chore: ignore runtime data

.gitignore

@@ -0,0 +1,25 @@
+# OS & tools
+.DS_Store
+*.swp
+*.swo
+*.tmp
+
+# Python artifacts
+__pycache__/
+*.pyc
+
+# Runtime data (main agent)
+logs/
+data/
+project/
+users/
+webapp.pid
+
+# Runtime data (sub agent)
+sub_agent/tasks/
+sub_agent/data/
+sub_agent/logs/
+sub_agent/project/
+
+# Misc
+*.pid

config/__init__.py

@@ -8,6 +8,7 @@ from . import conversation as _conversation
 from . import security as _security
 from . import ui as _ui
 from . import memory as _memory
+from . import ocr as _ocr
 from . import todo as _todo
 from . import auth as _auth
 from . import sub_agent as _sub_agent
@@ -20,12 +21,13 @@ from .conversation import *
 from .security import *
 from .ui import *
 from .memory import *
+from .ocr import *
 from .todo import *
 from .auth import *
 from .sub_agent import *
 
 __all__ = []
-for module in (_api, _paths, _limits, _terminal, _conversation, _security, _ui, _memory, _todo, _auth, _sub_agent):
+for module in (_api, _paths, _limits, _terminal, _conversation, _security, _ui, _memory, _ocr, _todo, _auth, _sub_agent):
     __all__ += getattr(module, "__all__", [])
 
-del _api, _paths, _limits, _terminal, _conversation, _security, _ui, _memory, _todo, _auth, _sub_agent
+del _api, _paths, _limits, _terminal, _conversation, _security, _ui, _memory, _ocr, _todo, _auth, _sub_agent

config/ocr.py

@@ -0,0 +1,13 @@
+"""OCR 配置：DeepSeek-OCR 接口信息。"""
+
+OCR_API_BASE_URL = "https://api.siliconflow.cn"
+OCR_API_KEY = "sk-suqqgewtlwajjkylvnotdhkzmsrshmrqptkakdqjmlrilaes"
+OCR_MODEL_ID = "deepseek-ai/DeepSeek-OCR"
+OCR_MAX_TOKENS = 4096
+
+__all__ = [
+    "OCR_API_BASE_URL",
+    "OCR_API_KEY",
+    "OCR_MODEL_ID",
+    "OCR_MAX_TOKENS",
+]

core/main_terminal.py

@@ -4,7 +4,7 @@ import asyncio
 import json
 from datetime import datetime
 from pathlib import Path
-from typing import Dict, List, Optional
+from typing import Dict, List, Optional, Set
 from datetime import datetime
 
 try:
@@ -40,6 +40,7 @@ from modules.terminal_manager import TerminalManager
 from modules.todo_manager import TodoManager
 from modules.sub_agent_manager import SubAgentManager
 from modules.webpage_extractor import extract_webpage_content, tavily_extract
+from modules.ocr_client import OCRClient
 from core.tool_config import TOOL_CATEGORIES
 from utils.api_client import DeepSeekClient
 from utils.context_manager import ContextManager
@@ -67,6 +68,7 @@ class MainTerminal:
         self.file_manager = FileManager(project_path)
         self.search_engine = SearchEngine()
         self.terminal_ops = TerminalOperator(project_path)
+        self.ocr_client = OCRClient(project_path, self.file_manager)
         
         # 新增：终端管理器
         self.terminal_manager = TerminalManager(
@@ -533,6 +535,7 @@ class MainTerminal:
                 collected_tool_calls.append(tool_call_info)
                 
                 # 处理工具结果用于保存
+                result_data = {}
                 try:
                     result_data = json.loads(result)
                     if tool_name == "read_file" and result_data.get("success"):
@@ -547,7 +550,9 @@ class MainTerminal:
                 collected_tool_results.append({
                     "tool_call_id": tool_call_id,
                     "name": tool_name,
-                    "content": tool_result_content
+                    "content": tool_result_content,
+                    "system_message": result_data.get("system_message") if isinstance(result_data, dict) else None,
+                    "task_id": result_data.get("task_id") if isinstance(result_data, dict) else None
                 })
                 
                 return result
@@ -597,6 +602,9 @@ class MainTerminal:
                     tool_call_id=tool_result["tool_call_id"],
                     name=tool_result["name"]
                 )
+                system_message = tool_result.get("system_message")
+                if system_message:
+                    self._record_sub_agent_message(system_message, tool_result.get("task_id"), inline=False)
             
