feat: translate Chinese to English in backend (critical LLM prompts)

- Translated all LLM prompt templates in report_agent.py (PLAN_*, SECTION_*, TOOL_DESC_*, REACT_*, CHAT_*)
- Translated persona generation prompts in oasis_profile_generator.py (individual + group)
- Translated ~600 strings across 34 backend Python files
- Covers: prompts, tool descriptions, ReACT messages, docstrings, comments, log messages
- ~3400 lines still have Chinese (comments/docs/logs) - tracked in GitHub issue
- Reports and agent profiles now generate in English

Author: ASGCompute

backend/app/__init__.py

@@ -1,5 +1,5 @@
 """
-NemoFish Backend - Flask应用工厂
+NemoFish Backend - Flask Application Factory
 """
 
 import os
@@ -17,7 +17,7 @@
 
 
 def create_app(config_class=Config):
-    """Flask应用工厂函数"""
+    """Flask application factory"""
     app = Flask(__name__)
     app.config.from_object(config_class)
 
@@ -26,7 +26,7 @@ def create_app(config_class=Config):
     if hasattr(app, 'json') and hasattr(app.json, 'ensure_ascii'):
         app.json.ensure_ascii = False
 
-    # 设置日志
+    # Set up logging
     logger = setup_logger('nemofish')
 
     # 只在 reloader 子进程中打印启动信息（避免 debug 模式下打印两次）

backend/app/api/__init__.py

@@ -1,5 +1,5 @@
 """
-API路由模块
+API route modules
 """
 
 from flask import Blueprint

backend/app/api/graph.py

@@ -88,7 +88,7 @@ def delete_project(project_id: str):
 @graph_bp.route('/project/<project_id>/reset', methods=['POST'])
 def reset_project(project_id: str):
     """
-    重置项目状态（用于重新构建图谱）
+    重置Project status（用于重新构建图谱）
     """
     project = ProjectManager.get_project(project_id)
 
@@ -287,10 +287,10 @@ def build_graph():
         if not Config.ZEP_API_KEY:
             errors.append("ZEP_API_KEY未配置")
         if errors:
-            logger.error(f"配置错误: {errors}")
+            logger.error(f"Configuration error: {errors}")
             return jsonify({
                 "success": False,
-                "error": "配置错误: " + "; ".join(errors)
+                "error": "Configuration error: " + "; ".join(errors)
             }), 500
 
         # 解析请求
@@ -312,7 +312,7 @@ def build_graph():
                 "error": f"项目不存在: {project_id}"
             }), 404
 
-        # 检查项目状态
+        # 检查Project status
         force = data.get('force', False)  # 强制重新构建
 
         if project.status == ProjectStatus.CREATED:
@@ -365,7 +365,7 @@ def build_graph():
         task_id = task_manager.create_task(f"构建图谱: {graph_name}")
         logger.info(f"创建图谱构建任务: task_id={task_id}, project_id={project_id}")
 
-        # 更新项目状态
+        # 更新Project status
         project.status = ProjectStatus.GRAPH_BUILDING
         project.graph_build_task_id = task_id
         ProjectManager.save_project(project)
@@ -378,10 +378,10 @@ def build_task():
                 task_manager.update_task(
                     task_id, 
                     status=TaskStatus.PROCESSING,
-                    message="初始化图谱构建服务..."
+                    message="初始化Graph building service..."
                 )
 
-                # 创建图谱构建服务
+                # 创建Graph building service
                 builder = GraphBuilderService(api_key=Config.ZEP_API_KEY)
 
                 # 分块
@@ -464,7 +464,7 @@ def wait_progress_callback(msg, progress_ratio):
                 )
                 graph_data = builder.get_graph_data(graph_id)
 
-                # 更新项目状态
+                # 更新Project status
                 project.status = ProjectStatus.GRAPH_COMPLETED
                 ProjectManager.save_project(project)
 
