diff --git a/.env.example b/.env.example
index d211818c..9b11faa8 100644
--- a/.env.example
+++ b/.env.example
@@ -26,3 +26,5 @@ SENTRY_DSN=
 # Configure these with your own Docker registry images
 DOCKER_IMAGE_BACKEND=backend
 DOCKER_IMAGE_FRONTEND=frontend
+
+MCP_IMAGE_PATH=/opt/sqlbot/images
diff --git a/backend/apps/api.py b/backend/apps/api.py
index c20d842c..9702c247 100644
--- a/backend/apps/api.py
+++ b/backend/apps/api.py
@@ -1,10 +1,11 @@
 from fastapi import APIRouter
 
-from apps.system.api import login, user, aimodel
-from apps.settings.api import terminology
-from apps.datasource.api import datasource
 from apps.chat.api import chat
 from apps.dashboard.api import dashboard_api
+from apps.datasource.api import datasource
+from apps.settings.api import terminology
+from apps.system.api import login, user, aimodel
+from apps.mcp import mcp
 
 api_router = APIRouter()
 api_router.include_router(login.router)
@@ -12,5 +13,9 @@ api_router.include_router(user.router)
 api_router.include_router(aimodel.router)
 api_router.include_router(terminology.router)
 api_router.include_router(datasource.router)
+# api_router.include_router(row_permission.router)
+# api_router.include_router(column_permission.router)
 api_router.include_router(chat.router)
 api_router.include_router(dashboard_api.router)
+api_router.include_router(mcp.router)
+
diff --git a/backend/apps/chat/api/chat.py b/backend/apps/chat/api/chat.py
index 8229b56d..32f338a7 100644
--- a/backend/apps/chat/api/chat.py
+++ b/backend/apps/chat/api/chat.py
@@ -1,20 +1,14 @@
 import traceback
-from typing import List
 
 import orjson
 from fastapi import APIRouter, HTTPException
 from fastapi.responses import StreamingResponse
-from sqlmodel import select
 
 from apps.chat.curd.chat import list_chats, get_chat_with_records, create_chat, rename_chat, \
-    delete_chat, list_records
-from apps.chat.models.chat_model import CreateChat, ChatRecord, RenameChat, Chat, ChatQuestion, ChatMcp
-from apps.chat.task.llm import LLMService
-from apps.datasource.crud.datasource import get_table_schema
-from apps.datasource.models.datasource import CoreDatasource
-from apps.system.crud.user import get_user_info
-from apps.system.models.system_model import AiModelDetail
-from common.core.deps import SessionDep, CurrentUser, get_current_user
+    delete_chat
+from apps.chat.models.chat_model import CreateChat, ChatRecord, RenameChat, ChatQuestion
+from apps.chat.task.llm import LLMService, run_task
+from common.core.deps import SessionDep, CurrentUser
 
 router = APIRouter(tags=["Data Q&A"], prefix="/chat")
 
@@ -57,18 +51,6 @@ async def delete(session: SessionDep, chart_id: int):
         )
 
 
-@router.post("/mcp_start", operation_id="mcp_start")
-async def mcp_start(session: SessionDep, chat: ChatMcp):
-    user = await get_current_user(session, chat.token)
-    return create_chat(session, user, CreateChat(), False)
-
-
-@router.post("/mcp_question", operation_id="mcp_question")
-async def mcp_question(session: SessionDep, chat: ChatMcp):
-    user = await get_current_user(session, chat.token)
-    return await stream_sql(session, user, chat)
-
-
 @router.post("/start")
 async def start_chat(session: SessionDep, current_user: CurrentUser, create_chat_obj: CreateChat):
     try:
@@ -80,7 +62,7 @@ async def start_chat(session: SessionDep, current_user: CurrentUser, create_chat
         )
 
 
-@router.post("/question", operation_id="question")
+@router.post("/question")
 async def stream_sql(session: SessionDep, current_user: CurrentUser, request_question: ChatQuestion):
     """Stream SQL analysis results
     
@@ -93,107 +75,17 @@ async def stream_sql(session: SessionDep, current_user: CurrentUser, request_que
         Streaming response with analysis results
     """
 
-    chat = session.query(Chat).filter(Chat.id == request_question.chat_id).first()
-    if not chat:
-        raise HTTPException(
-            status_code=400,
-            detail=f"Chat with id {request_question.chat_id} not found"
-        )
-    ds: CoreDatasource | None = None
-    if chat.datasource:
-        # Get available datasource
-        ds = session.query(CoreDatasource).filter(CoreDatasource.id == chat.datasource).first()
-        if not ds:
-            raise HTTPException(
-                status_code=500,
-                detail="No available datasource configuration found"
-            )
-
-        request_question.engine = ds.type_name if ds.type != 'excel' else 'PostgreSQL'
-
-    # Get available AI model
-    aimodel = session.exec(select(AiModelDetail).where(
-        AiModelDetail.status == True,
-        AiModelDetail.api_key.is_not(None)
-    )).first()
-    if not aimodel:
+    try:
+        llm_service = LLMService(session, current_user, request_question)
+        llm_service.init_record()
+    except Exception as e:
+        traceback.print_exc()
         raise HTTPException(
             status_code=500,
-            detail="No available AI model configuration found"
+            detail=str(e)
         )
 
