jarvis-ai-assistant 0.1.207__tar.gz → 0.1.209__tar.gz

{jarvis_ai_assistant-0.1.207/src/jarvis_ai_assistant.egg-info → jarvis_ai_assistant-0.1.209}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: jarvis-ai-assistant
-Version: 0.1.207
+Version: 0.1.209
 Summary: Jarvis: An AI assistant that uses tools to interact with the system
 Home-page: https://github.com/skyfireitdiy/Jarvis
 Author: skyfire
@@ -43,19 +43,18 @@ License-File: LICENSE
 Requires-Dist: requests==2.32.3
 Requires-Dist: colorama==0.4.6
 Requires-Dist: prompt_toolkit==3.0.50
-Requires-Dist: yaspin==2.4.0
 Requires-Dist: pygments==2.19.1
 Requires-Dist: fuzzywuzzy==0.18.0
 Requires-Dist: fastapi==0.115.12
 Requires-Dist: uvicorn==0.33.0
 Requires-Dist: rich==14.0.0
-Requires-Dist: transformers==4.46.3
-Requires-Dist: torch==2.4.1
 Requires-Dist: python-Levenshtein==0.25.1
+Requires-Dist: tiktoken==0.7.0
 Requires-Dist: pillow==10.2.0
 Requires-Dist: openai==1.78.1
 Requires-Dist: tabulate==0.9.0
 Requires-Dist: pyte==0.8.2
+Requires-Dist: pyyaml>=6.0.2
 Provides-Extra: dev
 Requires-Dist: pytest; extra == "dev"
 Requires-Dist: black; extra == "dev"
@@ -106,10 +105,9 @@ pip3 install jarvis-ai-assistant
 | `jarvis` | - | 使用通用代理 |
 | `jarvis-code-agent` | `jca` | 使用代码代理 |
 | `jarvis-smart-shell` | `jss` | 使用智能shell功能 |
-| `jarvis-platform-manager` | - | 使用平台管理功能 |
+| `jarvis-platform-manager` | `jpm` | 使用平台管理功能 |
 | `jarvis-code-review` | - | 使用代码审查功能 |
 | `jarvis-git-commit` | `jgc` | 使用自动化git commit功能 |
-| `jarvis-dev` | - | 使用dev功能（开发中） |
 | `jarvis-git-squash` | - | 使用git squash功能 |
 | `jarvis-multi-agent` | - | 使用多代理功能 |
 | `jarvis-agent` | - | 使用agent功能 |
@@ -483,7 +481,6 @@ JARVIS_THINKING_PLATFORM: yuanbao
 JARVIS_THINKING_MODEL: deep_seek
 ENV:
   YUANBAO_COOKIES: <元宝cookies>
-  YUANBAO_AGENT_ID: <元宝AgentID>
 ```
 
 #### Kimi
@@ -522,23 +519,18 @@ OPENAI_API_BASE: https://api.openai.com/v1
 | `ENV` | {} | 环境变量配置 |
 | `JARVIS_MAX_TOKEN_COUNT` | 960000 | 上下文窗口的最大token数量 |
 | `JARVIS_MAX_INPUT_TOKEN_COUNT` | 32000 | 输入的最大token数量 |
-| `JARVIS_AUTO_COMPLETE` | false | 是否启用自动完成功能 |
-| `JARVIS_SHELL_NAME` | bash | 系统shell名称 |
 | `JARVIS_PLATFORM` | yuanbao | 默认AI平台 |
 | `JARVIS_MODEL` | deep_seek_v3 | 默认模型 |
 | `JARVIS_THINKING_PLATFORM` | JARVIS_PLATFORM | 推理任务使用的平台 |
 | `JARVIS_THINKING_MODEL` | JARVIS_MODEL | 推理任务使用的模型 |
 | `JARVIS_EXECUTE_TOOL_CONFIRM` | false | 执行工具前是否需要确认 |
 | `JARVIS_CONFIRM_BEFORE_APPLY_PATCH` | false | 应用补丁前是否需要确认 |
-| `JARVIS_MAX_TOOL_CALL_COUNT` | 20 | 最大连续工具调用次数 |
-| `JARVIS_AUTO_UPDATE` | true | 是否自动更新Jarvis |
 | `JARVIS_MAX_BIG_CONTENT_SIZE` | 160000 | 最大大内容大小 |
 | `JARVIS_PRETTY_OUTPUT` | false | 是否启用PrettyOutput |
 | `JARVIS_GIT_COMMIT_PROMPT` | "" | 自定义git提交信息生成提示模板 |
 | `JARVIS_PRINT_PROMPT` | false | 是否打印提示 |
 | `JARVIS_USE_METHODOLOGY` | true | 是否启用方法论功能 |
 | `JARVIS_USE_ANALYSIS` | true | 是否启用任务分析功能 |
-| `JARVIS_USE_HISTORY_COUNT` | 0 | 使用N次历史记录作为上下文记忆 |
 | `JARVIS_DATA_PATH` | ~/.jarvis | Jarvis数据存储目录路径 |
 
 ## 🛠️ 工具说明 <a id="tools"></a>
@@ -547,15 +539,10 @@ OPENAI_API_BASE: https://api.openai.com/v1
 | 工具名称 | 描述 |
 |----------|------|
 | ask_user | 交互式用户输入收集 |
-| chdir | 更改当前工作目录 |
 | rewrite_file | 文件重写工具 |
 | edit_file | 代码编辑工具 |
-| code_plan | 理解需求并制定代码修改计划 |
-| create_code_agent | 代码开发工具 |
-| create_sub_agent | 创建子代理处理特定任务 |
 | execute_script | 执行脚本并返回结果 |
 | file_analyzer | 分析文件内容并提取关键信息 |
-| file_operation | 文件批量操作工具 |
 | methodology | 方法论管理工具 |
 | read_code | 代码阅读与分析工具 |
 | read_webpage | 读取网页内容并分析 |
@@ -572,9 +559,7 @@ OPENAI_API_BASE: https://api.openai.com/v1
 | `'ToolUsage'` | 显示可用工具列表及使用方法 |
 | `'ReloadConfig'` | 重新加载配置文件 |
 | `'Web'` | 网页搜索，支持多种提问格式 |
-| `'Plan'` | 生成代码修改计划，提供详细需求描述 |
 | `'FindRelatedFiles'` | 查找与功能相关的文件 |
-| `'Dev'` | 调用create_code_agent开发需求 |
 | `'Fix'` | 修复问题 |
 | `'Check'` | 执行静态代码检查，包括错误和风格问题 |
 

{jarvis_ai_assistant-0.1.207 → jarvis_ai_assistant-0.1.209}/README.md

@@ -36,10 +36,9 @@ pip3 install jarvis-ai-assistant
 | `jarvis` | - | 使用通用代理 |
 | `jarvis-code-agent` | `jca` | 使用代码代理 |
 | `jarvis-smart-shell` | `jss` | 使用智能shell功能 |
-| `jarvis-platform-manager` | - | 使用平台管理功能 |
+| `jarvis-platform-manager` | `jpm` | 使用平台管理功能 |
 | `jarvis-code-review` | - | 使用代码审查功能 |
 | `jarvis-git-commit` | `jgc` | 使用自动化git commit功能 |
-| `jarvis-dev` | - | 使用dev功能（开发中） |
 | `jarvis-git-squash` | - | 使用git squash功能 |
 | `jarvis-multi-agent` | - | 使用多代理功能 |
 | `jarvis-agent` | - | 使用agent功能 |
@@ -413,7 +412,6 @@ JARVIS_THINKING_PLATFORM: yuanbao
 JARVIS_THINKING_MODEL: deep_seek
 ENV:
   YUANBAO_COOKIES: <元宝cookies>
-  YUANBAO_AGENT_ID: <元宝AgentID>
 ```
 
