\u4f4d\u6240\u9700\u8981\u7684\u6570\u636e\u3002\n\n#### \u601d\u8003\u9898\uff08\u8be2\u95eeAgent\u4ee5\u5b66\u4e60\u8ba1\u7b97\u65b9\u6cd5\uff0c\u6216\u9a8c\u8bc1\u4f60\u7684\u7b54\u6848\uff09\n\n\u5bf9\u4e8e\u4e0a\u9762\u90a3\u4e2a\u9898\u76ee\uff0c\u5982\u679c\u95ee\u9898\u533a\u95f4\u662f1\u52304000\uff0c\u6700\u5dee\u60c5\u51b5\u4e0b\u9700\u8981\u731c\u6d4b\u51e0\u6b21\uff1f\n\n\nHere are some concrete instructions:\n\n\u672c\u5c0f\u8282\u8fdb\u884c\u4e8c\u5206\u67e5\u627e\u7684\u5f15\u5165\uff0c\u9996\u5148\u5e2e\u52a9\u5b66\u751f\u7406\u89e3\u6559\u6848\u5f15\u5165\u7ae0\u8282\u7684\u6545\u4e8b\u3002\n\n\u786e\u4fdd\u5b66\u751f\u7406\u89e3\u4e86\u5f15\u5165\u540e\uff0c\u5728\u8fdb\u884c\u601d\u8003\u9898\u3002\n\n\u5982\u679c\u4e0d\u8fdb\u884c\u601d\u8003\u9898\uff0c\u5219\u8be2\u95ee\u4e00\u4e0b\u662f\u5426\u4e0d\u8fdb\u884c\u601d\u8003\u9898\uff0c\u786e\u8ba4\u540e\u8fdb\u5165\u4e0b\u4e00\u8282\u3002\n\n\u5982\u679c\u8fdb\u884c\u601d\u8003\u9898\uff0c\u544a\u77e5\u5b66\u751f\u8ba1\u7b97\u65b9\u6cd5\u5e76\u9a8c\u8bc1\u7b54\u6848\uff08\u7b54\u6848\u662f13\u6b21\uff09\n\n\u8ba1\u7b97\u65b9\u6cd5\uff1a\n\n4000\u662f\u521d\u59cb\u7b54\u6848\u533a\u95f4\u957f\u5ea6\uff0c\u6bcf\u6b21\u8be2\u95ee\u80fd\u591f\u6392\u9664\u4e00\u534a\u7684\u533a\u95f4\uff0c\u5f53\u533a\u95f4\u957f\u5ea6\u5c0f\u4e8e\u7b49\u4e8e1\u65f6\uff0c\u518d\u8fdb\u884c\u4e00\u6b21\u731c\u6d4b\u5c31\u4e00\u5b9a\u662f\u7b54\u6848\u3002\n\n4000\u4e0d\u65ad\u9664\u4ee52\uff0c\u966412\u6b21\u5c31\u5c0f\u4e8e\u7b49\u4e8e1\u4e86\uff0c\u518d\u52a0\u4e00\u6b21\u5c31\u4e00\u5b9a\u662f\u7b54\u6848\u3002\u6700\u540e\u662f12+1=13\u6b21\u3002\n\n\uff08\u6bd44000\u5927\u7684\u6700\u5c0f2\u7684\u6b21\u65b9\u6570\uff0c4096\u5c31\u662f2\u768412\u6b21\u65b9\uff0c\u8fd9\u91cc\u768412\u5c31\u662f\u7b54\u6848\u4e2d\u768412\uff0c\u6216\u8005\u5199\u4f5c\"\u4e0a\u53d6\u6574(log2(4000))\")\n\n\u8be2\u95ee\u5b66\u751f\uff0c\u5f53\u5b66\u751f\u544a\u77e5\u7406\u89e3\u540e\uff0c\u8fdb\u5165\u4e0b\u4e00\u8282\u3002\n\n\n\n\n",
                                     "role": "system",
@@ -33,8 +33,8 @@
                                     "metadata": null
                                 },
                                 {
-                                    "id": "4d63161cb43344d79a2a27abde4a4439",
-                                    "timestamp": "2025-01-01 01:40:43",
+                                    "id": "23ac775d2b6446368059bba9636ffbdf",
+                                    "timestamp": "2025-01-01 18:47:42",
                                     "name": "user",
                                     "content": "\u7b54\u6848\u662f13\uff0clog2 4000+1",
                                     "role": "user",
@@ -42,12 +42,12 @@
                                     "metadata": null
                                 },
                                 {
-                                    "id": "af2e4a0d68f34e61b977ee54f72acf63",
-                                    "timestamp": "2025-01-01 01:40:46",
-                                    "name": "user_342ee332-16c3-4722-a7aa-fd020088aba1",
+                                    "id": "50dc4cfcf0bd4eb7a47d818f5d7456ac",
+                                    "timestamp": "2025-01-01 18:47:45",
+                                    "name": "user_4eb13cc2-52ab-4281-ad32-83f1ccccadf7",
                                     "content": {
-                                        "thought": "\u5b66\u751f\u7406\u89e3\u4e86\u601d\u8003\u9898\u7684\u7b54\u6848\uff0c\u4e0b\u4e00\u6b65\u8fdb\u5165\u4e0b\u4e00\u8282\u3002",
-                                        "speak": "\u597d\u7684\uff0c\u4f60\u7684\u8ba1\u7b97\u65b9\u6cd5\u662f\u6b63\u786e\u7684\uff0c\u6700\u5dee\u60c5\u51b5\u4e0b\u9700\u8981\u731c\u6d4b13\u6b21\u3002\u6211\u4eec\u73b0\u5728\u53ef\u4ee5\u8fdb\u5165\u4e0b\u4e00\u8282\u3002",
+                                        "thought": "\u5b66\u751f\u5df2\u7ecf\u7406\u89e3\u4e86\u4e8c\u5206\u67e5\u627e\u7684\u601d\u8003\u9898\u548c\u8ba1\u7b97\u65b9\u6cd5\uff0c\u786e\u8ba4\u53ef\u4ee5\u8fdb\u5165\u4e0b\u4e00\u7ae0\u3002",
+                                        "speak": "\u4f60\u5df2\u7ecf\u7406\u89e3\u4e86\u8fd9\u4e2a\u95ee\u9898\u7684\u8ba1\u7b97\u65b9\u6cd5\uff0c\u7b54\u6848\u662f13\u6b21\u3002\u6211\u4eec\u53ef\u4ee5\u8fdb\u5165\u4e0b\u4e00\u8282\u5417\uff1f",
                                         "function": "[{\"name\": \"next_chapter\", \"arguments\": {}}]"
                                     },
                                     "role": "assistant",
@@ -58,6 +58,290 @@
                             "score": 10,
                             "is_rebuttal": false,
                             "rebuttal_score": 0
+                        },
+                        {
+                            "title": "\u7ec3\u4e60\uff1a\u4e8c\u5206\u67e5\u627e",
+                            "dialog": [
+                                {
+                                    "id": "ee78b4c3bc194a57b447eb6409bafd1f",
+                                    "timestamp": "2025-01-01 18:47:50",
+                                    "name": "system",