-    history_records: List[ChatRecord] = list(filter(lambda r: True if r.first_chat != True else False,
-                                                    list_records(session=session, current_user=current_user,
-                                                                 chart_id=request_question.chat_id)))
-    # get schema
-    if ds:
-        request_question.db_schema = get_table_schema(session=session, ds=ds)
-
-    db_user = get_user_info(session=session, user_id=current_user.id)
-    request_question.lang = db_user.language
-
-    llm_service = LLMService(request_question, aimodel, history_records,
-                             CoreDatasource(**ds.model_dump()) if ds else None)
-
-    llm_service.init_record(session=session, current_user=current_user)
-
-    def run_task():
-        try:
-            # return id
-            yield orjson.dumps({'type': 'id', 'id': llm_service.get_record().id}).decode() + '\n\n'
-
-            # select datasource if datasource is none
-            if not ds:
-                ds_res = llm_service.select_datasource(session=session)
-                for chunk in ds_res:
-                    yield orjson.dumps({'content': chunk, 'type': 'datasource-result'}).decode() + '\n\n'
-                yield orjson.dumps({'id': llm_service.ds.id, 'datasource_name': llm_service.ds.name,
-                                    'engine_type': llm_service.ds.type_name, 'type': 'datasource'}).decode() + '\n\n'
-
-                llm_service.chat_question.db_schema = get_table_schema(session=session, ds=llm_service.ds)
-
-            # generate sql
-            sql_res = llm_service.generate_sql(session=session)
-            full_sql_text = ''
-            for chunk in sql_res:
-                full_sql_text += chunk
-                yield orjson.dumps({'content': chunk, 'type': 'sql-result'}).decode() + '\n\n'
-            yield orjson.dumps({'type': 'info', 'msg': 'sql generated'}).decode() + '\n\n'
-
-            # filter sql
-            print(full_sql_text)
-            sql = llm_service.check_save_sql(session=session, res=full_sql_text)
-            print(sql)
-            yield orjson.dumps({'content': sql, 'type': 'sql'}).decode() + '\n\n'
-
-            # execute sql
-            result = llm_service.execute_sql(sql=sql)
-            llm_service.save_sql_data(session=session, data_obj=result)
-            yield orjson.dumps({'content': orjson.dumps(result).decode(), 'type': 'sql-data'}).decode() + '\n\n'
-
-            # generate chart
-            chart_res = llm_service.generate_chart(session=session)
-            full_chart_text = ''
-            for chunk in chart_res:
-                full_chart_text += chunk
-                yield orjson.dumps({'content': chunk, 'type': 'chart-result'}).decode() + '\n\n'
-            yield orjson.dumps({'type': 'info', 'msg': 'chart generated'}).decode() + '\n\n'
-
-            # filter chart
-            print(full_chart_text)
-            chart = llm_service.check_save_chart(session=session, res=full_chart_text)
-            print(chart)
-            yield orjson.dumps({'content': orjson.dumps(chart).decode(), 'type': 'chart'}).decode() + '\n\n'
-
-            llm_service.finish(session=session)
-            yield orjson.dumps({'type': 'finish'}).decode() + '\n\n'
-
-        except Exception as e:
-            traceback.print_exc()
-            llm_service.save_error(session=session, message=str(e))
-            yield orjson.dumps({'content': str(e), 'type': 'error'}).decode() + '\n\n'
-
-    return StreamingResponse(run_task(), media_type="text/event-stream")
+    return StreamingResponse(run_task(llm_service, session), media_type="text/event-stream")
 
 
 @router.post("/record/{chart_record_id}/{action_type}")
@@ -217,35 +109,9 @@ async def analysis_or_predict(session: SessionDep, current_user: CurrentUser, ch
             detail=f"Chat record with id {chart_record_id} has not generated chart, do not support to analyze it"
         )
 
-    chat = session.query(Chat).filter(Chat.id == record.chat_id).first()
-    if not chat:
-        raise HTTPException(
-            status_code=400,
-            detail=f"Chat with id {record.chart_id} not found"
-        )
+    request_question = ChatQuestion(chat_id=record.chat_id, question='')
 