 #### Kimi
@@ -452,23 +450,18 @@ OPENAI_API_BASE: https://api.openai.com/v1
 | `ENV` | {} | 环境变量配置 |
 | `JARVIS_MAX_TOKEN_COUNT` | 960000 | 上下文窗口的最大token数量 |
 | `JARVIS_MAX_INPUT_TOKEN_COUNT` | 32000 | 输入的最大token数量 |
-| `JARVIS_AUTO_COMPLETE` | false | 是否启用自动完成功能 |
-| `JARVIS_SHELL_NAME` | bash | 系统shell名称 |
 | `JARVIS_PLATFORM` | yuanbao | 默认AI平台 |
 | `JARVIS_MODEL` | deep_seek_v3 | 默认模型 |
 | `JARVIS_THINKING_PLATFORM` | JARVIS_PLATFORM | 推理任务使用的平台 |
 | `JARVIS_THINKING_MODEL` | JARVIS_MODEL | 推理任务使用的模型 |
 | `JARVIS_EXECUTE_TOOL_CONFIRM` | false | 执行工具前是否需要确认 |
 | `JARVIS_CONFIRM_BEFORE_APPLY_PATCH` | false | 应用补丁前是否需要确认 |
-| `JARVIS_MAX_TOOL_CALL_COUNT` | 20 | 最大连续工具调用次数 |
-| `JARVIS_AUTO_UPDATE` | true | 是否自动更新Jarvis |
 | `JARVIS_MAX_BIG_CONTENT_SIZE` | 160000 | 最大大内容大小 |
 | `JARVIS_PRETTY_OUTPUT` | false | 是否启用PrettyOutput |
 | `JARVIS_GIT_COMMIT_PROMPT` | "" | 自定义git提交信息生成提示模板 |
 | `JARVIS_PRINT_PROMPT` | false | 是否打印提示 |
 | `JARVIS_USE_METHODOLOGY` | true | 是否启用方法论功能 |
 | `JARVIS_USE_ANALYSIS` | true | 是否启用任务分析功能 |
-| `JARVIS_USE_HISTORY_COUNT` | 0 | 使用N次历史记录作为上下文记忆 |
 | `JARVIS_DATA_PATH` | ~/.jarvis | Jarvis数据存储目录路径 |
 