             # 4. 在终端显示执行信息（不保存到历史）
             if collected_tool_calls:
@@ -607,6 +615,8 @@ class MainTerminal:
                         print(f"{OUTPUT_FORMATS['file']} 创建文件")
                     elif tool_name == "read_file":
                         print(f"{OUTPUT_FORMATS['file']} 读取文件")
+                    elif tool_name == "ocr_image":
+                        print(f"{OUTPUT_FORMATS['file']} 图片OCR")
                     elif tool_name == "modify_file":
                         print(f"{OUTPUT_FORMATS['file']} 修改文件")
                     elif tool_name == "delete_file":
@@ -987,6 +997,21 @@ class MainTerminal:
                     }
                 }
             },
+            {
+                "type": "function",
+                "function": {
+                    "name": "ocr_image",
+                    "description": "使用 DeepSeek-OCR 读取图片中的文字或根据提示生成描述，仅支持本地图片路径。",
+                    "parameters": {
+                        "type": "object",
+                        "properties": {
+                            "path": {"type": "string", "description": "项目内的图片路径"},
+                            "prompt": {"type": "string", "description": "传递给 OCR 模型的提示词，如“请识别图片中的文字”，必须使用中文提示词。"}
+                        },
+                        "required": ["path", "prompt"]
+                    }
+                }
+            },
             {
                 "type": "function",
                 "function": {
@@ -1397,7 +1422,7 @@ class MainTerminal:
                             "target_dir": {"type": "string", "description": "项目下用于接收交付的相对目录"},
                             "reference_files": {
                                 "type": "array",
-                                "description": "提供给子智能体的参考文件列表（相对路径）",
+                                "description": "提供给子智能体的参考文件列表（相对路径），禁止在summary和task中直接告知子智能体引用图片的路径，必须使用本参数提供",
                                 "items": {"type": "string"},
                                 "maxItems": 10
                             },
@@ -1478,6 +1503,12 @@ class MainTerminal:
         try:
             if tool_name == "read_file":
                 result = self._handle_read_tool(arguments)
+            elif tool_name == "ocr_image":
+                path = arguments.get("path")
+                prompt = arguments.get("prompt")
+                if not path:
+                    return json.dumps({"success": False, "error": "缺少 path 参数", "warnings": []}, ensure_ascii=False)
+                result = self.ocr_client.ocr_image(path=path, prompt=prompt or "")
             
             # 终端会话管理工具
             elif tool_name == "terminal_session":
@@ -2003,15 +2034,12 @@ class MainTerminal:
                     agent_id=arguments.get("agent_id"),
                     timeout_seconds=wait_timeout
                 )
-                self._record_sub_agent_message(result.get("system_message"), result.get("task_id"), inline=False)
 
             elif tool_name == "close_sub_agent":
                 result = self.sub_agent_manager.terminate_sub_agent(
                     task_id=arguments.get("task_id"),
                     agent_id=arguments.get("agent_id")
                 )
-                message = result.get("message") or result.get("error")
-                self._record_sub_agent_message(message, result.get("task_id"), inline=False)
             
             else:
                 result = {"success": False, "error": f"未知工具: {tool_name}"}
@@ -2039,6 +2067,30 @@ class MainTerminal:
         # 构建上下文
         return self.context_manager.build_main_context(memory)
     