-    if chat.create_by != current_user.id:
-        raise HTTPException(
-            status_code=401,
-            detail=f"You cannot use the chat with id {record.chart_id}"
-        )
-
-    # Get available AI model
-    aimodel = session.exec(select(AiModelDetail).where(
-        AiModelDetail.status == True,
-        AiModelDetail.api_key.is_not(None)
-    )).first()
-    if not aimodel:
-        raise HTTPException(
-            status_code=500,
-            detail="No available AI model configuration found"
-        )
-
-    request_question = ChatQuestion(chat_id=chat.id, question='')
-    db_user = get_user_info(session=session, user_id=current_user.id)
-    request_question.lang = db_user.language
-
-    llm_service = LLMService(request_question, aimodel)
+    llm_service = LLMService(session, current_user, request_question)
     llm_service.set_record(record)
 
     def run_task():
@@ -261,14 +127,14 @@ async def analysis_or_predict(session: SessionDep, current_user: CurrentUser, ch
 
             elif action_type == 'predict':
                 # generate predict
-                analysis_res = llm_service.generate_predict(session=session)
+                analysis_res = llm_service.generate_predict()
                 full_text = ''
                 for chunk in analysis_res:
                     yield orjson.dumps({'content': chunk, 'type': 'predict-result'}).decode() + '\n\n'
                     full_text += chunk
                 yield orjson.dumps({'type': 'info', 'msg': 'predict generated'}).decode() + '\n\n'
 
-                _data = llm_service.check_save_predict_data(session=session, res=full_text)
+                _data = llm_service.check_save_predict_data(res=full_text)
                 yield orjson.dumps({'type': 'predict', 'content': _data}).decode() + '\n\n'
 
                 yield orjson.dumps({'type': 'predict_finish'}).decode() + '\n\n'
diff --git a/backend/apps/chat/task/llm.py b/backend/apps/chat/task/llm.py
index 8cc9d80c..5418beb6 100644
--- a/backend/apps/chat/task/llm.py
+++ b/backend/apps/chat/task/llm.py
@@ -2,7 +2,9 @@ import logging
 import warnings
 from typing import Any, List, Union, Dict
 
+import numpy as np
 import orjson
+import pandas as pd
 from langchain_community.utilities import SQLDatabase
 from langchain_core.language_models import BaseLLM
 from langchain_core.messages import BaseMessage, SystemMessage, HumanMessage, AIMessage, AIMessageChunk
@@ -13,10 +15,12 @@ from apps.ai_model.model_factory import LLMConfig, LLMFactory, get_llm_config
 from apps.chat.curd.chat import save_question, save_full_sql_message, save_full_sql_message_and_answer, save_sql, \
     save_error_message, save_sql_exec_data, save_full_chart_message, save_full_chart_message_and_answer, save_chart, \
     finish_record, save_full_analysis_message_and_answer, save_full_predict_message_and_answer, save_predict_data, \
-    save_full_select_datasource_message_and_answer
+    save_full_select_datasource_message_and_answer, list_records
 from apps.chat.models.chat_model import ChatQuestion, ChatRecord, Chat
+from apps.datasource.crud.datasource import get_table_schema
 from apps.datasource.models.datasource import CoreDatasource
 from apps.db.db import exec_sql
+from apps.system.crud.user import get_user_info
 from apps.system.models.system_model import AiModelDetail
 from common.core.deps import SessionDep, CurrentUser
 
@@ -34,13 +38,49 @@ class LLMService:
     sql_message: List[Union[BaseMessage, dict[str, Any]]] = []
     chart_message: List[Union[BaseMessage, dict[str, Any]]] = []
     history_records: List[ChatRecord] = []
+    session: SessionDep
+    _current_user: CurrentUser
 
-    def __init__(self, chat_question: ChatQuestion, aimodel: AiModelDetail, history_records: List[ChatRecord] = [],
-                 ds: CoreDatasource = None):
-        self.ds = ds
+    def __init__(self, session: SessionDep, current_user: CurrentUser, chat_question: ChatQuestion):
+
+        self.session = session
+        self.current_user = current_user
+
+        chat = self.session.query(Chat).filter(Chat.id == chat_question.chat_id).first()
+        if not chat:
+            raise Exception(f"Chat with id {chat_question.chat_id} not found")
+        ds: CoreDatasource | None = None
+        if chat.datasource:
+            # Get available datasource
+            ds = self.session.query(CoreDatasource).filter(CoreDatasource.id == chat.datasource).first()
+            if not ds:
+                raise Exception("No available datasource configuration found")
+
+            chat_question.engine = ds.type_name if ds.type != 'excel' else 'PostgreSQL'
+
+        # Get available AI model
+        aimodel = self.session.exec(select(AiModelDetail).where(
+            AiModelDetail.status == True,
+            AiModelDetail.api_key.is_not(None)
+        )).first()
+        if not aimodel and aimodel[0]:
+            raise Exception("No available AI model configuration found")
+
+        history_records: List[ChatRecord] = list(
+            map(lambda x: ChatRecord(**x.model_dump()), filter(lambda r: True if r.first_chat != True else False,
+                                                  list_records(session=self.session, current_user=current_user,
+                                                               chart_id=chat_question.chat_id))))
+        # get schema
+        if ds:
+            chat_question.db_schema = get_table_schema(session=self.session, ds=ds)
+
+        db_user = get_user_info(session=self.session, user_id=current_user.id)
+        chat_question.lang = db_user.language
+
+        self.ds = CoreDatasource(**ds.model_dump()) if ds else None
         self.chat_question = chat_question
-        self.chat_question.ai_modal_id = aimodel.id
-        self.config = get_llm_config(aimodel)
+        self.chat_question.ai_modal_id = aimodel[0].id
+        self.config = get_llm_config(aimodel[0])
 