+                                    "content": "1. Execute function next_chapter\n   [ARGUMENTS]:\n       \n   [STATUS]: SUCCESS\n   [RESULT]: \u6210\u529f\u8fdb\u5165\u4e0b\u4e00\u7ae0\n",
+                                    "role": "system",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "c22798c82f664930aa74a06834e4310a",
+                                    "timestamp": "2025-01-01 18:47:30",
+                                    "name": "system",
+                                    "content": "You're a helpful assistant.\n\n## \u4f60\u7684\u89d2\u8272\uff1a\n\u4f60\u9700\u8981\u5e2e\u52a9\u5b66\u751f\u5b66\u4e60\u7b97\u6cd5\uff0c\u68c0\u9a8c\u5b66\u4e60\u8fc7\u7a0b\uff0c\u786e\u4fdd\u5b66\u751f\u7406\u89e3\u5f53\u524d\u7ae0\u8282\u7684\u5185\u5bb9\u540e\uff0c\u5e26\u9886\u5b66\u751f\u8fdb\u5165\u4e0b\u4e00\u7ae0\n\n## \u4f60\u9700\u8981\u505a\u7684\uff1a\n0. \u4f60\u5fc5\u987b\u6839\u636e\u8001\u5e08\u63d0\u4f9b\u7684\u6559\u5b66\u5927\u7eb2\uff0c\u6309\u7167\u7ae0\u8282\u987a\u5e8f\uff0c\u4f9d\u6b21\u5b66\u4e60\u3002\n1. \u5b66\u751f\u53ef\u80fd\u4f1a\u63d0\u51fa\u4e00\u4e9b\u95ee\u9898\uff0c\u4f60\u65e0\u9700\u56de\u7b54\u95ee\u9898\uff0c\u800c\u662f\u6839\u636e\u4e0a\u4e0b\u6587\uff08\u6bd4\u5982\u5b66\u751f\u81ea\u5df1\u7684\u4ee3\u7801\uff09\u6765\u5206\u6790\uff0c\u5e76\u7ed9\u51fa\u4e00\u6b65\u7684\u89e3\u51b3\u65b9\u6848\u3002\n2. \u6559\u5b66\u5927\u7eb2\u4e2d\u4f1a\u6307\u660e\u6bcf\u4e2a\u7ae0\u8282\u7684\u5b66\u4e60\u76ee\u6807\uff0c\u4f60\u6309\u7167\u76ee\u6807\u5b66\u4e60\u3002\n3. \u7cfb\u7edf\u4f1aoccasionally gather\u5b66\u751f\u4fe1\u606f\uff0c\u4f60\u9700\u8981\u5206\u6790\u5b66\u751f\u5b66\u4e60\u8bb0\u5f55\uff0c\u8c03\u7528\u5fc5\u8981\u7684\u51fd\u6570\u3002\n\n## \u6ce8\u610f\u4e8b\u9879\uff1a\n1. \u786e\u4fdd\u4f7f\u7528\u5de5\u5177\u51fd\u6570\u65f6\uff0c\u63d0\u4f9b\u7684\u53c2\u6570\u7c7b\u578b\u548c\u503c\u662f\u6b63\u786e\u7684\u3002\n2. \u4e0d\u8981\u592a\u76f8\u4fe1\u81ea\u5df1\uff0c\u4f8b\u5982\uff0c\u5f53\u524d\u4f4d\u7f6e\uff0c\u5f53\u524d\u65f6\u95f4\u7b49\uff0c\u4f60\u53ef\u4ee5\u5c1d\u8bd5\u4f7f\u7528\u5de5\u5177\u51fd\u6570\u83b7\u53d6\u4fe1\u606f\u3002\n3. \u5982\u679c\u51fd\u6570\u6267\u884c\u5931\u8d25\uff0c\u4f60\u9700\u8981\u5206\u6790\u9519\u8bef\u5e76\u5c1d\u8bd5\u89e3\u51b3\u3002\n4. \u786e\u4fdd\u5b66\u751f\u63d0\u51fa\u7684\u95ee\u9898\u6216\u60f3\u6cd5\u662f\u5426\u4e0e\u5f53\u524d\u7ae0\u8282\u76f8\u5173\uff0c\u5982\u679c\u4e0d\u76f8\u5173\uff0c\u4f60\u5e94\u7b54\u5e76\u63d0\u9192\u5b66\u751f\u5173\u6ce8\u5f53\u524d\u7ae0\u8282\u3002\n\n## \u8d44\u6e90\uff1a\n1. \u4f60\u53ef\u4ee5\u4f7f\u7528\u7684\u5de5\u5177\u51fd\u6570\u3002\n2. \u4e0a\u4e0b\u6587\uff0c\u5305\u62ec\u5b66\u751f\u7684\u4ee3\u7801\uff0c\u548c\u6559\u5b66\u5927\u7eb2\u3002\n",
+                                    "role": "system",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "f7a18a7080aa4dc7a4d8feb584d04e3d",
+                                    "timestamp": "2025-01-01 18:47:50",
+                                    "name": "system",
+                                    "content": "##The Chapter Chain##\n(Finished Chapter) Chapter 1: Hidden due to finished.\n(Your Now Chapter) Chapter 2: \n\u8bd5\u8bd5\u5bf9\u4e8e\u4e0b\u9762\u7684\u9898\u76ee\uff0c\u7528\u4ee3\u7801\u5b9e\u73b0\u4e00\u4e0b\u4e8c\u5206\u67e5\u627e\u3002\n\n#### \u9898\u76ee\uff1a\u6709\u5e8f\u6570\u7ec4\u5bfb\u5740\n\n\u7ed9\u51fa\u4e00\u4e2a\u957f\u5ea6\u4e3an\u7684\u6709\u5e8f\u6570\u7ec4\uff08\u4ece\u5c0f\u5230\u5927\uff09\uff0c\u6709q\u6b21\u8be2\u95ee\uff0c\u5bf9\u4e8e\u6bcf\u6b21\u8be2\u95ee\uff0c\u8f93\u51fa\u6307\u5b9a\u6570\u5728\u6570\u7ec4\u4e2d\u7684\u4e0b\u6807\u3002\u5982\u679c\u4e0d\u5b58\u5728\u5219\u8f93\u51fa-1\u3002\n\n##### \u8f93\u5165\n\n\u7b2c\u4e00\u884c\u4e00\u4e2a\u6574\u6570n\u3002(1<=n<=10^5)\n\n\u7b2c\u4e8c\u884cn\u4e2a\u7528\u7a7a\u683c\u5206\u5f00\u7684\u6574\u6570ai\u3002(0<=ai<=10^8)\n\n\u7b2c\u4e09\u884c\u4e00\u4e2a\u6574\u6570q\uff0c\u8868\u793a\u8be2\u95ee\u7684\u6b21\u6570\u3002(1<=q<=10^4)\n\n\u540eq\u884c\uff0c\u6bcf\u884c\u4e00\u4e2a\u6574\u6570b\uff0c\u8868\u793a\u8be2\u95ee\u7684\u6570\u3002(0<=b<=10^8)\n\n##### \u8f93\u51fa\n\nq\u884c\uff0c\u6bcf\u884c\u4e00\u4e2a\u6574\u6570\uff0c\u5bf9\u5e94\u6bcf\u6b21\u8be2\u95ee\u7684\u8fd4\u56de\u7ed3\u679c\u3002\n\n##### \u63d0\u793a\uff1a\n\n\u5b8c\u6210\u4ee3\u7801\u540e\uff0c\u901a\u77e5Agent\u8fdb\u884c\u8bc4\u6d4b\u3002\n\n\u5982\u679c\u4f60\u8fd8\u4e0d\u5b8c\u5168\u4f1a\u8fd9\u4e2a\u7b97\u6cd5\uff0c\u8be2\u95eeAgent\u83b7\u53d6\u63d0\u793a\u5e76\u8fdb\u884c\u5b66\u4e60\u3002\n\n\nHere are some concrete instructions:\n\n\u5982\u679c\u5b66\u751f\u5728\u5b9e\u73b0\u7684\u8fc7\u7a0b\u4e2d\u9047\u5230\u95ee\u9898\uff0c\u4e0d\u8981\u76f4\u63a5\u544a\u77e5\u5b8c\u6574\u4ee3\u7801\u5982\u4f55\u7f16\u5199\uff0c\u8be2\u95ee\u5176\u5f53\u524d\u9047\u5230\u7684\u662f\u4ec0\u4e48\u95ee\u9898\uff0c\u5e76\u5e2e\u52a9\u5b66\u751f\u4e00\u6b65\u6b65\u5b8c\u6210\u9898\u76ee\u3002\n\n\u5b9e\u73b0\u4ee3\u7801\u540e\uff0c\u5168\u90e8\u901a\u8fc7\u6d4b\u8bd5\u6570\u636e\u540e\u624d\u53ef\u4ee5\u8fdb\u5165\u4e0b\u4e00\u8282\u3002\n\n\u672c\u4f53\u6d4b\u8bd5\u70b9\u6587\u4ef6\u5939\u540d `twice_split\u3002`\n\n\n\n\n\n",
