fix: harden runtime API and simulation handling

backend/app/api/graph.py

@@ -295,7 +295,7 @@ def build_graph():
             }), 500
         
         # 解析请求
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         project_id = data.get('project_id')
         logger.debug(f"请求参数: project_id={project_id}")
         

backend/app/api/report.py

@@ -48,7 +48,7 @@ def generate_report():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         simulation_id = data.get('simulation_id')
         if not simulation_id:
@@ -223,7 +223,7 @@ def get_generate_status():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         task_id = data.get('task_id')
         simulation_id = data.get('simulation_id')
@@ -497,7 +497,7 @@ def chat_with_report_agent():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         simulation_id = data.get('simulation_id')
         message = data.get('message')
@@ -945,7 +945,7 @@ def search_graph_tool():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         graph_id = data.get('graph_id')
         query = data.get('query')
@@ -991,7 +991,7 @@ def get_graph_statistics_tool():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         graph_id = data.get('graph_id')
         

backend/app/api/simulation.py

@@ -192,7 +192,7 @@ def create_simulation():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         project_id = data.get('project_id')
         if not project_id:
@@ -291,7 +291,7 @@ def _check_simulation_prepared(simulation_id: str) -> tuple:
     state_file = os.path.join(simulation_dir, "state.json")
     try:
         import json
-        with open(state_file, 'r', encoding='utf-8') as f:
+        with open(state_file, 'r', encoding='utf-8-sig') as f:
             state_data = json.load(f)
         
         status = state_data.get("status", "")
@@ -403,7 +403,7 @@ def prepare_simulation():
     from ..config import Config
     
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         simulation_id = data.get('simulation_id')
         if not simulation_id:
@@ -670,7 +670,7 @@ def get_prepare_status():
     from ..models.task import TaskManager
     
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         task_id = data.get('task_id')
         simulation_id = data.get('simulation_id')
@@ -1104,7 +1104,7 @@ def get_simulation_profiles_realtime(simulation_id: str):
         state_file = os.path.join(sim_dir, "state.json")
         if os.path.exists(state_file):
             try:
-                with open(state_file, 'r', encoding='utf-8') as f:
+                with open(state_file, 'r', encoding='utf-8-sig') as f:
                     state_data = json.load(f)
                     status = state_data.get("status", "")
                     is_generating = status == "preparing"
@@ -1200,7 +1200,7 @@ def get_simulation_config_realtime(simulation_id: str):
         state_file = os.path.join(sim_dir, "state.json")
         if os.path.exists(state_file):
             try:
-                with open(state_file, 'r', encoding='utf-8') as f:
+                with open(state_file, 'r', encoding='utf-8-sig') as f:
                     state_data = json.load(f)
                     status = state_data.get("status", "")
                     is_generating = status == "preparing"
@@ -1388,7 +1388,7 @@ def generate_profiles():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         graph_id = data.get('graph_id')
         if not graph_id:
@@ -1490,7 +1490,7 @@ def start_simulation():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
 
         simulation_id = data.get('simulation_id')
         if not simulation_id:
@@ -1662,7 +1662,7 @@ def stop_simulation():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         simulation_id = data.get('simulation_id')
         if not simulation_id:
@@ -2191,7 +2191,7 @@ def interview_agent():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         simulation_id = data.get('simulation_id')
         agent_id = data.get('agent_id')
@@ -2313,7 +2313,7 @@ def interview_agents_batch():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
 
         simulation_id = data.get('simulation_id')
         interviews = data.get('interviews')
@@ -2440,7 +2440,7 @@ def interview_all_agents():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
 
         simulation_id = data.get('simulation_id')
         prompt = data.get('prompt')
@@ -2544,7 +2544,7 @@ def get_interview_history():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         simulation_id = data.get('simulation_id')
         platform = data.get('platform')  # 不指定则返回两个平台的历史
@@ -2606,7 +2606,7 @@ def get_env_status():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         simulation_id = data.get('simulation_id')
         