         # Create LLM instance through factory
         llm_instance = LLMFactory.create_llm(self.config)
@@ -112,8 +152,8 @@ class LLMService:
                     _msg = AIMessage(content=last_chart_message['content'])
                     self.chart_message.append(_msg)
 
-    def init_record(self, session: SessionDep, current_user: CurrentUser) -> ChatRecord:
-        self.record = save_question(session=session, current_user=current_user, question=self.chat_question)
+    def init_record(self) -> ChatRecord:
+        self.record = save_question(session=self.session, current_user=self.current_user, question=self.chat_question)
         return self.record
 
     def get_record(self):
@@ -183,7 +223,7 @@ class LLMService:
                                                                                        in
                                                                                        analysis_msg]).decode())
 
-    def generate_predict(self, session: SessionDep):
+    def generate_predict(self):
         fields = self.get_fields_from_chart()
 
         self.chat_question.fields = orjson.dumps(fields).decode()
@@ -196,7 +236,7 @@ class LLMService:
         if self.record.full_predict_message and self.record.full_predict_message.strip() != '':
             history_msg = orjson.loads(self.record.full_predict_message)
 
-        self.record = save_full_predict_message_and_answer(session=session, record_id=self.record.id, answer='',
+        self.record = save_full_predict_message_and_answer(session=self.session, record_id=self.record.id, answer='',
                                                            data='',
                                                            full_message=orjson.dumps(history_msg +
                                                                                      [{'type': msg.type,
@@ -218,7 +258,7 @@ class LLMService:
                 continue
 
         predict_msg.append(AIMessage(full_predict_text))
-        self.record = save_full_predict_message_and_answer(session=session, record_id=self.record.id,
+        self.record = save_full_predict_message_and_answer(session=self.session, record_id=self.record.id,
                                                            answer=full_predict_text, data='',
                                                            full_message=orjson.dumps(history_msg +
                                                                                      [{'type': msg.type,
@@ -226,10 +266,10 @@ class LLMService:
                                                                                       in
                                                                                       predict_msg]).decode())
 
-    def select_datasource(self, session: SessionDep):
+    def select_datasource(self):
         datasource_msg: List[Union[BaseMessage, dict[str, Any]]] = []
         datasource_msg.append(SystemMessage(self.chat_question.datasource_sys_question()))
-        _ds_list = session.exec(select(CoreDatasource).options(
+        _ds_list = self.session.exec(select(CoreDatasource).options(
             load_only(CoreDatasource.id, CoreDatasource.name, CoreDatasource.description))).all()
         _ds_list_dict = []
         for _ds in _ds_list:
@@ -241,7 +281,7 @@ class LLMService:
         if self.record.full_select_datasource_message and self.record.full_select_datasource_message.strip() != '':
             history_msg = orjson.loads(self.record.full_select_datasource_message)
 
-        self.record = save_full_select_datasource_message_and_answer(session=session, record_id=self.record.id,
+        self.record = save_full_select_datasource_message_and_answer(session=self.session, record_id=self.record.id,
                                                                      answer='',
                                                                      full_message=orjson.dumps(history_msg +
                                                                                                [{'type': msg.type,
@@ -252,7 +292,6 @@ class LLMService:
         full_text = ''
         res = self.llm.stream(datasource_msg)
         for chunk in res:
-            print(chunk)
             if isinstance(chunk, dict):
                 full_text += chunk['content']
                 yield chunk['content']
@@ -273,7 +312,7 @@ class LLMService:
 
             if data['id'] and data['id'] != 0:
                 _datasource = data['id']
-                _ds = session.query(CoreDatasource).filter(CoreDatasource.id == _datasource).first()
+                _ds = self.session.query(CoreDatasource).filter(CoreDatasource.id == _datasource).first()
                 if not _ds:
                     _datasource = None
                     raise Exception(f"Datasource configuration with id {_datasource} not found")
@@ -281,14 +320,14 @@ class LLMService:
                 self.chat_question.engine = _ds.type_name if _ds.type != 'excel' else 'PostgreSQL'
                 _engine_type = self.chat_question.engine
                 # save chat
-                _chat = session.query(Chat).filter(Chat.id == self.record.chat_id).first()
+                _chat = self.session.query(Chat).filter(Chat.id == self.record.chat_id).first()
                 _chat.datasource = _datasource
                 _chat.engine_type = _ds.type_name
 