 ## 🛠️ 工具说明 <a id="tools"></a>
@@ -477,15 +470,10 @@ OPENAI_API_BASE: https://api.openai.com/v1
 | 工具名称 | 描述 |
 |----------|------|
 | ask_user | 交互式用户输入收集 |
-| chdir | 更改当前工作目录 |
 | rewrite_file | 文件重写工具 |
 | edit_file | 代码编辑工具 |
-| code_plan | 理解需求并制定代码修改计划 |
-| create_code_agent | 代码开发工具 |
-| create_sub_agent | 创建子代理处理特定任务 |
 | execute_script | 执行脚本并返回结果 |
 | file_analyzer | 分析文件内容并提取关键信息 |
-| file_operation | 文件批量操作工具 |
 | methodology | 方法论管理工具 |
 | read_code | 代码阅读与分析工具 |
 | read_webpage | 读取网页内容并分析 |
@@ -502,9 +490,7 @@ OPENAI_API_BASE: https://api.openai.com/v1
 | `'ToolUsage'` | 显示可用工具列表及使用方法 |
 | `'ReloadConfig'` | 重新加载配置文件 |
 | `'Web'` | 网页搜索，支持多种提问格式 |
-| `'Plan'` | 生成代码修改计划，提供详细需求描述 |
 | `'FindRelatedFiles'` | 查找与功能相关的文件 |
-| `'Dev'` | 调用create_code_agent开发需求 |
 | `'Fix'` | 修复问题 |
 | `'Check'` | 执行静态代码检查，包括错误和风格问题 |
 

{jarvis_ai_assistant-0.1.207 → jarvis_ai_assistant-0.1.209}/pyproject.toml

@@ -2,13 +2,9 @@
 requires = ["setuptools>=45", "wheel"]
 build-backend = "setuptools.build_meta"
 
-[[tool.uv.index]]
-url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple"
-default = true
-
 [project]
 name = "jarvis-ai-assistant"
-version = "0.1.207"
+version = "0.1.209"
 description = "Jarvis: An AI assistant that uses tools to interact with the system"
 readme = "README.md"
 authors = [{ name = "skyfire", email = "skyfireitdiy@hotmail.com" }]
@@ -29,19 +25,18 @@ dependencies = [
     "requests==2.32.3",
     "colorama==0.4.6",
     "prompt_toolkit==3.0.50",
-    "yaspin==2.4.0",
     "pygments==2.19.1",
     "fuzzywuzzy==0.18.0",
     "fastapi==0.115.12",
     "uvicorn==0.33.0",
     "rich==14.0.0",
-    "transformers==4.46.3",
-    "torch==2.4.1",
     "python-Levenshtein==0.25.1",
+    "tiktoken==0.7.0",
     "pillow==10.2.0",
     "openai==1.78.1",
     "tabulate==0.9.0",
     "pyte==0.8.2",
+    "pyyaml>=6.0.2",
 ]
 requires-python = ">=3.8"
 