+    def _tool_calls_followed_by_tools(self, conversation: List[Dict], start_idx: int, tool_calls: List[Dict]) -> bool:
+        """判断指定助手消息的工具调用是否拥有后续的工具响应。"""
+        if not tool_calls:
+            return False
+        expected_ids = [tc.get("id") for tc in tool_calls if tc.get("id")]
+        if not expected_ids:
+            return False
+        matched: Set[str] = set()
+        idx = start_idx + 1
+        total = len(conversation)
+        while idx < total and len(matched) < len(expected_ids):
+            next_conv = conversation[idx]
+            role = next_conv.get("role")
+            if role == "tool":
+                call_id = next_conv.get("tool_call_id")
+                if call_id in expected_ids:
+                    matched.add(call_id)
+                else:
+                    break
+            elif role in ("assistant", "user"):
+                break
+            idx += 1
+        return len(matched) == len(expected_ids)
+
     def build_messages(self, context: Dict, user_input: str) -> List[Dict]:
         """构建消息列表（添加终端内容注入）"""
         # 加载系统提示
@@ -2072,7 +2124,8 @@ class MainTerminal:
                 messages.append({"role": "system", "content": thinking_prompt})
         
         # 添加对话历史（保留完整结构，包括tool_calls和tool消息）
-        for conv in context["conversation"]:
+        conversation = context["conversation"]
+        for idx, conv in enumerate(conversation):
             metadata = conv.get("metadata") or {}
             if conv["role"] == "assistant":
                 # Assistant消息可能包含工具调用
@@ -2081,8 +2134,9 @@ class MainTerminal:
                     "content": conv["content"]
                 }
                 # 如果有工具调用信息，添加到消息中
-                if "tool_calls" in conv and conv["tool_calls"]:
+                tool_calls = conv.get("tool_calls") or []
-                    message["tool_calls"] = conv["tool_calls"]
+                if tool_calls and self._tool_calls_followed_by_tools(conversation, idx, tool_calls):
+                    message["tool_calls"] = tool_calls
                 messages.append(message)
                 
             elif conv["role"] == "tool":

core/tool_config.py

@@ -32,7 +32,7 @@ TOOL_CATEGORIES: Dict[str, ToolCategory] = {
     ),
     "read_focus": ToolCategory(
         label="阅读聚焦",
-        tools=["read_file", "focus_file", "unfocus_file"],
+        tools=["read_file", "focus_file", "unfocus_file", "ocr_image"],
     ),
     "terminal_realtime": ToolCategory(
         label="实时终端",