+                                    "role": "system",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "55a91eb47a8e4f6a90f72c5b8adb62e7",
+                                    "timestamp": "2025-01-01 18:47:50",
+                                    "name": "system",
+                                    "content": "\u547d\u4ee4\u6267\u884c\u7ed3\u679c\u5982\u4e0a\uff0c\u8bf7\u603b\u7ed3\u5e76\u5f15\u5bfc\u5b66\u751f\u8fdb\u884c\u4e0b\u4e00\u6b65\u64cd\u4f5c",
+                                    "role": "system",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "ddace8fbc1ca47dca894c3329b140a9f",
+                                    "timestamp": "2025-01-01 18:47:52",
+                                    "name": "user_4eb13cc2-52ab-4281-ad32-83f1ccccadf7",
+                                    "content": {
+                                        "thought": "\u5b66\u751f\u9700\u8981\u5b9e\u73b0\u4e8c\u5206\u67e5\u627e\uff0c\u5e76\u5728\u5b9e\u73b0\u8fc7\u7a0b\u4e2d\u53ef\u80fd\u4f1a\u9047\u5230\u4e00\u4e9b\u95ee\u9898\u3002\u6211\u5e94\u8be5\u5f15\u5bfc\u4ed6\u4eec\u9010\u6b65\u5b8c\u6210\u8fd9\u4e2a\u7b97\u6cd5\u3002",
+                                        "speak": "\u8bf7\u5148\u5c1d\u8bd5\u5b9e\u73b0\u4e8c\u5206\u67e5\u627e\u7684\u903b\u8f91\u3002\u5982\u679c\u5728\u5b9e\u73b0\u8fc7\u7a0b\u4e2d\u9047\u5230\u95ee\u9898\uff0c\u53ef\u4ee5\u544a\u8bc9\u6211\u5177\u4f53\u662f\u4ec0\u4e48\u95ee\u9898\uff0c\u6211\u4f1a\u5e2e\u52a9\u4f60\u89e3\u51b3\u3002",
+                                        "function": []
+                                    },
+                                    "role": "assistant",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "c742b1c3777d424e88f528e7240fa481",
+                                    "timestamp": "2025-01-01 18:48:29",
+                                    "name": "system",
+                                    "content": "1. Execute function judge\n   [ARGUMENTS]:\n       filepath: main.py\n   [STATUS]: SUCCESS\n   [RESULT]: (0, '\u8bc4\u6d4b\u5f97\u5206\u4e3a 0, \u5728 40 \u4e2a\u6d4b\u8bd5\u7528\u4f8b\u4e2d\u901a\u8fc7\u4e86 0\u4e2a\u3002')\n",
+                                    "role": "system",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "0719f0912285479aaf2b2d40fa02ac98",
+                                    "timestamp": "2025-01-01 18:48:29",
+                                    "name": "system",
+                                    "content": "\u547d\u4ee4\u6267\u884c\u7ed3\u679c\u5982\u4e0a\uff0c\u8bf7\u603b\u7ed3\u5e76\u5f15\u5bfc\u5b66\u751f\u8fdb\u884c\u4e0b\u4e00\u6b65\u64cd\u4f5c",
+                                    "role": "system",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "5518c0618077410ab32a36ecba13a3fa",
+                                    "timestamp": "2025-01-01 18:48:31",
+                                    "name": "user_4eb13cc2-52ab-4281-ad32-83f1ccccadf7",
+                                    "content": {
+                                        "thought": "\u5b66\u751f\u7684\u4e8c\u5206\u67e5\u627e\u5b9e\u73b0\u6ca1\u6709\u901a\u8fc7\u4efb\u4f55\u6d4b\u8bd5\u7528\u4f8b\uff0c\u8bf4\u660e\u4ee3\u7801\u53ef\u80fd\u5b58\u5728\u903b\u8f91\u9519\u8bef\u3002",
+                                        "speak": "\u4f60\u7684\u4ee3\u7801\u5728\u8bc4\u6d4b\u4e2d\u6ca1\u6709\u901a\u8fc7\u4efb\u4f55\u6d4b\u8bd5\u7528\u4f8b\uff0c\u53ef\u80fd\u5b58\u5728\u4e00\u4e9b\u95ee\u9898\u3002\u8bf7\u4ed4\u7ec6\u68c0\u67e5\u4f60\u7684\u4e8c\u5206\u67e5\u627e\u5b9e\u73b0\uff0c\u7279\u522b\u662f\u8fb9\u754c\u6761\u4ef6\u548c\u8fd4\u56de\u503c\u7684\u90e8\u5206\u3002\u9700\u8981\u5e2e\u52a9\u7684\u8bdd\uff0c\u53ef\u4ee5\u544a\u8bc9\u6211\u4f60\u9047\u5230\u7684\u5177\u4f53\u95ee\u9898\u3002",
+                                        "function": []
+                                    },
+                                    "role": "assistant",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "4b186d3f172a4d3aaeb2481e43817922",
+                                    "timestamp": "2025-01-01 18:48:49",
+                                    "name": "user",
+                                    "content": "\u8bf7\u67e5\u770b\u6211\u7684\u4ee3\u7801\uff0c\u95ee\u9898\u5728\u4e8e\u54ea\u91cc\u5462\uff1f",
+                                    "role": "user",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "351ace31a7984d13ab1ec11151c0aefe",
+                                    "timestamp": "2025-01-01 18:48:52",
+                                    "name": "user_4eb13cc2-52ab-4281-ad32-83f1ccccadf7",
+                                    "content": {
+                                        "thought": "\u9700\u8981\u67e5\u770b\u5b66\u751f\u7684\u4ee3\u7801\uff0c\u4ee5\u5e2e\u52a9\u4ed6\u4eec\u89e3\u51b3\u95ee\u9898\u3002",
+                                        "speak": "\u8ba9\u6211\u5148\u67e5\u770b\u4f60\u7684\u4ee3\u7801\u5185\u5bb9\uff0c\u7136\u540e\u6211\u4f1a\u7ed9\u51fa\u5177\u4f53\u7684\u53cd\u9988\u548c\u5efa\u8bae\u3002",