@@ -2673,7 +2673,7 @@ def close_simulation_env():
         }
     """
     try:
-        data = request.get_json() or {}
+        data = request.get_json(silent=True) or {}
         
         simulation_id = data.get('simulation_id')
         timeout = data.get('timeout', 30)

backend/app/services/oasis_profile_generator.py

@@ -195,7 +195,8 @@ class OasisProfileGenerator:
         
         self.client = OpenAI(
             api_key=self.api_key,
-            base_url=self.base_url
+            base_url=self.base_url,
+            default_headers={"User-Agent": "python-requests/2.32.5"}
         )
         
         # Zep客户端用于检索丰富上下文
@@ -315,6 +316,7 @@ class OasisProfileGenerator:
             return results
         
         comprehensive_query = t('progress.zepSearchQuery', name=entity_name)
+        zep_query = comprehensive_query[:400]
         
         def search_edges():
             """搜索边（事实/关系）- 带重试机制"""
@@ -325,7 +327,7 @@ class OasisProfileGenerator:
             for attempt in range(max_retries):
                 try:
                     return self.zep_client.graph.search(
-                        query=comprehensive_query,
+                        query=zep_query,
                         graph_id=self.graph_id,
                         limit=30,
                         scope="edges",
@@ -350,7 +352,7 @@ class OasisProfileGenerator:
             for attempt in range(max_retries):
                 try:
                     return self.zep_client.graph.search(
-                        query=comprehensive_query,
+                        query=zep_query,
                         graph_id=self.graph_id,
                         limit=20,
                         scope="nodes",

backend/app/services/simulation_config_generator.py

@@ -237,7 +237,8 @@ class SimulationConfigGenerator:
         
         self.client = OpenAI(
             api_key=self.api_key,
-            base_url=self.base_url
+            base_url=self.base_url,
+            default_headers={"User-Agent": "python-requests/2.32.5"}
         )
     
     def generate_config(

backend/app/services/simulation_ipc.py

@@ -278,7 +278,7 @@ class SimulationIPCClient:
             return False
         
         try:
-            with open(status_file, 'r', encoding='utf-8') as f:
+            with open(status_file, 'r', encoding='utf-8-sig') as f:
                 status = json.load(f)
             return status.get("status") == "alive"
         except (json.JSONDecodeError, OSError):

backend/app/services/simulation_manager.py

@@ -165,7 +165,7 @@ class SimulationManager:
         if not os.path.exists(state_file):
             return None
         
-        with open(state_file, 'r', encoding='utf-8') as f:
+        with open(state_file, 'r', encoding='utf-8-sig') as f:
             data = json.load(f)
         
         state = SimulationState(

backend/app/services/simulation_runner.py

@@ -12,6 +12,7 @@ import threading
 import subprocess
 import signal
 import atexit
+import psutil
 from typing import Dict, Any, List, Optional, Union
 from dataclasses import dataclass, field
 from datetime import datetime
@@ -247,7 +248,7 @@ class SimulationRunner:
             return None
         
         try:
-            with open(state_file, 'r', encoding='utf-8') as f:
+            with open(state_file, 'r', encoding='utf-8-sig') as f:
                 data = json.load(f)
             
             state = SimulationRunState(
@@ -1232,12 +1233,18 @@ class SimulationRunner:
                     
                     # 更新 run_state.json
                     state = cls.get_run_state(simulation_id)
+                    keep_completed = bool(
+                        state and (
+                            state.runner_status == RunnerStatus.COMPLETED or
+                            (state.progress_percent >= 100 and state.twitter_completed and state.reddit_completed)
+                        )
+                    )
                     if state:
-                        state.runner_status = RunnerStatus.STOPPED
+                        state.runner_status = RunnerStatus.COMPLETED if keep_completed else RunnerStatus.STOPPED
                         state.twitter_running = False
                         state.reddit_running = False
-                        state.completed_at = datetime.now().isoformat()
-                        state.error = "服务器关闭，模拟被终止"
+                        state.completed_at = state.completed_at or datetime.now().isoformat()
+                        state.error = None if keep_completed else "服务器关闭，模拟被终止"
                         cls._save_run_state(state)
                     