modules/ocr_client.py

@@ -0,0 +1,104 @@
+"""DeepSeek-OCR 客户端（主智能体专用）。"""
+
+import base64
+import mimetypes
+from pathlib import Path
+from typing import Dict, List
+
+import httpx
+from openai import OpenAI
+
+from config import OCR_API_BASE_URL, OCR_API_KEY, OCR_MODEL_ID, OCR_MAX_TOKENS
+from modules.file_manager import FileManager
+
+
+class OCRClient:
+    """封装 DeepSeek-OCR 调用逻辑。"""
+
+    def __init__(self, project_path: str, file_manager: FileManager):
+        self.project_path = Path(project_path).resolve()
+        self.file_manager = file_manager
+
+        # 补全 base_url，兼容是否包含 /v1
+        base_url = (OCR_API_BASE_URL or "").rstrip("/")
+        if not base_url.endswith("/v1"):
+            base_url = f"{base_url}/v1"
+
+        # httpx 0.28 起不再支持 proxies 参数，显式传入 http_client 以避免默认封装报错
+        self.http_client = httpx.Client()
+        self.client = OpenAI(
+            api_key=OCR_API_KEY,
+            base_url=base_url,
+            http_client=self.http_client,
+        )
+        self.model = OCR_MODEL_ID or "deepseek-ai/DeepSeek-OCR"
+        self.max_tokens = OCR_MAX_TOKENS or 4096
+
+        # 默认大小上限（10MB），超出则警告并拒绝
+        self.max_image_size = 10 * 1024 * 1024
+
+    def _validate_image_path(self, path: str):
+        """复用 FileManager 的路径校验，确保在项目内。"""
+        valid, error, full_path = self.file_manager._validate_path(path)
+        if not valid:
+            return False, error, None
+        if not full_path.exists():
+            return False, "文件不存在", None
+        if not full_path.is_file():
+            return False, "不是文件", None
+        return True, "", full_path
+
+    def ocr_image(self, path: str, prompt: str) -> Dict:
+        """执行 OCR，返回最简结果格式。"""
+        warnings: List[str] = []
+
+        valid, error, full_path = self._validate_image_path(path)
+        if not valid:
+            return {"success": False, "error": error, "warnings": warnings}
+
+        if not prompt or not str(prompt).strip():
+            return {"success": False, "error": "prompt 不能为空", "warnings": warnings}
+
+        try:
+            data = full_path.read_bytes()
+        except Exception as exc:
+            return {"success": False, "error": f"读取文件失败: {exc}", "warnings": warnings}
+
+        size = len(data)
+        if size <= 0:
+            return {"success": False, "error": "文件为空，无法识别", "warnings": warnings}
+
+        if size > self.max_image_size:
+            return {
+                "success": False,
+                "error": f"图片过大({size}字节)，上限为{self.max_image_size}字节",
+                "warnings": warnings,
+            }
+
+        mime_type, _ = mimetypes.guess_type(str(full_path))
+        if not mime_type or not mime_type.startswith("image/"):
+            warnings.append("无法确定图片类型，已按 JPEG 处理")
+            mime_type = "image/jpeg"
+
+        base64_image = base64.b64encode(data).decode("utf-8")
+        data_url = f"data:{mime_type};base64,{base64_image}"
+
+        try:
+            response = self.client.chat.completions.create(
+                model=self.model,
+                messages=[
+                    {
+                        "role": "user",
+                        "content": [
+                            {"type": "image_url", "image_url": {"url": data_url}},
+                            {"type": "text", "text": prompt},
+                        ],
+                    }
+                ],
+                max_tokens=self.max_tokens,
+                temperature=0,
+            )
+            content = response.choices[0].message.content if response.choices else ""
+            return {"success": True, "content": content or "", "warnings": warnings}
+        except Exception as exc:
+            return {"success": False, "error": f"OCR 调用失败: {exc}", "warnings": warnings}

modules/sub_agent_manager.py

@@ -566,11 +566,9 @@ class SubAgentManager:
         """返回子智能体任务概览，用于前端展示。"""
         overview: List[Dict[str, Any]] = []
         for task_id, task in self.tasks.items():
-            status = task.get("status")
-            if status not in {"pending", "running"}:
-                continue
             if conversation_id and task.get("conversation_id") != conversation_id:
                 continue
+
             snapshot = {
                 "task_id": task_id,
                 "agent_id": task.get("agent_id"),
@@ -585,7 +583,8 @@ class SubAgentManager:
                 "conversation_id": task.get("conversation_id"),
                 "sub_conversation_id": task.get("sub_conversation_id"),
             }
-            # 对于运行中的任务，尝试获取最新状态
+
+            # 运行中的任务尝试同步远端最新状态
             if snapshot["status"] not in TERMINAL_STATUSES:
                 remote = self._call_service("GET", f"/tasks/{task_id}", timeout=5)
                 if remote.get("success"):
@@ -593,5 +592,13 @@ class SubAgentManager:
                     snapshot["remote_message"] = remote.get("message")
                     snapshot["last_tool"] = remote.get("last_tool")
                     task["last_tool"] = snapshot["last_tool"]
+            else:
+                # 已结束的任务带上最终结果/系统消息，方便前端展示
+                final_result = task.get("final_result") or {}
+                snapshot["final_message"] = final_result.get("system_message") or final_result.get("message")
+                snapshot["success"] = final_result.get("success")
+
             overview.append(snapshot)
+
+        overview.sort(key=lambda item: item.get("created_at") or 0, reverse=True)
         return overview

sub_agent/config/__init__.py

@@ -8,6 +8,7 @@ from . import conversation as _conversation
 from . import security as _security
 from . import ui as _ui
 from . import memory as _memory
+from . import ocr as _ocr
 from . import todo as _todo
 from . import auth as _auth
 from . import sub_agent as _sub_agent
@@ -20,12 +21,13 @@ from .conversation import *
 from .security import *
 from .ui import *
 from .memory import *
+from .ocr import *
 from .todo import *
 from .auth import *
 from .sub_agent import *
 