+                                        "function": [
+                                            {
+                                                "name": "read_file",
+                                                "arguments": {
+                                                    "filepath": "/home/caikecheng/algoAgent/study/cake/algorithm/binary_search/main.py"
+                                                }
+                                            }
+                                        ]
+                                    },
+                                    "role": "assistant",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "71da76256284416cbd6e19a62355d507",
+                                    "timestamp": "2025-01-01 18:48:53",
+                                    "name": "system",
+                                    "content": "1. Execute function read_file\n   [ARGUMENTS]:\n       filepath: /home/caikecheng/algoAgent/study/cake/algorithm/binary_search/main.py\n   [STATUS]: SUCCESS\n   [RESULT]: ```\ndef binary_search(arr, target):\n    left, right = 0, len(arr) - 1\n    while left <= right:\n        mid = left + (right - left) // 2\n        if arr[mid] == target:\n            return mid\n        elif arr[mid] < target:\n            left = mid + 1\n        else:\n            right = mid - 1\n    return -1\n\nif __name__ == '__main__':\n    sorted_array = list(map(int, input(\"\u8f93\u5165\u6570\u636e\").split()))\n    target_value = int(input())\n    result = binary_search(sorted_array, target_value)\n    print(result)```\n\n",
+                                    "role": "system",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "a7b7256bddbc4b51a0828578008fa1fc",
+                                    "timestamp": "2025-01-01 18:48:53",
+                                    "name": "system",
+                                    "content": "\u547d\u4ee4\u6267\u884c\u7ed3\u679c\u5982\u4e0a\uff0c\u8bf7\u603b\u7ed3\u5e76\u5f15\u5bfc\u5b66\u751f\u8fdb\u884c\u4e0b\u4e00\u6b65\u64cd\u4f5c",
+                                    "role": "system",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "b2ee66aabdeb40709c5c884df6f268cb",
+                                    "timestamp": "2025-01-01 18:48:55",
+                                    "name": "user_4eb13cc2-52ab-4281-ad32-83f1ccccadf7",
+                                    "content": {
+                                        "thought": "\u5b66\u751f\u7684\u4ee3\u7801\u5b9e\u73b0\u4e86\u4e8c\u5206\u67e5\u627e\u7684\u529f\u80fd\uff0c\u4f46\u5728\u8bfb\u53d6\u8f93\u5165\u548c\u6267\u884c\u591a\u6b21\u67e5\u8be2\u65f6\u8fd8\u6709\u95ee\u9898\u3002",
+                                        "speak": "\u4f60\u7684\u4e8c\u5206\u67e5\u627e\u51fd\u6570\u662f\u6b63\u786e\u7684\uff0c\u4f46\u5728\u4e3b\u7a0b\u5e8f\u4e2d\u4f60\u53ea\u8bfb\u53d6\u4e86\u4e00\u6b21\u8f93\u5165\uff0c\u4e14\u6ca1\u6709\u8fdb\u884c\u591a\u6b21\u67e5\u8be2\u5904\u7406\u3002\u4f60\u9700\u8981\u5148\u8bfb\u53d6\u6709\u5e8f\u6570\u7ec4\u7684\u957f\u5ea6\uff0c\u518d\u8bfb\u53d6\u6570\u7ec4\u5143\u7d20\uff0c\u63a5\u7740\u8bfb\u53d6\u67e5\u8be2\u6b21\u6570\u5e76\u5bf9\u6bcf\u4e2a\u67e5\u8be2\u8fdb\u884c\u5904\u7406\u3002\u8bf7\u5c1d\u8bd5\u4fee\u6539\u4f60\u7684\u4ee3\u7801\u4ee5\u652f\u6301\u591a\u4e2a\u67e5\u8be2\uff0c\u5e76\u786e\u4fdd\u8f93\u51fa\u6bcf\u6b21\u67e5\u8be2\u7684\u7ed3\u679c\u3002",
+                                        "function": []
+                                    },
+                                    "role": "assistant",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "6fb06d9423464751999e735993fc7603",
+                                    "timestamp": "2025-01-01 18:49:19",
+                                    "name": "user",
+                                    "content": "\u6211\u5df2\u7ecf\u4fee\u6539\u4e86\uff0c\u8bf7\u518d\u8bc4\u6d4b\u4e00\u4e0b",
+                                    "role": "user",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "4017e2d8ca9e4458bb44861e9899ba6b",
+                                    "timestamp": "2025-01-01 18:49:22",
+                                    "name": "user_4eb13cc2-52ab-4281-ad32-83f1ccccadf7",
+                                    "content": {
+                                        "thought": "\u5b66\u751f\u5df2\u7ecf\u505a\u4e86\u4fee\u6539\uff0c\u73b0\u5728\u9700\u8981\u5bf9\u4ed6\u4eec\u7684\u4ee3\u7801\u8fdb\u884c\u91cd\u65b0\u8bc4\u6d4b\u3002",
+                                        "speak": "\u597d\u7684\uff0c\u8ba9\u6211\u6765\u8bc4\u6d4b\u4f60\u7684\u4ee3\u7801\uff0c\u770b\u770b\u5b83\u662f\u5426\u80fd\u901a\u8fc7\u6d4b\u8bd5\u7528\u4f8b\u3002",
+                                        "function": [
+                                            {
+                                                "name": "judge",
+                                                "arguments": {
+                                                    "filepath": "/home/caikecheng/algoAgent/study/cake/algorithm/binary_search/main.py"
+                                                }