                     # 同时更新 state.json，将状态设为 stopped
@@ -1246,13 +1253,13 @@ class SimulationRunner:
                         state_file = os.path.join(sim_dir, "state.json")
                         logger.info(f"尝试更新 state.json: {state_file}")
                         if os.path.exists(state_file):
-                            with open(state_file, 'r', encoding='utf-8') as f:
+                            with open(state_file, 'r', encoding='utf-8-sig') as f:
                                 state_data = json.load(f)
-                            state_data['status'] = 'stopped'
+                            state_data['status'] = 'completed' if keep_completed else 'stopped'
                             state_data['updated_at'] = datetime.now().isoformat()
                             with open(state_file, 'w', encoding='utf-8') as f:
                                 json.dump(state_data, f, indent=2, ensure_ascii=False)
-                            logger.info(f"已更新 state.json 状态为 stopped: {simulation_id}")
+                            logger.info(f"已更新 state.json 状态为 {state_data['status']}: {simulation_id}")
                         else:
                             logger.warning(f"state.json 不存在: {state_file}")
                     except Exception as state_err:
@@ -1386,7 +1393,23 @@ class SimulationRunner:
             return False
 
         ipc_client = SimulationIPCClient(sim_dir)
-        return ipc_client.check_env_alive()
+        if not ipc_client.check_env_alive():
+            return False
+
+        process = cls._processes.get(simulation_id)
+        if process is not None:
+            return process.poll() is None
+
+        state = cls.get_run_state(simulation_id)
+        pid = state.process_pid if state else None
+        if not pid:
+            return False
+
+        try:
+            proc = psutil.Process(pid)
+            return proc.is_running() and proc.status() != psutil.STATUS_ZOMBIE
+        except psutil.Error:
+            return False
 
     @classmethod
     def get_env_status_detail(cls, simulation_id: str) -> Dict[str, Any]:
@@ -1413,12 +1436,13 @@ class SimulationRunner:
             return default_status
         
         try:
-            with open(status_file, 'r', encoding='utf-8') as f:
+            with open(status_file, 'r', encoding='utf-8-sig') as f:
                 status = json.load(f)
+            env_alive = cls.check_env_alive(simulation_id)
             return {
-                "status": status.get("status", "stopped"),
-                "twitter_available": status.get("twitter_available", False),
-                "reddit_available": status.get("reddit_available", False),
+                "status": status.get("status", "stopped") if env_alive else "stopped",
+                "twitter_available": status.get("twitter_available", False) if env_alive else False,
+                "reddit_available": status.get("reddit_available", False) if env_alive else False,
                 "timestamp": status.get("timestamp")
             }
         except (json.JSONDecodeError, OSError):
@@ -1459,7 +1483,7 @@ class SimulationRunner:
 
         ipc_client = SimulationIPCClient(sim_dir)
 
-        if not ipc_client.check_env_alive():
+        if not cls.check_env_alive(simulation_id):
             raise ValueError(f"模拟环境未运行或已关闭，无法执行Interview: {simulation_id}")
 
         logger.info(f"发送Interview命令: simulation_id={simulation_id}, agent_id={agent_id}, platform={platform}")
@@ -1521,7 +1545,7 @@ class SimulationRunner:
 
         ipc_client = SimulationIPCClient(sim_dir)
 
-        if not ipc_client.check_env_alive():
+        if not cls.check_env_alive(simulation_id):
             raise ValueError(f"模拟环境未运行或已关闭，无法执行Interview: {simulation_id}")
 
         logger.info(f"发送批量Interview命令: simulation_id={simulation_id}, count={len(interviews)}, platform={platform}")
@@ -1631,7 +1655,7 @@ class SimulationRunner:
         
         ipc_client = SimulationIPCClient(sim_dir)
         
-        if not ipc_client.check_env_alive():
+        if not cls.check_env_alive(simulation_id):
             return {
                 "success": True,
                 "message": "环境已经关闭"

backend/app/services/zep_tools.py

@@ -484,13 +484,14 @@ class ZepToolsService:
             SearchResult: 搜索结果
         """
         logger.info(t("console.graphSearch", graphId=graph_id, query=query[:50]))
+        zep_query = query[:400]
         