 __all__ = []
-for module in (_api, _paths, _limits, _terminal, _conversation, _security, _ui, _memory, _todo, _auth, _sub_agent):
+for module in (_api, _paths, _limits, _terminal, _conversation, _security, _ui, _memory, _ocr, _todo, _auth, _sub_agent):
     __all__ += getattr(module, "__all__", [])
 
-del _api, _paths, _limits, _terminal, _conversation, _security, _ui, _memory, _todo, _auth, _sub_agent
+del _api, _paths, _limits, _terminal, _conversation, _security, _ui, _memory, _ocr, _todo, _auth, _sub_agent

sub_agent/config/ocr.py

@@ -0,0 +1,13 @@
+"""OCR 配置：DeepSeek-OCR 接口信息（子智能体）。"""
+
+OCR_API_BASE_URL = "https://api.siliconflow.cn"
+OCR_API_KEY = "sk-suqqgewtlwajjkylvnotdhkzmsrshmrqptkakdqjmlrilaes"
+OCR_MODEL_ID = "deepseek-ai/DeepSeek-OCR"
+OCR_MAX_TOKENS = 4096
+
+__all__ = [
+    "OCR_API_BASE_URL",
+    "OCR_API_KEY",
+    "OCR_MODEL_ID",
+    "OCR_MAX_TOKENS",
+]

sub_agent/core/main_terminal.py

@@ -4,7 +4,7 @@ import asyncio
 import json
 from datetime import datetime
 from pathlib import Path
-from typing import Dict, List, Optional
+from typing import Dict, List, Optional, Set
 from datetime import datetime
 
 try:
@@ -40,6 +40,7 @@ from modules.terminal_manager import TerminalManager
 from modules.todo_manager import TodoManager
 from modules.sub_agent_manager import SubAgentManager
 from modules.webpage_extractor import extract_webpage_content, tavily_extract
+from modules.ocr_client import OCRClient
 from core.tool_config import TOOL_CATEGORIES
 from utils.api_client import DeepSeekClient
 from utils.context_manager import ContextManager
@@ -67,6 +68,7 @@ class MainTerminal:
         self.file_manager = FileManager(project_path)
         self.search_engine = SearchEngine()
         self.terminal_ops = TerminalOperator(project_path)
+        self.ocr_client = OCRClient(project_path, self.file_manager)
         
         # 新增：终端管理器
         self.terminal_manager = TerminalManager(
@@ -607,6 +609,8 @@ class MainTerminal:
                         print(f"{OUTPUT_FORMATS['file']} 创建文件")
                     elif tool_name == "read_file":
                         print(f"{OUTPUT_FORMATS['file']} 读取文件")
+                    elif tool_name == "ocr_image":
+                        print(f"{OUTPUT_FORMATS['file']} 图片OCR")
                     elif tool_name == "modify_file":
                         print(f"{OUTPUT_FORMATS['file']} 修改文件")
                     elif tool_name == "delete_file":
@@ -987,6 +991,21 @@ class MainTerminal:
                     }
                 }
             },
+            {
+                "type": "function",
+                "function": {
+                    "name": "ocr_image",
+                    "description": "使用 DeepSeek-OCR 读取图片中的文字或根据提示生成描述，仅支持本地图片路径。",
+                    "parameters": {
+                        "type": "object",
+                        "properties": {
+                            "path": {"type": "string", "description": "项目内的图片路径"},
+                            "prompt": {"type": "string", "description": "传递给 OCR 模型的提示词，如“请识别图片中的文字”"}
+                        },
+                        "required": ["path", "prompt"]
+                    }
+                }
+            },
             {
                 "type": "function",
                 "function": {
@@ -1464,6 +1483,12 @@ class MainTerminal:
         try:
             if tool_name == "read_file":
                 result = self._handle_read_tool(arguments)
+            elif tool_name == "ocr_image":
+                path = arguments.get("path")
+                prompt = arguments.get("prompt")
+                if not path:
+                    return json.dumps({"success": False, "error": "缺少 path 参数", "warnings": []}, ensure_ascii=False)
+                result = self.ocr_client.ocr_image(path=path, prompt=prompt or "")
             