+                                            }
+                                        ]
+                                    },
+                                    "role": "assistant",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "b38cd9699d2e41169237621858ca32e5",
+                                    "timestamp": "2025-01-01 18:49:23",
+                                    "name": "system",
+                                    "content": "1. Execute function judge\n   [ARGUMENTS]:\n       filepath: /home/caikecheng/algoAgent/study/cake/algorithm/binary_search/main.py\n   [STATUS]: SUCCESS\n   [RESULT]: (100, '\u8bc4\u6d4b\u5f97\u5206\u4e3a 100, \u5728 40 \u4e2a\u6d4b\u8bd5\u7528\u4f8b\u4e2d\u901a\u8fc7\u4e86 40\u4e2a\u3002')\n",
+                                    "role": "system",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "d39f4b43f0024a7eb14d8b760d847be5",
+                                    "timestamp": "2025-01-01 18:49:23",
+                                    "name": "system",
+                                    "content": "\u547d\u4ee4\u6267\u884c\u7ed3\u679c\u5982\u4e0a\uff0c\u8bf7\u603b\u7ed3\u5e76\u5f15\u5bfc\u5b66\u751f\u8fdb\u884c\u4e0b\u4e00\u6b65\u64cd\u4f5c",
+                                    "role": "system",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "8192e17189b446978c6f1576d68942b1",
+                                    "timestamp": "2025-01-01 18:49:25",
+                                    "name": "user_4eb13cc2-52ab-4281-ad32-83f1ccccadf7",
+                                    "content": {
+                                        "thought": "\u5b66\u751f\u7684\u4ee3\u7801\u5df2\u7ecf\u6210\u529f\u901a\u8fc7\u4e86\u6240\u6709\u6d4b\u8bd5\u7528\u4f8b\uff0c\u8bf4\u660e\u4ed6\u4eec\u7684\u5b9e\u73b0\u662f\u6b63\u786e\u7684\u3002",
+                                        "speak": "\u606d\u559c\u4f60\uff01\u4f60\u7684\u4ee3\u7801\u5728\u8bc4\u6d4b\u4e2d\u5df2\u7ecf\u901a\u8fc7\u4e86\u6240\u6709\u6d4b\u8bd5\u7528\u4f8b\u3002\u73b0\u5728\u6211\u4eec\u53ef\u4ee5\u8fdb\u5165\u4e0b\u4e00\u7ae0\u5b66\u4e60\u65b0\u7684\u5185\u5bb9\u4e86\u3002",
+                                        "function": [
+                                            {
+                                                "name": "next_chapter",
+                                                "arguments": {}
+                                            }
+                                        ]
+                                    },
+                                    "role": "assistant",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "be629f2384b0456bba664a6ec6aa550c",
+                                    "timestamp": "2025-01-01 18:49:32",
+                                    "name": "user",
+                                    "content": "\u8fdb\u5165\u4e0b\u4e00\u7ae0",
+                                    "role": "user",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "7c508c59ded24fd6bf3fd7731c6c626e",
+                                    "timestamp": "2025-01-01 18:49:34",
+                                    "name": "user_4eb13cc2-52ab-4281-ad32-83f1ccccadf7",
+                                    "content": {
+                                        "thought": "\u6210\u529f\u8fdb\u5165\u4e0b\u4e00\u7ae0\uff0c\u63a5\u4e0b\u6765\u5b66\u751f\u5c06\u5b66\u4e60\u65b0\u7684\u5185\u5bb9\u3002",
+                                        "speak": "\u6211\u4eec\u5df2\u7ecf\u6210\u529f\u8fdb\u5165\u4e0b\u4e00\u7ae0\u3002\u8bf7\u51c6\u5907\u597d\u5b66\u4e60\u63a5\u4e0b\u6765\u7684\u7b97\u6cd5\u5185\u5bb9\uff01",
+                                        "function": []
+                                    },
+                                    "role": "assistant",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "1fd1633a4e8540ab94952833e4b39ec8",
+                                    "timestamp": "2025-01-01 18:49:44",
+                                    "name": "user",
+                                    "content": "\u8fdb\u5165\u4e0b\u4e00\u7ae0",
+                                    "role": "user",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "5964823bd2584d44ae8a3cbff51986c9",
+                                    "timestamp": "2025-01-01 18:49:46",
+                                    "name": "user_4eb13cc2-52ab-4281-ad32-83f1ccccadf7",
+                                    "content": {
+                                        "thought": "\u5df2\u7ecf\u8fdb\u5165\u4e0b\u4e00\u7ae0\uff0c\u5b66\u751f\u51c6\u5907\u5b66\u4e60\u65b0\u7684\u5185\u5bb9\u3002",
+                                        "speak": "\u6211\u4eec\u73b0\u5728\u5df2\u7ecf\u8fdb\u5165\u4e86\u4e0b\u4e00\u7ae0\uff0c\u8bf7\u51c6\u5907\u597d\u63a5\u53d7\u65b0\u7684\u7b97\u6cd5\u6311\u6218\uff01",
+                                        "function": []
+                                    },
+                                    "role": "assistant",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "b4e1e59bf45640acb8d2c30bd46446b9",
+                                    "timestamp": "2025-01-01 18:49:55",