         # 尝试使用Zep Cloud Search API
         try:
             search_results = self._call_with_retry(
                 func=lambda: self.client.graph.search(
                     graph_id=graph_id,
-                    query=query,
+                    query=zep_query,
                     limit=limit,
                     scope=scope,
                     reranker="cross_encoder"

backend/app/utils/llm_client.py

@@ -29,7 +29,8 @@ class LLMClient:
         
         self.client = OpenAI(
             api_key=self.api_key,
-            base_url=self.base_url
+            base_url=self.base_url,
+            default_headers={"User-Agent": "python-requests/2.32.5"}
         )
     
     def chat(

backend/pyproject.toml

@@ -32,6 +32,7 @@ dependencies = [
     # 工具库
     "python-dotenv>=1.0.0",
     "pydantic>=2.0.0",
+    "psutil>=5.9.0",
 ]
 
 [project.optional-dependencies]

backend/requirements.txt

@@ -33,3 +33,4 @@ python-dotenv>=1.0.0
 
 # 数据验证
 pydantic>=2.0.0
+psutil>=5.9.0

backend/scripts/run_parallel_simulation.py

@@ -678,8 +678,9 @@ def fetch_new_actions_from_db(
     if not os.path.exists(db_path):
         return actions, new_last_rowid
     
+    conn = None
     try:
-        conn = sqlite3.connect(db_path)
+        conn = sqlite3.connect(db_path, timeout=5)
         cursor = conn.cursor()
         
         # 使用 rowid 来追踪已处理的记录（rowid 是 SQLite 的内置自增字段）
@@ -739,13 +740,35 @@ def fetch_new_actions_from_db(
                 'action_args': simplified_args,
             })
         
-        conn.close()
     except Exception as e:
         print(f"读取数据库动作失败: {e}")
+    finally:
+        if conn is not None:
+            conn.close()
     
     return actions, new_last_rowid
 
 
+def get_latest_trace_rowid(db_path: str) -> int:
+    """Return the latest processed trace row id for a platform DB."""
+    if not os.path.exists(db_path):
+        return 0
+
+    conn = None
+    try:
+        conn = sqlite3.connect(db_path, timeout=5)
+        cursor = conn.cursor()
+        cursor.execute("SELECT COALESCE(MAX(rowid), 0) FROM trace")
+        rowid = int(cursor.fetchone()[0] or 0)
+        return rowid
+    except Exception as e:
+        print(f"读取最新动作游标失败: {e}")
+        return 0
+    finally:
+        if conn is not None:
+            conn.close()
+
+
 def _enrich_action_context(
     cursor,
     action_type: str,
@@ -1034,6 +1057,9 @@ def create_model(config: Dict[str, Any], use_boost: bool = False):
     return ModelFactory.create(
         model_platform=ModelPlatformType.OPENAI,
         model_type=llm_model,
+        api_key=llm_api_key or None,
+        url=llm_base_url or None,
+        default_headers={"User-Agent": "python-requests/2.32.5"},
     )
 
 
@@ -1184,10 +1210,16 @@ async def run_twitter_simulation(
             content = post.get("content", "")
             try:
                 agent = result.env.agent_graph.get_agent(agent_id)
-                initial_actions[agent] = ManualAction(
+                manual_action = ManualAction(
                     action_type=ActionType.CREATE_POST,
                     action_args={"content": content}
                 )
+                if agent in initial_actions:
+                    if not isinstance(initial_actions[agent], list):
+                        initial_actions[agent] = [initial_actions[agent]]
+                    initial_actions[agent].append(manual_action)
+                else:
+                    initial_actions[agent] = manual_action
                 
                 if action_logger:
                     action_logger.log_action(
@@ -1204,7 +1236,9 @@ async def run_twitter_simulation(
         
         if initial_actions:
             await result.env.step(initial_actions)
-            log_info(f"已发布 {len(initial_actions)} 条初始帖子")
+            last_rowid = get_latest_trace_rowid(db_path)
+            posted_count = sum(len(action) if isinstance(action, list) else 1 for action in initial_actions.values())
+            log_info(f"已发布 {posted_count} 条初始帖子")
     