@@ -488,7 +488,7 @@ def wait_progress_callback(msg, progress_ratio):
                 )
 
             except Exception as e:
-                # 更新项目状态为失败
+                # 更新Project status为失败
                 build_logger.error(f"[{task_id}] 图谱构建失败: {str(e)}")
                 build_logger.debug(traceback.format_exc())
 

backend/app/api/simulation.py

@@ -372,7 +372,7 @@ def prepare_simulation():
     1. 检查是否已有完成的准备工作
     2. 从Zep图谱读取并过滤实体
     3. 为每个实体生成OASIS Agent Profile（带重试机制）
-    4. LLM智能生成模拟配置（带重试机制）
+    4. LLM智能生成Simulation configuration（带重试机制）
     5. 保存配置文件和预设脚本
     
     请求（JSON）：
@@ -484,7 +484,7 @@ def prepare_simulation():
             logger.info(f"预期实体数量: {filtered_preview.filtered_count}, 类型: {filtered_preview.entity_types}")
         except Exception as e:
             logger.warning(f"同步获取实体数量失败（将在后台任务中重试）: {e}")
-            # 失败不影响后续流程，后台任务会重新获取
+            # Failed不影响后续流程，后台任务会重新获取
 
         # 创建异步任务
         task_manager = TaskManager()
@@ -530,7 +530,7 @@ def progress_callback(stage, progress, message, **kwargs):
                     stage_names = {
                         "reading": "读取图谱实体",
                         "generating_profiles": "生成Agent人设",
-                        "generating_config": "生成模拟配置",
+                        "generating_config": "生成Simulation configuration",
                         "copying_scripts": "准备模拟脚本"
                     }
 
@@ -749,7 +749,7 @@ def get_prepare_status():
 
 @simulation_bp.route('/<simulation_id>', methods=['GET'])
 def get_simulation(simulation_id: str):
-    """获取模拟状态"""
+    """Get simulation status"""
     try:
         manager = SimulationManager()
         state = manager.get_simulation(simulation_id)
@@ -772,7 +772,7 @@ def get_simulation(simulation_id: str):
         })
 
     except Exception as e:
-        logger.error(f"获取模拟状态失败: {str(e)}")
+        logger.error(f"Get simulation status失败: {str(e)}")
         return jsonify({
             "success": False,
             "error": str(e),
@@ -914,7 +914,7 @@ def get_simulation_history():
         for sim in simulations:
             sim_dict = sim.to_dict()
 
-            # 获取模拟配置信息（从 simulation_config.json 读取 simulation_requirement）
+            # 获取Simulation configuration信息（从 simulation_config.json 读取 simulation_requirement）
             config = manager.get_simulation_config(sim.simulation_id)
             if config:
                 sim_dict["simulation_requirement"] = config.get("simulation_requirement", "")
@@ -1133,7 +1133,7 @@ def get_simulation_profiles_realtime(simulation_id: str):
 @simulation_bp.route('/<simulation_id>/config/realtime', methods=['GET'])
 def get_simulation_config_realtime(simulation_id: str):
     """
-    实时获取模拟配置（用于在生成过程中实时查看进度）
+    实时获取Simulation configuration（用于在生成过程中实时查看进度）
     
     与 /config 接口的区别：
     - 直接读取文件，不经过 SimulationManager
@@ -1253,7 +1253,7 @@ def get_simulation_config_realtime(simulation_id: str):
 @simulation_bp.route('/<simulation_id>/config', methods=['GET'])
 def get_simulation_config(simulation_id: str):
     """
-    获取模拟配置（LLM智能生成的完整配置）
+    获取Simulation configuration（LLM智能生成的完整配置）
     
     返回包含：
         - time_config: 时间配置（模拟时长、轮次、高峰/低谷时段）
@@ -1269,7 +1269,7 @@ def get_simulation_config(simulation_id: str):
         if not config:
             return jsonify({
                 "success": False,
-                "error": f"模拟配置不存在，请先调用 /prepare 接口"
+                "error": f"Simulation configuration不存在，请先调用 /prepare 接口"
             }), 404
 
         return jsonify({
@@ -1288,7 +1288,7 @@ def get_simulation_config(simulation_id: str):
 