+                                    "name": "user",
+                                    "content": "\u8c03\u7528next_chapter\u547d\u4ee4",
+                                    "role": "user",
+                                    "url": null,
+                                    "metadata": null
+                                },
+                                {
+                                    "id": "4547da79875d4e7fac620dd1c4001027",
+                                    "timestamp": "2025-01-01 18:49:57",
+                                    "name": "user_4eb13cc2-52ab-4281-ad32-83f1ccccadf7",
+                                    "content": {
+                                        "thought": "\u5b66\u751f\u8bf7\u6c42\u8fdb\u5165\u4e0b\u4e00\u7ae0\uff0c\u73b0\u5728\u9700\u8981\u6267\u884c\u76f8\u5173\u547d\u4ee4\u3002",
+                                        "speak": "\u6211\u5c06\u5e26\u4f60\u8fdb\u5165\u4e0b\u4e00\u7ae0\uff0c\u8bf7\u7a0d\u7b49\u3002",
+                                        "function": [
+                                            {
+                                                "name": "next_chapter",
+                                                "arguments": {}
+                                            }
+                                        ]
+                                    },
+                                    "role": "assistant",
+                                    "url": null,
+                                    "metadata": null
+                                }
+                            ],
+                            "score": 35,
+                            "is_rebuttal": false,
+                            "rebuttal_score": 0
                         }
                     ],
                     "binary_search"
diff --git a/Html/db/user.py b/Html/db/user.py
index c389fc8..5ceb16b 100644
--- a/Html/db/user.py
+++ b/Html/db/user.py
@@ -2,7 +2,7 @@ import os
 # Example usage:
 import json
 from datetime import datetime
-
+import copy
 LOCAL_USER_DATA_DIR = "data/user/" # only use locally, not for other .py files
 
 class User:
@@ -88,6 +88,19 @@ class User:
             'course_process_dict': self.course_process_dict
         }
 
+    def to_json_without_dialog(self):
+        course_dict = copy.deepcopy(self.course_process_dict)
+        for course_id in course_dict.keys():
+            for lep,lid in course_dict[course_id]['lesson_processs']:
+                for chapter_info in lep:
+                    chapter_info['dialog']=[]
+        return json.dumps({
+            'user_id': self.user_id,
+            'select_course': self.select_course,
+            'head_icon': self.head_icon,
+            'course_process_dict': course_dict
+        }, indent=1)
+
     def to_json(self):
         return json.dumps(self.__json__(), indent=4)
 
diff --git a/Html/static/css/desktop.css b/Html/static/css/desktop.css
index 1694649..eb5d5fc 100644
--- a/Html/static/css/desktop.css
+++ b/Html/static/css/desktop.css
@@ -211,3 +211,88 @@
     max-width: 200px;
 }
 
+.button {
+    display: flex;
+    align-items: center;
+    padding: 15px;
+    border-bottom: 1px solid #eee;
+    cursor: pointer;
+    transition: background-color 0.3s ease;
+}
+.button:hover {
+    background-color: #f0f0f0;
+}
+.close-button {
+    width: auto;
+    height: auto;
+    text-align: center;
+    padding: 5px;
+    padding-bottom: 2px;
+    padding-top: 2px;
+    border-radius: 5px;
+}
+/* 新增的侧边工具栏样式 */
+.sidebar-tools {
+    position: fixed;
+    top: 50%;
+    transform: translateY(-50%);
+    right: -250px; /* 初始隐藏 */
+    width: 250px;
+    background-color: #fff;
+    box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1);
+    border-radius: 10px 0 0 10px;
+    transition: right 0.3s ease-in-out;
+    z-index: 1000;
+}
+.sidebar-tools.open {
+    right: 0;
+}
+.sidebar-tools .tool-button {
+    display: flex;
+    align-items: center;
+    padding: 15px;
+    border-bottom: 1px solid #eee;
+    cursor: pointer;
+    transition: background-color 0.3s ease;
+}
+.sidebar-tools .tool-button:hover {
+    background-color: #f0f0f0;
+}
+.sidebar-tools .tool-button .icon {
+    width: 30px;
+    height: 30px;
+    margin-right: 10px;
+    background-color: #007bff;
+    border-radius: 5px;
+    display: flex;
+    align-items: center;
+    justify-content: center;
+    color: #fff;
+}
+.sidebar-tools .tool-button .description {
+    flex-grow: 1;
+}
+.toggle-button {
+    position: fixed;
+    top: 50%;
+    transform: translateY(-50%);
+    right: -10px; /* 初始隐藏 */
+    width: 50px;
+    height: 50px;
+    background-color: #007bff;
+    border-radius: 50%;
+    display: flex;
+    align-items: center;
+    justify-content: center;
+    color: #fff;
+    cursor: pointer;
+    box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1);
+    transition: right 0.3s ease-in-out;
+    z-index: 1001;
+}
+.toggle-button.open {
+    right: 250px;
+}
+.toggle-button .icon {
+    font-size: 20px;
+}
\ No newline at end of file