     # 记录 round 0 结束
     if action_logger:
@@ -1403,7 +1437,9 @@ async def run_reddit_simulation(
         
         if initial_actions:
             await result.env.step(initial_actions)
-            log_info(f"已发布 {len(initial_actions)} 条初始帖子")
+            last_rowid = get_latest_trace_rowid(db_path)
+            posted_count = sum(len(action) if isinstance(action, list) else 1 for action in initial_actions.values())
+            log_info(f"已发布 {posted_count} 条初始帖子")
     
     # 记录 round 0 结束
     if action_logger:

backend/scripts/run_reddit_simulation.py

@@ -464,6 +464,9 @@ class RedditSimulationRunner:
         return ModelFactory.create(
             model_platform=ModelPlatformType.OPENAI,
             model_type=llm_model,
+            api_key=llm_api_key or None,
+            url=llm_base_url or None,
+            default_headers={"User-Agent": "python-requests/2.32.5"},
         )
     
     def _get_active_agents_for_round(
@@ -617,7 +620,8 @@ class RedditSimulationRunner:
             
             if initial_actions:
                 await self.env.step(initial_actions)
-                print(f"  已发布 {len(initial_actions)} 条初始帖子")
+                posted_count = sum(len(action) if isinstance(action, list) else 1 for action in initial_actions.values())
+                print(f"  已发布 {posted_count} 条初始帖子")
         
         # 主模拟循环
         print("\n开始模拟循环...")

backend/scripts/run_twitter_simulation.py

@@ -457,6 +457,9 @@ class TwitterSimulationRunner:
         return ModelFactory.create(
             model_platform=ModelPlatformType.OPENAI,
             model_type=llm_model,
+            api_key=llm_api_key or None,
+            url=llm_base_url or None,
+            default_headers={"User-Agent": "python-requests/2.32.5"},
         )
     
     def _get_active_agents_for_round(
@@ -615,16 +618,23 @@ class TwitterSimulationRunner:
                 content = post.get("content", "")
                 try:
                     agent = self.env.agent_graph.get_agent(agent_id)
-                    initial_actions[agent] = ManualAction(
+                    manual_action = ManualAction(
                         action_type=ActionType.CREATE_POST,
                         action_args={"content": content}
                     )
+                    if agent in initial_actions:
+                        if not isinstance(initial_actions[agent], list):
+                            initial_actions[agent] = [initial_actions[agent]]
+                        initial_actions[agent].append(manual_action)
+                    else:
+                        initial_actions[agent] = manual_action
                 except Exception as e:
                     print(f"  警告: 无法为Agent {agent_id}创建初始帖子: {e}")
             
             if initial_actions:
                 await self.env.step(initial_actions)
-                print(f"  已发布 {len(initial_actions)} 条初始帖子")
+                posted_count = sum(len(action) if isinstance(action, list) else 1 for action in initial_actions.values())
+                print(f"  已发布 {posted_count} 条初始帖子")
         
         # 主模拟循环
         print("\n开始模拟循环...")

backend/uv.lock

@@ -995,6 +995,7 @@ dependencies = [
     { name = "flask" },
     { name = "flask-cors" },
     { name = "openai" },
+    { name = "psutil" },
     { name = "pydantic" },
     { name = "pymupdf" },
     { name = "python-dotenv" },
@@ -1024,6 +1025,7 @@ requires-dist = [
     { name = "flask-cors", specifier = ">=6.0.0" },
     { name = "openai", specifier = ">=1.0.0" },
     { name = "pipreqs", marker = "extra == 'dev'", specifier = ">=0.5.0" },
+    { name = "psutil", specifier = ">=5.9.0" },
     { name = "pydantic", specifier = ">=2.0.0" },
     { name = "pymupdf", specifier = ">=1.24.0" },
     { name = "pytest", marker = "extra == 'dev'", specifier = ">=8.0.0" },