@@ -58,10 +53,10 @@ jca = "jarvis.jarvis_code_agent.code_agent:main"
 jarvis-smart-shell = "jarvis.jarvis_smart_shell.main:main"
 jss = "jarvis.jarvis_smart_shell.main:main"
 jarvis-platform-manager = "jarvis.jarvis_platform_manager.main:main"
+jpm = "jarvis.jarvis_platform_manager.main:main"
 jarvis-git-commit = "jarvis.jarvis_git_utils.git_commiter:main"
 jarvis-code-review = "jarvis.jarvis_code_analysis.code_review:main"
 jgc = "jarvis.jarvis_git_utils.git_commiter:main"
-jarvis-dev = "jarvis.jarvis_dev.main:main"
 jarvis-git-squash = "jarvis.jarvis_git_squash.main:main"
 jarvis-multi-agent = "jarvis.jarvis_multi_agent.main:main"
 jarvis-agent = "jarvis.jarvis_agent.main:main"

{jarvis_ai_assistant-0.1.207 → jarvis_ai_assistant-0.1.209}/setup.py

@@ -1,9 +1,9 @@
 # -*- coding: utf-8 -*-
-from setuptools import setup, find_packages
+from setuptools import setup, find_packages  # type: ignore
 
 setup(
     name="jarvis-ai-assistant",
-    version="0.1.207",
+    version="0.1.209",
     author="skyfire",
     author_email="skyfireitdiy@hotmail.com",
     description="An AI assistant that uses various tools to interact with the system",
@@ -14,32 +14,26 @@ setup(
     package_dir={"": "src"},
     include_package_data=True,
     package_data={
-        "jarvis": [
-            "jarvis_data/huggingface.tar.gz",
-            "jarvis_data/config_schema.json"
-        ],
+        "jarvis": ["jarvis_data/config_schema.json"],
     },
     install_requires=[
         "requests==2.32.3",
         "colorama==0.4.6",
         "prompt_toolkit==3.0.50",
-        "yaspin==2.4.0",
         "pygments==2.19.1",
         "fuzzywuzzy==0.18.0",
         "fastapi==0.115.12",
         "uvicorn==0.33.0",
         "rich==14.0.0",
-        "transformers==4.46.3",
-        "torch==2.4.1",
         "python-Levenshtein==0.25.1",
+        "tiktoken==0.7.0",
         "pillow==10.2.0",
         "openai==1.78.1",
         "tabulate==0.9.0",
         "pyte==0.8.2",
+        "pyyaml>=6.0.2",
     ],
-    extras_require={
-        "dev": ["pytest", "black", "isort", "mypy", "build", "twine"]
-    },
+    extras_require={"dev": ["pytest", "black", "isort", "mypy", "build", "twine"]},
     entry_points={
         "console_scripts": [
             "jarvis=jarvis.jarvis_agent.jarvis:main",
@@ -48,10 +42,10 @@ setup(
             "jarvis-smart-shell=jarvis.jarvis_smart_shell.main:main",
             "jss=jarvis.jarvis_smart_shell.main:main",
             "jarvis-platform-manager=jarvis.jarvis_platform_manager.main:main",
+            "jpm=jarvis.jarvis_platform_manager.main:main",
             "jarvis-code-review=jarvis.jarvis_code_analysis.code_review:main",
-            "jarvis-git-commit=jarvis.jarvis_git_utils.git_commiter:main", 
+            "jarvis-git-commit=jarvis.jarvis_git_utils.git_commiter:main",
             "jgc=jarvis.jarvis_git_utils.git_commiter:main",
-            "jarvis-dev=jarvis.jarvis_dev.main:main",
             "jarvis-git-squash=jarvis.jarvis_git_squash.main:main",
             "jarvis-multi-agent=jarvis.jarvis_multi_agent.main:main",
             "jarvis-agent=jarvis.jarvis_agent.main:main",

{jarvis_ai_assistant-0.1.207 → jarvis_ai_assistant-0.1.209}/src/jarvis/__init__.py

@@ -1,4 +1,4 @@
 # -*- coding: utf-8 -*-
 """Jarvis AI Assistant"""
 
-__version__ = "0.1.207"
+__version__ = "0.1.209"

{jarvis_ai_assistant-0.1.207 → jarvis_ai_assistant-0.1.209}/src/jarvis/jarvis_agent/__init__.py