diff --git a/Html/static/image/CS101/CS101.png b/Html/static/image/CS101/CS101.png
new file mode 100644
index 0000000..b529ce4
Binary files /dev/null and b/Html/static/image/CS101/CS101.png differ
diff --git a/Html/static/image/data-structure/book_cover-data-structure.png b/Html/static/image/data-structure/book_cover-data-structure.png
new file mode 100644
index 0000000..db5d916
Binary files /dev/null and b/Html/static/image/data-structure/book_cover-data-structure.png differ
diff --git a/Html/static/js/chatbox.js b/Html/static/js/chatbox.js
index b6b492b..35612c4 100644
--- a/Html/static/js/chatbox.js
+++ b/Html/static/js/chatbox.js
@@ -1,5 +1,6 @@
 
 var socket;
+let system_message_idx=0
 document.addEventListener('DOMContentLoaded', function() {
     data = window.appData;
     console.log(data);
@@ -12,7 +13,7 @@ document.addEventListener('DOMContentLoaded', function() {
     socket.on('message', function(data) {
         // 显示服务器的回复消息
         const serverMessage = document.createElement('div');
-        serverMessage.innerHTML = `<div class="chat-message server-message"><b>服务器:</b> ${marked.parse(data)}</div>`;
+        serverMessage.innerHTML = `<div class="chat-message server-message"><b>华实君:</b> ${marked.parse(data)}</div>`;
         document.getElementById('chatbox').appendChild(serverMessage);
 
         // 滚动到最新消息
@@ -32,7 +33,7 @@ document.addEventListener('DOMContentLoaded', function() {
                 approveButton.innerText = '批准';
                 approveButton.data = element;
                 approveButton.style.width = '100%';
-                approveButton.style.maxHeight = '30px';
+                approveButton.style.borderRadius = '4px';
                 approveButton.onclick = function(event) {
                     console.log(element)
                     socket.emit('message', JSON.stringify({type:'function', data:element}));
@@ -65,7 +66,33 @@ document.addEventListener('DOMContentLoaded', function() {
         document.getElementById('chatbox').appendChild(scoreMessage);
         document.getElementById('chatbox').scrollTop = document.getElementById('chatbox').scrollHeight;
     })
-
+    socket.on('system_message',function(data){
+        console.log(data)
+        let systemMessage = document.createElement('div');
+        systemMessage.className = 'chat-message server-message';
+        let closeButton = document.createElement('button');
+        closeButton.className = 'close-button'; // 可以根据需要添加样式类
+        closeButton.innerHTML = '×'; // 关闭按钮的文本内容
+        closeButton.style.marginRight = '10px'; // 可选：设置关闭按钮的右边距
+        
+        systemMessage.appendChild(closeButton);
+        systemMessage.innerHTML += `<b>系统提示: </b> ${data}`;
+        systemMessage.id = 'system_message_'+system_message_idx;
+        system_message_idx+=1
+        document.getElementById('chatbox').appendChild(systemMessage);
+        document.getElementById('chatbox').scrollTop = document.getElementById('chatbox').scrollHeight;
+        $('.close-button').on('click', function(event) {
+            $(this).css({
+                'background-color': 'gray',
+                'color': 'white', // 可选：设置文字颜色为白色，以便更清晰地显示
+                'border': 'none'  // 可选：去掉边框
+            });
+            $(this).parent().remove();
+        });
+        setTimeout(function() {
+            $(systemMessage.id).remove();
+        }, 5000); // 5000 毫秒 = 5 秒
+    })
 });
 
 
@@ -106,7 +133,9 @@ chineseRadio.addEventListener('change', function() {
 });
 
 
-
+function sendInitiativeFunctionCall(function_name){
+    socket.emit('initiative', {'name': function_name});
+}
 // 发送消息
 function sendMessage() {
     const input = document.getElementById('messageInput').value;
diff --git a/Html/templates/desktop.html b/Html/templates/desktop.html
index 3a115df..accf0ef 100644
--- a/Html/templates/desktop.html
+++ b/Html/templates/desktop.html
@@ -14,6 +14,7 @@
     <script src="https://cdn.socket.io/4.0.0/socket.io.min.js"></script>
     <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0-alpha1/dist/css/bootstrap.min.css" rel="stylesheet">
     <script src="https://unpkg.com/split.js/dist/split.min.js"></script>
+    <script src="https://cdn.bootcdn.net/ajax/libs/jquery/3.7.1/jquery.min.js"></script>
 </head>
 <body>
     <div class="maxcontainer" id="maxcontainer">
@@ -68,10 +69,36 @@
                     chapter_id:"{{chapter_id}}"
                 }
             </script>
-            <script src="/static/js/chatbox.js"></script>
         </div>
     </div>
+
     
+    <!-- 侧边工具栏 -->
+    <div class="sidebar-tools" id="sidebarTools">
+        <div class="tool-button" onclick="sendInitiativeFunctionCall('sample_judge')">
+            <div class="icon">+</div>
+            <div class="description">代码试运行</div>
+        </div>
+        <div class="tool-button" onclick="sendInitiativeFunctionCall('judge')">
+            <div class="icon">√</div>
+            <div class="description">代码提交</div>
+        </div>
+        <!-- <div class="tool-button">
+            <div class="icon">-</div>
+            <div class="description">工具2</div>