-                session.add(_chat)
-                session.flush()
-                session.refresh(_chat)
-                session.commit()
+                self.session.add(_chat)
+                self.session.flush()
+                self.session.refresh(_chat)
+                self.session.commit()
 
             elif data['fail']:
                 raise Exception(data['fail'])
@@ -298,7 +337,7 @@ class LLMService:
         except Exception as e:
             _error = e
 
-        self.record = save_full_select_datasource_message_and_answer(session=session, record_id=self.record.id,
+        self.record = save_full_select_datasource_message_and_answer(session=self.session, record_id=self.record.id,
                                                                      answer=full_text, datasource=_datasource,
                                                                      engine_type=_engine_type,
                                                                      full_message=orjson.dumps(history_msg +
@@ -312,10 +351,10 @@ class LLMService:
         if _error:
             raise _error
 
-    def generate_sql(self, session: SessionDep):
+    def generate_sql(self):
         # append current question
         self.sql_message.append(HumanMessage(self.chat_question.sql_user_question()))
-        self.record = save_full_sql_message(session=session, record_id=self.record.id,
+        self.record = save_full_sql_message(session=self.session, record_id=self.record.id,
                                             full_message=orjson.dumps(
                                                 [{'type': msg.type, 'content': msg.content} for msg in
                                                  self.sql_message]).decode())
@@ -333,16 +372,16 @@ class LLMService:
                 continue
 
         self.sql_message.append(AIMessage(full_sql_text))
-        self.record = save_full_sql_message_and_answer(session=session, record_id=self.record.id,
+        self.record = save_full_sql_message_and_answer(session=self.session, record_id=self.record.id,
                                                        answer=full_sql_text,
                                                        full_message=orjson.dumps(
                                                            [{'type': msg.type, 'content': msg.content} for msg in
                                                             self.sql_message]).decode())
 
-    def generate_chart(self, session: SessionDep):
+    def generate_chart(self):
         # append current question
         self.chart_message.append(HumanMessage(self.chat_question.chart_user_question()))
-        self.record = save_full_chart_message(session=session, record_id=self.record.id,
+        self.record = save_full_chart_message(session=self.session, record_id=self.record.id,
                                               full_message=orjson.dumps(
                                                   [{'type': msg.type, 'content': msg.content} for msg in
                                                    self.chart_message]).decode())
@@ -359,13 +398,13 @@ class LLMService:
                 continue
 
         self.chart_message.append(AIMessage(full_chart_text))
-        self.record = save_full_chart_message_and_answer(session=session, record_id=self.record.id,
+        self.record = save_full_chart_message_and_answer(session=self.session, record_id=self.record.id,
                                                          answer=full_chart_text,
                                                          full_message=orjson.dumps(
                                                              [{'type': msg.type, 'content': msg.content} for msg in
                                                               self.chart_message]).decode())
 
-    def check_save_sql(self, session: SessionDep, res: str) -> str:
+    def check_save_sql(self, res: str) -> str:
 
         json_str = extract_nested_json(res)
         data = orjson.loads(json_str)
@@ -385,13 +424,13 @@ class LLMService:
         if sql.strip() == '':
             raise Exception("SQL query is empty")
 
-        save_sql(session=session, sql=sql, record_id=self.record.id)
+        save_sql(session=self.session, sql=sql, record_id=self.record.id)
 
         self.chat_question.sql = sql
 
         return sql
 
-    def check_save_chart(self, session: SessionDep, res: str) -> Dict[str, Any]:
+    def check_save_chart(self, res: str) -> Dict[str, Any]:
 
         json_str = extract_nested_json(res)
         data = orjson.loads(json_str)
@@ -413,30 +452,30 @@ class LLMService:
         if error:
             raise Exception(message)
 
-        save_chart(session=session, chart=orjson.dumps(chart).decode(), record_id=self.record.id)
+        save_chart(session=self.session, chart=orjson.dumps(chart).decode(), record_id=self.record.id)
 
         return chart
 
-    def check_save_predict_data(self, session: SessionDep, res: str) -> Dict[str, Any]:
+    def check_save_predict_data(self, res: str) -> Dict[str, Any]:
 
         json_str = extract_nested_json(res)
 
         if not json_str:
             json_str = ''
 
-        save_predict_data(session=session, record_id=self.record.id, data=json_str)
+        save_predict_data(session=self.session, record_id=self.record.id, data=json_str)
 
         return json_str
 
-    def save_error(self, session: SessionDep, message: str):
-        return save_error_message(session=session, record_id=self.record.id, message=message)
+    def save_error(self, message: str):
+        return save_error_message(session=self.session, record_id=self.record.id, message=message)
 