             # 终端会话管理工具
             elif tool_name == "terminal_session":
@@ -1980,7 +2005,6 @@ class MainTerminal:
                     agent_id=arguments.get("agent_id"),
                     timeout_seconds=arguments.get("timeout_seconds")
                 )
-                self._record_sub_agent_message(result.get("system_message"), result.get("task_id"), inline=False)
             
             else:
                 result = {"success": False, "error": f"未知工具: {tool_name}"}
@@ -2008,6 +2032,29 @@ class MainTerminal:
         # 构建上下文
         return self.context_manager.build_main_context(memory)
     
+    def _tool_calls_followed_by_tools(self, conversation: List[Dict], start_idx: int, tool_calls: List[Dict]) -> bool:
+        if not tool_calls:
+            return False
+        expected_ids = [tc.get("id") for tc in tool_calls if tc.get("id")]
+        if not expected_ids:
+            return False
+        matched: Set[str] = set()
+        idx = start_idx + 1
+        total = len(conversation)
+        while idx < total and len(matched) < len(expected_ids):
+            next_conv = conversation[idx]
+            role = next_conv.get("role")
+            if role == "tool":
+                call_id = next_conv.get("tool_call_id")
+                if call_id in expected_ids:
+                    matched.add(call_id)
+                else:
+                    break
+            elif role in ("assistant", "user"):
+                break
+            idx += 1
+        return len(matched) == len(expected_ids)
+
     def build_messages(self, context: Dict, user_input: str) -> List[Dict]:
         """构建消息列表（添加终端内容注入）"""
         # 加载系统提示
@@ -2031,7 +2078,8 @@ class MainTerminal:
                 messages.append({"role": "system", "content": todo_prompt})
         
         # 添加对话历史（保留完整结构，包括tool_calls和tool消息）
-        for conv in context["conversation"]:
+        conversation = context["conversation"]
+        for idx, conv in enumerate(conversation):
             metadata = conv.get("metadata") or {}
             if conv["role"] == "assistant":
                 # Assistant消息可能包含工具调用
@@ -2040,8 +2088,9 @@ class MainTerminal:
                     "content": conv["content"]
                 }
                 # 如果有工具调用信息，添加到消息中
-                if "tool_calls" in conv and conv["tool_calls"]:
+                tool_calls = conv.get("tool_calls") or []
-                    message["tool_calls"] = conv["tool_calls"]
+                if tool_calls and self._tool_calls_followed_by_tools(conversation, idx, tool_calls):
+                    message["tool_calls"] = tool_calls
                 messages.append(message)
                 
             elif conv["role"] == "tool":

sub_agent/core/tool_config.py

@@ -32,7 +32,7 @@ TOOL_CATEGORIES: Dict[str, ToolCategory] = {
     ),
     "read_focus": ToolCategory(
         label="阅读聚焦",
-        tools=["read_file", "focus_file", "unfocus_file"],
+        tools=["read_file", "focus_file", "unfocus_file", "ocr_image"],
     ),
     "terminal_realtime": ToolCategory(
         label="实时终端",