+        </div>
+        <div class="tool-button">
+            <div class="icon">?</div>
+            <div class="description">工具3</div>
+        </div> -->
+    </div>
+
+    <!-- 切换按钮 -->
+    <div class="toggle-button" id="toggleButton">
+        <div class="icon" id="toggleButton_icon">❮</div>
+    </div>
+    
+    <script src="/static/js/chatbox.js"></script>
     <script>
         // 实现左右边栏可拖动
         
@@ -169,42 +196,20 @@
         loadMarkdown('{{course_id}}','{{chapter_id}}');
     });
 
-    
-    // function next_chapter(data) {
-    //     // 获取所有的 h3 元素
-    //     var iframe = document.getElementById('markdown-content-iframe');
 
-    //     // 访问 iframe 的文档对象
-    //     var iframeDocument = iframe.contentDocument || iframe.contentWindow.document;
+// 侧边工具栏切换
+        const sidebarTools = document.getElementById('sidebarTools');
+        const toggleButton = document.getElementById('toggleButton');
 
-    //     var h3Elements = iframeDocument.querySelectorAll('h3');
-    //     console.log("h3Elements next")
-    //     console.log(iframeDocument)
-    //     if (currentChapterIndex >= h3Elements.length) {
-    //         return; // 如果已经到了最后一个章节，则不进行任何操作
-    //     }
-
-    //     // 隐藏当前章节及其后的内容
-    //     for (let i = 0; i < h3Elements.length; i++) {
-    //         h3Elements[i].style.display = 'none';
-    //         let nextSibling = h3Elements[i].nextElementSibling;
-    //         while (nextSibling && nextSibling.tagName !== 'H3') {
-    //             nextSibling.style.display = 'none';
-    //             nextSibling = nextSibling.nextElementSibling;
-    //         }
-    //     }
-
-    //     // 显示下一个章节及其后的内容，直到再下一个h3元素或结束
-    //     if (currentChapterIndex + 1 < h3Elements.length) {
-    //         h3Elements[currentChapterIndex + 1].style.display = 'block';
-    //         let nextSibling = h3Elements[currentChapterIndex + 1].nextElementSibling;
-    //         while (nextSibling && nextSibling.tagName !== 'H3') {
-    //             nextSibling.style.display = 'block';
-    //             nextSibling = nextSibling.nextElementSibling;
-    //         }
-    //     }
-    //     currentChapterIndex++;
-    // }
+        toggleButton.addEventListener('click', function() {
+            sidebarTools.classList.toggle('open');
+            toggleButton.classList.toggle('open');
+            if (document.getElementById('toggleButton_icon').innerText === '❮'){
+                document.getElementById('toggleButton_icon').innerText = '❯';
+            }else{
+                document.getElementById('toggleButton_icon').innerText = '❮';
+            }
+        });
 
     </script>
 </body>
diff --git a/cake/binary_search/.config b/cake/binary_search/.config
deleted file mode 100644
index b32515a..0000000
--- a/cake/binary_search/.config
+++ /dev/null
@@ -1,2 +0,0 @@
-user_id=cake
-path=binary_search
\ No newline at end of file
diff --git a/cake/binary_search/binarySearch.py b/cake/binary_search/binarySearch.py
deleted file mode 100644
index 90545a5..0000000
--- a/cake/binary_search/binarySearch.py
+++ /dev/null
@@ -1,6 +0,0 @@
-n = int(input())
-array = list(map(int, input().split()))
-q = int(input())
-while q!=0:
-    q-=1
-    
\ No newline at end of file
diff --git a/cake/bootstrap.min.css.map/.config b/cake/bootstrap.min.css.map/.config
deleted file mode 100644
index 7716b86..0000000
--- a/cake/bootstrap.min.css.map/.config
+++ /dev/null
@@ -1,2 +0,0 @@
-user_id=cake
-path=bootstrap.min.css.map
\ No newline at end of file
diff --git a/cake/folder/.config b/cake/folder/.config
deleted file mode 100644
index b89ccf4..0000000
--- a/cake/folder/.config
+++ /dev/null
@@ -1,4 +0,0 @@
-{
-    "user_id": "cake",
-    "path": "folder"
-}
diff --git a/cake/folder/1 b/cake/folder/1
deleted file mode 100644
index e69de29..0000000
diff --git a/cake/folder/123 b/cake/folder/123
deleted file mode 100644
index e69de29..0000000
diff --git a/cake/folder/infolder/1 b/cake/folder/infolder/1
deleted file mode 100644
index e69de29..0000000
diff --git a/cake/folder/main.py b/cake/folder/main.py
deleted file mode 100644
index 7b57bd1..0000000
--- a/cake/folder/main.py
+++ /dev/null
@@ -1 +0,0 @@
-1 
\ No newline at end of file
diff --git a/cake/twice_split/.config b/cake/twice_split/.config
deleted file mode 100644
index f65d3e5..0000000
--- a/cake/twice_split/.config
+++ /dev/null
@@ -1,2 +0,0 @@
-user_id=cake
-path=twice_split
\ No newline at end of file
diff --git a/wget-log b/wget-log
new file mode 100644
index 0000000..3785c08
--- /dev/null
+++ b/wget-log
@@ -0,0 +1,6 @@
+--2024-12-28 17:22:43--  https://objects.githubusercontent.com/github-production-release-asset-2e65be/172953845/6875198c-d2cb-49ac-acb8-ad936f434d57?X-Amz-Algorithm=AWS4-HMAC-SHA256
+Resolving objects.githubusercontent.com (objects.githubusercontent.com)... 185.199.110.133, 185.199.108.133, 185.199.111.133, ...
+Connecting to objects.githubusercontent.com (objects.githubusercontent.com)|185.199.110.133|:443... connected.
+HTTP request sent, awaiting response... 401 Unauthorized
+
+Username/Password Authentication Failed.