-    def save_sql_data(self, session: SessionDep, data_obj: Dict[str, Any]):
-        return save_sql_exec_data(session=session, record_id=self.record.id,
+    def save_sql_data(self, data_obj: Dict[str, Any]):
+        return save_sql_exec_data(session=self.session, record_id=self.record.id,
                                   data=orjson.dumps(data_obj).decode())
 
-    def finish(self, session: SessionDep):
-        return finish_record(session=session, record_id=self.record.id)
+    def finish(self):
+        return finish_record(session=self.session, record_id=self.record.id)
 
     def execute_sql(self, sql: str):
         """Execute SQL query
@@ -503,3 +542,109 @@ def execute_sql_with_db(db: SQLDatabase, sql: str) -> str:
         error_msg = f"SQL execution failed: {str(e)}"
         logging.error(error_msg)
         raise RuntimeError(error_msg)
+
+
+def run_task(llm_service: LLMService, session: SessionDep, in_chat: bool = True):
+    try:
+        # return id
+        if in_chat:
+            yield orjson.dumps({'type': 'id', 'id': llm_service.get_record().id}).decode() + '\n\n'
+
+        # select datasource if datasource is none
+        if not llm_service.ds:
+            ds_res = llm_service.select_datasource()
+
+            for chunk in ds_res:
+                print(chunk)
+                if in_chat:
+                    yield orjson.dumps({'content': chunk, 'type': 'datasource-result'}).decode() + '\n\n'
+            if in_chat:
+                yield orjson.dumps({'id': llm_service.ds.id, 'datasource_name': llm_service.ds.name,
+                                    'engine_type': llm_service.ds.type_name, 'type': 'datasource'}).decode() + '\n\n'
+
+            llm_service.chat_question.db_schema = get_table_schema(session=session, ds=llm_service.ds)
+
+        # generate sql
+        sql_res = llm_service.generate_sql()
+        full_sql_text = ''
+        for chunk in sql_res:
+            full_sql_text += chunk
+            if in_chat:
+                yield orjson.dumps({'content': chunk, 'type': 'sql-result'}).decode() + '\n\n'
+        if in_chat:
+            yield orjson.dumps({'type': 'info', 'msg': 'sql generated'}).decode() + '\n\n'
+
+        # filter sql
+        print(full_sql_text)
+        sql = llm_service.check_save_sql(res=full_sql_text)
+        print(sql)
+        if in_chat:
+            yield orjson.dumps({'content': sql, 'type': 'sql'}).decode() + '\n\n'
+        else:
+            yield f'```sql\n{sql}\n```\n\n'
+
+        # execute sql
+        result = llm_service.execute_sql(sql=sql)
+        llm_service.save_sql_data(data_obj=result)
+        if in_chat:
+            yield orjson.dumps({'content': orjson.dumps(result).decode(), 'type': 'sql-data'}).decode() + '\n\n'
+
+        # generate chart
+        chart_res = llm_service.generate_chart()
+        full_chart_text = ''
+        for chunk in chart_res:
+            full_chart_text += chunk
+            if in_chat:
+                yield orjson.dumps({'content': chunk, 'type': 'chart-result'}).decode() + '\n\n'
+        if in_chat:
+            yield orjson.dumps({'type': 'info', 'msg': 'chart generated'}).decode() + '\n\n'
+
+        # filter chart
+        print(full_chart_text)
+        chart = llm_service.check_save_chart(res=full_chart_text)
+        print(chart)
+        if in_chat:
+            yield orjson.dumps({'content': orjson.dumps(chart).decode(), 'type': 'chart'}).decode() + '\n\n'
+        else:
+            data = []
+            _fields = {}
+            if chart.get('columns'):
+                for _column in chart.get('columns'):
+                    if _column:
+                        _fields[_column.get('value')] = _column.get('name')
+            if chart.get('axis'):
+                if chart.get('axis').get('x'):
+                    _fields[chart.get('axis').get('x').get('value')] = chart.get('axis').get('x').get('name')
+                if chart.get('axis').get('y'):
+                    _fields[chart.get('axis').get('y').get('value')] = chart.get('axis').get('y').get('name')
+                if chart.get('axis').get('series'):
+                    _fields[chart.get('axis').get('series').get('value')] = chart.get('axis').get('series').get('name')
+            _fields_list = []
+            _fields_skip = False
+            for _data in result.get('data'):
+                _row = []
+                for field in result.get('fields'):
+                    _row.append(_data.get(field))
+                    if not _fields_skip:
+                        _fields_list.append(field if not _fields.get(field) else _fields.get(field))
+                data.append(_row)
+                _fields_skip = True
+            df = pd.DataFrame(np.array(data), columns=_fields_list)
+            markdown_table = df.to_markdown(index=False)
+            yield markdown_table + '\n\n'
+
+        record = llm_service.finish()
+        if in_chat:
+            yield orjson.dumps({'type': 'finish'}).decode() + '\n\n'
+        else:
+            # todo generate picture
+            if chart['type'] != 'table':
+                yield '# todo generate chart picture'
+
+            yield f'![{chart["type"]}](https://sqlbot.fit2cloud.cn/images/111.png)'
+    except Exception as e:
+        llm_service.save_error(message=str(e))
+        if in_chat:
+            yield orjson.dumps({'content': str(e), 'type': 'error'}).decode() + '\n\n'
+        else:
+            yield f'> ❌ **ERROR**\n\n> \n\n> {str(e)}。'
diff --git a/backend/apps/datasource/api/datasource.py b/backend/apps/datasource/api/datasource.py
index 08f2e6f7..b0921673 100644
--- a/backend/apps/datasource/api/datasource.py
+++ b/backend/apps/datasource/api/datasource.py
@@ -18,7 +18,7 @@ router = APIRouter(tags=["datasource"], prefix="/datasource")
 path = "/opt/sqlbot/data/excel"
 