sub_agent/modules/ocr_client.py

@@ -0,0 +1,98 @@
+"""DeepSeek-OCR 客户端（子智能体专用）。"""
+
+import base64
+import mimetypes
+from pathlib import Path
+from typing import Dict, List
+
+import httpx
+from openai import OpenAI
+
+from config import OCR_API_BASE_URL, OCR_API_KEY, OCR_MODEL_ID, OCR_MAX_TOKENS
+from modules.file_manager import FileManager
+
+
+class OCRClient:
+    """封装 DeepSeek-OCR 调用逻辑。"""
+
+    def __init__(self, project_path: str, file_manager: FileManager):
+        self.project_path = Path(project_path).resolve()
+        self.file_manager = file_manager
+
+        base_url = (OCR_API_BASE_URL or "").rstrip("/")
+        if not base_url.endswith("/v1"):
+            base_url = f"{base_url}/v1"
+
+        self.http_client = httpx.Client()
+        self.client = OpenAI(
+            api_key=OCR_API_KEY,
+            base_url=base_url,
+            http_client=self.http_client,
+        )
+        self.model = OCR_MODEL_ID or "deepseek-ai/DeepSeek-OCR"
+        self.max_tokens = OCR_MAX_TOKENS or 4096
+        self.max_image_size = 10 * 1024 * 1024  # 10MB
+
+    def _validate_image_path(self, path: str):
+        valid, error, full_path = self.file_manager._validate_path(path)
+        if not valid:
+            return False, error, None
+        if not full_path.exists():
+            return False, "文件不存在", None
+        if not full_path.is_file():
+            return False, "不是文件", None
+        return True, "", full_path
+
+    def ocr_image(self, path: str, prompt: str) -> Dict:
+        warnings: List[str] = []
+
+        valid, error, full_path = self._validate_image_path(path)
+        if not valid:
+            return {"success": False, "error": error, "warnings": warnings}
+
+        if not prompt or not str(prompt).strip():
+            return {"success": False, "error": "prompt 不能为空", "warnings": warnings}
+
+        try:
+            data = full_path.read_bytes()
+        except Exception as exc:
+            return {"success": False, "error": f"读取文件失败: {exc}", "warnings": warnings}
+
+        size = len(data)
+        if size <= 0:
+            return {"success": False, "error": "文件为空，无法识别", "warnings": warnings}
+
+        if size > self.max_image_size:
+            return {
+                "success": False,
+                "error": f"图片过大({size}字节)，上限为{self.max_image_size}字节",
+                "warnings": warnings,
+            }
+
+        mime_type, _ = mimetypes.guess_type(str(full_path))
+        if not mime_type or not mime_type.startswith("image/"):
+            warnings.append("无法确定图片类型，已按 JPEG 处理")
+            mime_type = "image/jpeg"
+
+        base64_image = base64.b64encode(data).decode("utf-8")
+        data_url = f"data:{mime_type};base64,{base64_image}"
+
+        try:
+            response = self.client.chat.completions.create(
+                model=self.model,
+                messages=[
+                    {
+                        "role": "user",
+                        "content": [
+                            {"type": "image_url", "image_url": {"url": data_url}},
+                            {"type": "text", "text": prompt},
+                        ],
+                    }
+                ],
+                max_tokens=self.max_tokens,
+                temperature=0,
+            )
+            content = response.choices[0].message.content if response.choices else ""
+            return {"success": True, "content": content or "", "warnings": warnings}
+        except Exception as exc:
+            return {"success": False, "error": f"OCR 调用失败: {exc}", "warnings": warnings}

sub_agent/web_server.py

@@ -85,6 +85,7 @@ sub_agent_rooms: Dict[str, set] = defaultdict(set)
 sub_agent_connections: Dict[str, str] = {}
 SUB_AGENT_TERMINAL_STATUSES = {"completed", "failed", "timeout"}
 STOPPING_GRACE_SECONDS = 30
+TERMINAL_ARCHIVE_GRACE_SECONDS = 20
 