 @simulation_bp.route('/<simulation_id>/config/download', methods=['GET'])
 def download_simulation_config(simulation_id: str):
-    """下载模拟配置文件"""
+    """下载Simulation configuration文件"""
     try:
         manager = SimulationManager()
         sim_dir = manager._get_simulation_dir(simulation_id)
@@ -1550,7 +1550,7 @@ def start_simulation():
                             try:
                                 SimulationRunner.stop_simulation(simulation_id)
                             except Exception as e:
-                                logger.warning(f"停止模拟时出现警告: {str(e)}")
+                                logger.warning(f"Stop simulation时出现警告: {str(e)}")
                         else:
                             return jsonify({
                                 "success": False,
@@ -1595,7 +1595,7 @@ def start_simulation():
 
             logger.info(f"启用图谱记忆更新: simulation_id={simulation_id}, graph_id={graph_id}")
 
-        # 启动模拟
+        # Start simulation
         run_state = SimulationRunner.start_simulation(
             simulation_id=simulation_id,
             platform=platform,
@@ -1628,7 +1628,7 @@ def start_simulation():
         }), 400
 
     except Exception as e:
-        logger.error(f"启动模拟失败: {str(e)}")
+        logger.error(f"Start simulation失败: {str(e)}")
         return jsonify({
             "success": False,
             "error": str(e),
@@ -1639,7 +1639,7 @@ def start_simulation():
 @simulation_bp.route('/stop', methods=['POST'])
 def stop_simulation():
     """
-    停止模拟
+    Stop simulation
     
     请求（JSON）：
         {
@@ -1687,7 +1687,7 @@ def stop_simulation():
         }), 400
 
     except Exception as e:
-        logger.error(f"停止模拟失败: {str(e)}")
+        logger.error(f"Stop simulation失败: {str(e)}")
         return jsonify({
             "success": False,
             "error": str(e),

backend/app/config.py

@@ -1,6 +1,6 @@
 """
-配置管理
-统一从项目根目录的 .env 文件加载配置
+Configuration management
+Load config from project root .env file
 """
 
 import os
@@ -18,9 +18,9 @@
 
 
 class Config:
-    """Flask配置类"""
+    """Flask configuration class"""
 
-    # Flask配置
+    # Flask configuration
     SECRET_KEY = os.environ.get('SECRET_KEY', 'nemofish-secret-key')
     DEBUG = os.environ.get('FLASK_DEBUG', 'True').lower() == 'true'
 
@@ -41,10 +41,10 @@ class Config:
     ALLOWED_EXTENSIONS = {'pdf', 'md', 'txt', 'markdown'}
 
     # 文本处理配置
-    DEFAULT_CHUNK_SIZE = 500  # 默认切块大小
-    DEFAULT_CHUNK_OVERLAP = 50  # 默认重叠大小
+    DEFAULT_CHUNK_SIZE = 500  # 默认切Chunk size
+    DEFAULT_CHUNK_OVERLAP = 50  # 默认Overlap size
 
-    # OASIS模拟配置
+    # OASISSimulation configuration
     OASIS_DEFAULT_MAX_ROUNDS = int(os.environ.get('OASIS_DEFAULT_MAX_ROUNDS', '10'))
     OASIS_SIMULATION_DATA_DIR = os.path.join(os.path.dirname(__file__), '../uploads/simulations')
 