 
-@router.get("/list", operation_id="get_datasource_list")
+@router.get("/list")
 async def datasource_list(session: SessionDep):
     return get_datasource_list(session=session)
 
diff --git a/backend/apps/mcp/__init__.py b/backend/apps/mcp/__init__.py
new file mode 100644
index 00000000..a14a79d9
--- /dev/null
+++ b/backend/apps/mcp/__init__.py
@@ -0,0 +1,2 @@
+# Author: Junjun
+# Date: 2025/7/1
diff --git a/backend/apps/mcp/mcp.py b/backend/apps/mcp/mcp.py
new file mode 100644
index 00000000..a9c266b1
--- /dev/null
+++ b/backend/apps/mcp/mcp.py
@@ -0,0 +1,63 @@
+# Author: Junjun
+# Date: 2025/7/1
+
+from datetime import timedelta
+from typing import Annotated
+
+from fastapi import APIRouter, Depends, HTTPException
+from fastapi.responses import StreamingResponse
+from fastapi.security import OAuth2PasswordRequestForm
+
+from apps.chat.api.chat import create_chat
+from apps.chat.models.chat_model import ChatMcp, CreateChat
+from apps.chat.task.llm import LLMService, run_task
+from apps.datasource.crud.datasource import get_datasource_list
+from apps.system.crud.user import authenticate
+from apps.system.models.system_model import AiModelDetail
+from common.core.config import settings
+from common.core.deps import SessionDep, get_current_user
+from common.core.schemas import Token
+from common.core.security import create_access_token
+
+router = APIRouter(tags=["mcp"], prefix="/mcp")
+
+
+@router.post("/access_token", operation_id="access_token")
+def local_login(
+        session: SessionDep,
+        form_data: Annotated[OAuth2PasswordRequestForm, Depends()]
+) -> Token:
+    user = authenticate(session=session, account=form_data.username, password=form_data.password)
+    if not user:
+        raise HTTPException(status_code=400, detail="Incorrect account or password")
+    access_token_expires = timedelta(minutes=settings.ACCESS_TOKEN_EXPIRE_MINUTES)
+    user_dict = user.to_dict()
+    return Token(access_token=create_access_token(
+        user_dict, expires_delta=access_token_expires
+    ))
+
+
+@router.get("/ds_list", operation_id="get_datasource_list")
+async def datasource_list(session: SessionDep):
+    return get_datasource_list(session=session)
+
+
+@router.get("/model_list", operation_id="get_model_list")
+async def get_model_list(session: SessionDep):
+    return session.query(AiModelDetail).all()
+
+
+@router.post("/mcp_start", operation_id="mcp_start")
+async def mcp_start(session: SessionDep, chat: ChatMcp):
+    user = await get_current_user(session, chat.token)
+    return create_chat(session, user, CreateChat(), False)
+
+
+@router.post("/mcp_question", operation_id="mcp_question")
+async def mcp_question(session: SessionDep, chat: ChatMcp):
+    user = await get_current_user(session, chat.token)
+
+    llm_service = LLMService(session, user, chat)
+    llm_service.init_record()
+
+    return StreamingResponse(run_task(llm_service, session, False), media_type="text/event-stream")
diff --git a/backend/apps/system/api/aimodel.py b/backend/apps/system/api/aimodel.py
index a3bf6f21..eabc1fd6 100644
--- a/backend/apps/system/api/aimodel.py
+++ b/backend/apps/system/api/aimodel.py
@@ -88,8 +88,3 @@ async def status(session: SessionDep, dto: model_status):
         session.add(term)
     session.commit()
     return {"message": f"AiModel with IDs {ids} updated successfully."}
-
-
-@router.get("/list", operation_id="get_model_list")
-async def get_model_list(session: SessionDep):
-    return session.query(AiModelDetail).all()
diff --git a/backend/common/core/config.py b/backend/common/core/config.py
index 078f682e..b54fa12c 100644
--- a/backend/common/core/config.py
+++ b/backend/common/core/config.py
@@ -79,6 +79,7 @@ class Settings(BaseSettings):
     SMTP_PASSWORD: str | None = None
     EMAILS_FROM_EMAIL: EmailStr | None = None
     EMAILS_FROM_NAME: EmailStr | None = None
+    MCP_IMAGE_PATH: str
 