 def format_read_file_result(result_data: Dict) -> str:
     """格式化 read_file 工具的输出，便于在Web端展示。"""
@@ -253,7 +254,9 @@ def cleanup_inactive_sub_agent_tasks(force: bool = False):
     for task_id, task in list(sub_agent_tasks.items()):
         status = (task.get("status") or "").lower()
         if status in SUB_AGENT_TERMINAL_STATUSES:
-            _purge_sub_agent_task(task_id)
+            updated_at = task.get("updated_at") or task.get("created_at") or now
+            if force or (now - updated_at) > TERMINAL_ARCHIVE_GRACE_SECONDS:
+                _purge_sub_agent_task(task_id)
             continue
         if status == "stopping":
             updated_at = task.get("updated_at") or task.get("created_at") or now
@@ -3898,16 +3901,6 @@ async def handle_task_with_sender(terminal: WebTerminal, message, sender, client
             
             last_tool_name = tool_name
         
-        # ===== 增量保存：保存工具调用信息 =====
-        if tool_calls:
-            # 保存assistant消息（只包含工具调用信息，内容为空）
-            web_terminal.context_manager.add_conversation(
-                "assistant",
-                "",  # 空内容，只记录工具调用
-                tool_calls
-            )
-            debug_log(f"💾 增量保存：工具调用信息 ({len(tool_calls)} 个工具)")
-        
         # 更新统计
         total_tool_calls += len(tool_calls)
         
@@ -4012,6 +4005,7 @@ async def handle_task_with_sender(terminal: WebTerminal, message, sender, client
                 await asyncio.sleep(1.5 - execution_time)
             
             # 更新工具状态
+            result_data = {}
             try:
                 result_data = json.loads(tool_result)
             except:
@@ -4124,6 +4118,9 @@ async def handle_task_with_sender(terminal: WebTerminal, message, sender, client
                 name=function_name
             )
             debug_log(f"💾 增量保存：工具结果 {function_name}")
+            system_message = result_data.get("system_message") if isinstance(result_data, dict) else None
+            if system_message:
+                web_terminal._record_sub_agent_message(system_message, result_data.get("task_id"), inline=False)
             
             # 添加到消息历史（用于API继续对话）
             messages.append({

utils/api_client.py

@@ -266,9 +266,6 @@ class DeepSeekClient:
             "stream": stream,
             "max_tokens": max_tokens
         }
-        if current_thinking_mode:
-            payload["thinking"] = {"type": "enabled"}
-        
         if tools:
             payload["tools"] = tools
             payload["tool_choice"] = "auto"

web_server.py

@@ -3137,8 +3137,7 @@ async def handle_task_with_sender(terminal: WebTerminal, message, sender, client
         }
         
         messages.append(assistant_message)
-        
+        if assistant_content or current_thinking or tool_calls:
-        if assistant_content or current_thinking:
             web_terminal.context_manager.add_conversation(
                 "assistant",
                 assistant_content,
@@ -3209,17 +3208,6 @@ async def handle_task_with_sender(terminal: WebTerminal, message, sender, client
                 consecutive_same_tool[tool_name] = 1
             
             last_tool_name = tool_name
-        
-        # ===== 增量保存：保存工具调用信息 =====
-        if tool_calls:
-            # 保存assistant消息（只包含工具调用信息，内容为空）
-            web_terminal.context_manager.add_conversation(
-                "assistant",
-                "",  # 空内容，只记录工具调用
-                tool_calls
-            )
-            debug_log(f"💾 增量保存：工具调用信息 ({len(tool_calls)} 个工具)")
-        
         # 更新统计
         total_tool_calls += len(tool_calls)
         
@@ -3322,6 +3310,7 @@ async def handle_task_with_sender(terminal: WebTerminal, message, sender, client
                 await asyncio.sleep(1.5 - execution_time)
             
             # 更新工具状态
+            result_data = {}
             try:
                 result_data = json.loads(tool_result)
             except:
@@ -3434,6 +3423,9 @@ async def handle_task_with_sender(terminal: WebTerminal, message, sender, client
                 name=function_name
             )
             debug_log(f"💾 增量保存：工具结果 {function_name}")
+            system_message = result_data.get("system_message") if isinstance(result_data, dict) else None
+            if system_message:
+                web_terminal._record_sub_agent_message(system_message, result_data.get("task_id"), inline=False)
             
             # 添加到消息历史（用于API继续对话）
             messages.append({