@@ -7,27 +7,30 @@ import platform
 from typing import Any, Callable, Dict, List, Optional, Protocol, Tuple, Union
 
 # 第三方库导入
-from yaspin import yaspin  # type: ignore
 
 # 本地库导入
 # jarvis_agent 相关
 # jarvis_platform 相关
 from jarvis.jarvis_platform.base import BasePlatform
 from jarvis.jarvis_platform.registry import PlatformRegistry
+
 # jarvis_utils 相关
-from jarvis.jarvis_utils.config import (get_data_dir, get_max_token_count,
-                                        get_max_tool_call_count,
-                                        is_auto_complete,
-                                        is_execute_tool_confirm,
-                                        is_use_analysis, get_history_count, is_use_methodology)
+from jarvis.jarvis_utils.config import (
+    get_max_token_count,
+    is_execute_tool_confirm,
+    is_use_analysis,
+    is_use_methodology,
+)
 from jarvis.jarvis_utils.embedding import get_context_token_count
-from jarvis.jarvis_utils.globals import (delete_agent, get_interrupt,
-                                         make_agent_name, set_agent,
-                                         set_interrupt)
+from jarvis.jarvis_utils.globals import (
+    delete_agent,
+    get_interrupt,
+    make_agent_name,
+    set_agent,
+    set_interrupt,
+)
 from jarvis.jarvis_utils.input import get_multiline_input
-from jarvis.jarvis_utils.jarvis_history import JarvisHistory
-from jarvis.jarvis_utils.methodology import (load_methodology,
-                                             upload_methodology)
+from jarvis.jarvis_utils.methodology import load_methodology, upload_methodology
 from jarvis.jarvis_utils.output import OutputType, PrettyOutput
 from jarvis.jarvis_utils.tag import ct, ot
 from jarvis.jarvis_utils.utils import user_confirm
@@ -139,7 +142,7 @@ class Agent:
         self.prompt = ""
 
     def __del__(self):
-        self.history.stop_record()
+        # 只有在记录启动时才停止记录
         delete_agent(self.name)
 
     def __init__(
@@ -150,7 +153,7 @@ class Agent:
         platform: Union[Optional[BasePlatform], Optional[str]] = None,
         model_name: Optional[str] = None,
         summary_prompt: Optional[str] = None,
-        auto_complete: Optional[bool] = None,
+        auto_complete: bool = False,
         output_handler: List[OutputHandlerProtocol] = [],
         use_tools: List[str] = [],
         input_handler: Optional[List[Callable[[str, Any], Tuple[str, bool]]]] = None,
@@ -160,7 +163,6 @@ class Agent:
         use_methodology: Optional[bool] = None,
         use_analysis: Optional[bool] = None,
         files: List[str] = [],
-        history_count: Optional[int] = None,
     ):
         self.files = files
         """初始化Jarvis Agent实例
@@ -235,17 +237,8 @@ class Agent:
         # Load configuration from environment variables
         self.addon_prompt = ""
 
-        self.tool_call_count = 0
-        self.max_tool_call_count = get_max_tool_call_count()
         self.after_tool_call_cb: Optional[Callable[[Agent], None]] = None
 
-        self.history = JarvisHistory()
-        self.history_dir = str(Path(get_data_dir())/"history")
-        self.history.start_record(self.history_dir)
-
-        self.history_count = history_count if history_count is not None else get_history_count()
-
-
         self.execute_tool_confirm = (
             execute_tool_confirm
             if execute_tool_confirm is not None
@@ -274,9 +267,7 @@ class Agent:
         )
 
         self.max_token_count = get_max_token_count()
-        self.auto_complete = (
-            auto_complete if auto_complete is not None else is_auto_complete()
-        )
+        self.auto_complete = auto_complete
         welcome_message = f"{name} 初始化完成 - 使用 {self.model.name()} 模型"
 