     @model_validator(mode="after")
     def _set_default_emails_from(self) -> Self:
diff --git a/backend/common/utils/whitelist.py b/backend/common/utils/whitelist.py
index 33b621fe..abf414bb 100644
--- a/backend/common/utils/whitelist.py
+++ b/backend/common/utils/whitelist.py
@@ -21,14 +21,11 @@ wlist = [
     "*.ttf",
     "*.eot",
     "*.otf",
-    "/mcp",
-    "/mcp/message",
-    "/datasource/list",
-    "/system/aimodel/list",
-    "/chat/mcp_question",
-    "/chat/mcp_start",
+    "/mcp*",
     "/system/license",
-    "/system/config/key"
+    "/system/config/key",
+    "/images/*",
+    "/sse",
 ]
 
 class WhitelistChecker:
diff --git a/backend/main.py b/backend/main.py
index 4131bb3b..85d14420 100644
--- a/backend/main.py
+++ b/backend/main.py
@@ -12,6 +12,7 @@ from common.core.response_middleware import ResponseMiddleware, exception_handle
 from alembic.config import Config
 from alembic import command
 from fastapi_mcp import FastApiMCP
+from fastapi.staticfiles import StaticFiles
 import sqlbot_xpack
 
 def run_migrations():
@@ -39,6 +40,8 @@ app = FastAPI(
     generate_unique_id_function=custom_generate_unique_id,
     lifespan=lifespan
 )
+# mcp server, images path
+app.mount("/images", StaticFiles(directory=settings.MCP_IMAGE_PATH), name="images")
 
 mcp_app = FastAPI()
 
@@ -48,7 +51,7 @@ mcp = FastApiMCP(
     description="SQLBot MCP Server",
     describe_all_responses=True,
     describe_full_response_schema=True,
-    include_operations=["get_datasource_list", "get_model_list", "mcp_question", "mcp_start"]
+    include_operations=["get_datasource_list", "get_model_list", "mcp_question", "mcp_start", "access_token"]
 )
 
 mcp.mount(mcp_app)
diff --git a/backend/pyproject.toml b/backend/pyproject.toml
index 5e4981f8..2f894061 100644
--- a/backend/pyproject.toml
+++ b/backend/pyproject.toml
@@ -35,14 +35,15 @@ dependencies = [
     "oracledb (>=3.1.1,<4.0.0)",
     "pyyaml (>=6.0.2,<7.0.0)",
     "fastapi-mcp (>=0.3.4,<0.4.0)",
-    "sqlbot-xpack==0.0.3",
+    "tabulate>=0.9.0",
+    "sqlbot-xpack==0.0.3.2",
 ]
 [[tool.uv.index]]
 name = "default"
 url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple"
 default = true
-[[tool.uv.index]]
 
+[[tool.uv.index]]
 name = "testpypi"
 url = "https://test.pypi.org/simple"
 explicit = true
diff --git a/frontend/src/router/watch.ts b/frontend/src/router/watch.ts
index 98224f6b..075140db 100644
--- a/frontend/src/router/watch.ts
+++ b/frontend/src/router/watch.ts
@@ -38,9 +38,19 @@ const loadXpackStatic = () => {
     return Promise.resolve()
   }
   const url = '/xpack_static/license-generator.umd.js'
-  return request.loadRemoteScript(url, 'sqlbot_xpack_static', () => {
-    // eslint-disable-next-line @typescript-eslint/ban-ts-comment
-    // @ts-ignore
-    LicenseGenerator?.init(import.meta.env.VITE_API_BASE_URL)
+  return new Promise((resolve, reject) => {
+    request
+      .loadRemoteScript(url, 'sqlbot_xpack_static', () => {
+        // eslint-disable-next-line @typescript-eslint/ban-ts-comment
+        // @ts-ignore
+        LicenseGenerator?.init(import.meta.env.VITE_API_BASE_URL).then(() => {
+          resolve(true)
+        })
+      })
+      .catch((error) => {
+        console.error('Failed to load xpack_static script:', error)
+        ElMessage.error('Failed to load license generator script')
+        reject(error)
+      })
   })
 }