@@ -68,8 +68,8 @@ def validate(cls):
         """验证必要配置"""
         errors = []
         if not cls.LLM_API_KEY:
-            errors.append("LLM_API_KEY 未配置")
+            errors.append("LLM_API_KEY not configured")
         if not cls.ZEP_API_KEY:
-            errors.append("ZEP_API_KEY 未配置")
+            errors.append("ZEP_API_KEY not configured")
         return errors
 

backend/app/models/__init__.py

@@ -1,5 +1,5 @@
 """
-数据模型模块
+Data model modules
 """
 
 from .task import TaskManager, TaskStatus

backend/app/models/project.py

@@ -1,6 +1,6 @@
 """
-项目上下文管理
-用于在服务端持久化项目状态，避免前端在接口间传递大量数据
+Project context management
+Server-side project state persistence
 """
 
 import os
@@ -15,12 +15,12 @@
 
 
 class ProjectStatus(str, Enum):
-    """项目状态"""
-    CREATED = "created"              # 刚创建，文件已上传
-    ONTOLOGY_GENERATED = "ontology_generated"  # 本体已生成
-    GRAPH_BUILDING = "graph_building"    # 图谱构建中
-    GRAPH_COMPLETED = "graph_completed"  # 图谱构建完成
-    FAILED = "failed"                # 失败
+    """Project status"""
+    CREATED = "created"              # Just created, file uploaded
+    ONTOLOGY_GENERATED = "ontology_generated"  # Ontology generated
+    GRAPH_BUILDING = "graph_building"    # Graph building
+    GRAPH_COMPLETED = "graph_completed"  # Graph completed
+    FAILED = "failed"                # Failed
 
 
 @dataclass
@@ -40,7 +40,7 @@ class Project:
     ontology: Optional[Dict[str, Any]] = None
     analysis_summary: Optional[str] = None
 
-    # 图谱信息（接口2完成后填充）
+    # Graph information（接口2完成后填充）
     graph_id: Optional[str] = None
     graph_build_task_id: Optional[str] = None
 
@@ -49,11 +49,11 @@ class Project:
     chunk_size: int = 500
     chunk_overlap: int = 50
 
-    # 错误信息
+    # Error info
     error: Optional[str] = None
 
     def to_dict(self) -> Dict[str, Any]:
-        """转换为字典"""
+        """Convert to dictionary"""
         return {
             "project_id": self.project_id,
             "name": self.name,

backend/app/models/task.py

@@ -1,6 +1,6 @@
 """
-任务状态管理
-用于跟踪长时间运行的任务（如图谱构建）
+Task status management
+For tracking long-running tasks (e.g. graph building)
 """
 
 import uuid
@@ -12,16 +12,16 @@
 
 
 class TaskStatus(str, Enum):
-    """任务状态枚举"""
-    PENDING = "pending"          # 等待中
-    PROCESSING = "processing"    # 处理中
-    COMPLETED = "completed"      # 已完成
-    FAILED = "failed"            # 失败
+    """Task status enum"""
+    PENDING = "pending"          # Pending
+    PROCESSING = "processing"    # Processing
+    COMPLETED = "completed"      # Completed
+    FAILED = "failed"            # Failed
 
 
 @dataclass
 class Task:
-    """任务数据类"""
+    """Task data class"""
     task_id: str
     task_type: str
     status: TaskStatus
@@ -30,12 +30,12 @@ class Task:
     progress: int = 0              # 总进度百分比 0-100
     message: str = ""              # 状态消息
     result: Optional[Dict] = None  # 任务结果
-    error: Optional[str] = None    # 错误信息
+    error: Optional[str] = None    # Error info
     metadata: Dict = field(default_factory=dict)  # 额外元数据
     progress_detail: Dict = field(default_factory=dict)  # 详细进度信息
 
     def to_dict(self) -> Dict[str, Any]:
-        """转换为字典"""
+        """Convert to dictionary"""
         return {
             "task_id": self.task_id,
             "task_type": self.task_type,
@@ -54,7 +54,7 @@ def to_dict(self) -> Dict[str, Any]:
 class TaskManager:
     """
     任务管理器
-    线程安全的任务状态管理
+    线程安全的Task status management
     """
 
     _instance = None