         PrettyOutput.print(welcome_message, OutputType.SYSTEM)
@@ -407,10 +398,10 @@ class Agent:
         if self.conversation_length > self.max_token_count:
             message = self._summarize_and_clear_history() + "\n\n" + message
             self.conversation_length += get_context_token_count(message)
-        self.history.append_msg("user", message)
+
         response = self.model.chat_until_success(message)  # type: ignore
         self.conversation_length += get_context_token_count(response)
-        self.history.append_msg("assistant", response)
+
         return response
 
     def generate_summary(self) -> str:
@@ -422,8 +413,8 @@ class Agent:
         注意:
             仅生成摘要，不修改对话状态
         """
-        with yaspin(text="正在总结对话历史...", color="cyan") as spinner:
-            summary_prompt = """
+        print("📄 正在总结对话历史...")
+        summary_prompt = """
 <summary_request>
 <objective>
 请对当前对话历史进行简明扼要的总结，提取关键信息和重要决策点。这个总结将作为上下文继续任务，因此需要保留对后续对话至关重要的内容。
@@ -447,14 +438,11 @@ class Agent:
 """
 
         try:
-            with spinner.hidden():
-                summary = self.model.chat_until_success(self.prompt + "\n" + summary_prompt)  # type: ignore
-            spinner.text = "总结对话历史完成"
-            spinner.ok("✅")
+            summary = self.model.chat_until_success(self.prompt + "\n" + summary_prompt)  # type: ignore
+            print("✅ 总结对话历史完成")
             return summary
         except Exception as e:
-            spinner.text = "总结对话历史失败"
-            spinner.fail("❌")
+            print("❌ 总结对话历史失败")
             return ""
 
     def _summarize_and_clear_history(self) -> str:
@@ -482,9 +470,9 @@ class Agent:
                 summary = self.generate_summary()
             else:
                 import tempfile
+
                 tmp_file = tempfile.NamedTemporaryFile(delete=False)
                 tmp_file_name = tmp_file.name
-                self.history.save_history(tmp_file_name)
             self.clear_history()  # type: ignore
 
             if need_summary:
@@ -540,32 +528,16 @@ class Agent:
             )
         if len(tool_list) == 0:
             return False, ""
-        if (
-            self.max_tool_call_count > 0
-            and self.tool_call_count >= self.max_tool_call_count
-        ):
-            if user_confirm(f"工具调用次数超过限制，是否继续执行？", True):
-                self.reset_tool_call_count()
-            else:
-                return False, ""
-        if self.execute_tool_confirm:
-            self.reset_tool_call_count()
+
         if not self.execute_tool_confirm or user_confirm(
             f"需要执行{tool_list[0].name()}确认执行？", True
         ):
-            with yaspin(
-                text=f"正在执行{tool_list[0].name()}...", color="cyan"
-            ) as spinner:
-                with spinner.hidden():
-                    result = tool_list[0].handle(response, self)
-                spinner.text = f"{tool_list[0].name()}执行完成"
-                spinner.ok("✅")
-                self.tool_call_count += 1
-                return result
-        return False, ""
+            print(f"🔧 正在执行{tool_list[0].name()}...")
+            result = tool_list[0].handle(response, self)
+            print(f"✅ {tool_list[0].name()}执行完成")
 
-    def reset_tool_call_count(self):
-        self.tool_call_count = 0
+            return result
+        return False, ""
 
     def _complete_task(self) -> str:
         """完成任务并生成总结(如果需要)
@@ -581,21 +553,19 @@ class Agent:
         if self.use_analysis:
             self._analysis_task()
         if self.need_summary:
-            with yaspin(text="正在生成总结...", color="cyan") as spinner:
-                self.prompt = self.summary_prompt
-                with spinner.hidden():
-                    ret = self.model.chat_until_success(self.prompt)  # type: ignore
-                    spinner.text = "总结生成完成"
-                    spinner.ok("✅")
-                    return ret
+            print("📄 正在生成总结...")
+            self.prompt = self.summary_prompt
+            ret = self.model.chat_until_success(self.prompt)  # type: ignore
+            print("✅ 总结生成完成")
+            return ret
 
         return "任务完成"
 
     def _analysis_task(self):
-        with yaspin(text="正在分析任务...", color="cyan") as spinner:
-            try:
-                # 让模型判断是否需要生成方法论
-                analysis_prompt = f"""<task_analysis>
+        print("🔍 正在分析任务...")
+        try:
+            # 让模型判断是否需要生成方法论
+            analysis_prompt = f"""<task_analysis>
 <request>
 当前任务已结束，请分析该任务的解决方案：
 1. 首先检查现有工具或方法论是否已经可以完成该任务，如果可以，直接说明即可，无需生成新内容
@@ -741,17 +711,12 @@ arguments:
 </output_requirements>
 </task_analysis>"""
 
-                self.prompt = analysis_prompt
-                with spinner.hidden():
-                    response = self.model.chat_until_success(self.prompt)  # type: ignore
-
-                with spinner.hidden():
-                    self._call_tools(response)
-                spinner.text = "分析完成"
-                spinner.ok("✅")
-            except Exception as e:
-                spinner.text = "分析失败"
-                spinner.fail("❌")
+            self.prompt = analysis_prompt
+            response = self.model.chat_until_success(self.prompt)  # type: ignore
+            self._call_tools(response)
+            print("✅ 分析完成")
+        except Exception as e:
+            print("❌ 分析失败")
 
     def make_default_addon_prompt(self, need_complete: bool) -> str:
         """生成附加提示。
@@ -808,9 +773,8 @@ arguments:
             set_agent(self.name, self)
 
             while True:
-                history_md = ""
                 if self.first:
-                    history_md = self._first_run()
+                    self._first_run()
                 try:
                     current_response = self._call_model(self.prompt, True)
                     self.prompt = ""
@@ -822,8 +786,13 @@ arguments:
                         )
                         if user_input:
                             # 如果有工具调用且用户确认继续，则将干预信息和工具执行结果拼接为prompt
-                            if any(handler.can_handle(current_response) for handler in self.output_handler):
-                                if user_confirm("检测到有工具调用，是否继续处理工具调用？", True):
+                            if any(
+                                handler.can_handle(current_response)
+                                for handler in self.output_handler
+                            ):
+                                if user_confirm(
+                                    "检测到有工具调用，是否继续处理工具调用？", True
+                                ):
                                     self.prompt = f"{user_input}\n\n{current_response}"
                                     continue
                             self.prompt += f"{user_input}"
@@ -843,8 +812,6 @@ arguments:
                     if self.auto_complete and ot("!!!COMPLETE!!!") in current_response:
                         return self._complete_task()
 
-                    self.reset_tool_call_count()
-
                     # 获取用户输入
                     user_input = self.multiline_inputer(
                         f"{self.name}: 请输入，或输入空行来结束当前任务："
@@ -858,8 +825,6 @@ arguments:
                         return self._complete_task()
 
                 except Exception as e:
-                    if history_md:
-                        os.remove(history_md)
                     PrettyOutput.print(f"任务失败: {str(e)}", OutputType.ERROR)
                     return f"Task failed: {str(e)}"
 
@@ -868,21 +833,15 @@ arguments:
             return f"Task failed: {str(e)}"
 
     def _first_run(self):
-        history_md = ""
-        if self.history_count > 0 and self.model and self.model.support_upload_files():
-            import tempfile
-            timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
-            history_md = str(Path(tempfile.gettempdir())/f"{self.name}_history_{timestamp}.md")
-            self.history.export_history_to_markdown(self.history_dir, history_md, max_files=self.history_count)
-            self.files.append(history_md)
-
-                        # 如果有上传文件，先上传文件
+        # 如果有上传文件，先上传文件
         if self.model and self.model.support_upload_files():
             if self.use_methodology:
                 if not upload_methodology(self.model, other_files=self.files):
                     if self.files:
-                        PrettyOutput.print("文件上传失败，将忽略文件列表", OutputType.WARNING)
-                                    # 上传失败则回退到本地加载
+                        PrettyOutput.print(
+                            "文件上传失败，将忽略文件列表", OutputType.WARNING
+                        )
+                        # 上传失败则回退到本地加载
                     msg = self.prompt
                     for handler in self.input_handler:
                         msg, _ = handler(msg, self)
@@ -894,7 +853,9 @@ arguments:
                         self.prompt = f"{self.prompt}\n\n上传的文件包含历史对话信息，可以从中获取一些经验信息。"
             elif self.files:
                 if not self.model.upload_files(self.files):
-                    PrettyOutput.print("文件上传失败，将忽略文件列表", OutputType.WARNING)
+                    PrettyOutput.print(
+                        "文件上传失败，将忽略文件列表", OutputType.WARNING
+                    )
                 else:
                     self.prompt = f"{self.prompt}\n\n上传的文件包含历史对话信息，可以从中获取一些经验信息。"
         else:
@@ -907,7 +868,6 @@ arguments:
                 self.prompt = f"{self.prompt}\n\n以下是历史类似问题的执行经验，可参考：\n{load_methodology(msg, self.get_tool_registry())}"
 
         self.first = False
-        return history_md
 
     def clear_history(self):
         """清空对话历史但保留系统提示