diff --git a/case_utils/assert_handle.py b/case_utils/assert_handle.py
index 47c768d..8c6a92d 100644
--- a/case_utils/assert_handle.py
+++ b/case_utils/assert_handle.py
@@ -10,11 +10,12 @@
 from requests import Response
 from case_utils.data_handle import json_extractor, re_extract
 from loguru import logger
-from case_utils.requests_handle import response_type
-from case_utils.data_handle import eval_data_process
+from case_utils.request_data_handle import response_type
+from common_utils.mysql_handle import MysqlServer
+from config.settings import db_info
 
 
-def assert_result(response: Response, expected: dict) -> None:
+def assert_response(response: Response, expected: dict) -> None:
     """ 断言方法
     :param response: 实际响应对象
     :param expected: 预期响应内容，从excel中或者yaml读取、或者手动传入,格式如下：
@@ -27,9 +28,9 @@ def assert_result(response: Response, expected: dict) -> None:
     return None
     """
     if expected is None:
-        logger.info("当前用例无断言！")
+        logger.info("当前用例无响应断言！")
         return
-    logger.info(f"预期结果：{expected}, {type(expected)}")
+    logger.debug(f"响应断言预期结果：{expected}, {type(expected)}")
     index = 0
     for k, v in expected.items():
         # 获取需要断言的实际结果部分
@@ -37,7 +38,6 @@ def assert_result(response: Response, expected: dict) -> None:
             if _k == "http_code":
                 actual = response.status_code
             else:
-                logger.debug("根据响应类型的不同，从响应数据中提取实际结果")
                 if response_type(response) == "json":
                     # 如果响应数据是json格式
                     actual = json_extractor(response.json(), _k)
@@ -45,27 +45,70 @@ def assert_result(response: Response, expected: dict) -> None:
                     # 响应数据不是json格式
                     actual = re_extract(response.text, _k)
             index += 1
-            # 对预期结果进行数据处理
-            _v = eval_data_process(_v)
-            logger.info(f'第{index}个断言 -|- 预期结果: {_v}, {type(_v)}   {k}   实际结果: {actual},{type(actual)}')
+            logger.info(f'第{index}个响应断言 -|- 预期结果: {_v}, {type(_v)}   {k}   实际结果: {actual}, {type(actual)}')
             try:
                 if k == "eq":  # 预期结果 = 实际结果
                     assert _v == actual
-                    logger.info("断言成功！")
+                    logger.info(f"预期结果: {_v} ==  实际结果: {actual}, 断言成功！")
                 elif k == "in":  # 实际结果 包含 预期结果
                     assert _v in actual
-                    logger.info("断言成功！")
+                    logger.info(f"预期结果: {_v} in  实际结果: {actual}, 断言成功！")
                 elif k == "gt":  # 预期结果 > 实际结果 (值应该为数值型)
-                    assert actual < _v
-                    logger.info("断言成功！")
+                    assert _v > actual
+                    logger.info(f"预期结果: {_v} >  实际结果: {actual}, 断言成功！")
                 elif k == "lt":  # 预期结果 < 实际结果 (值应该为数值型)
-                    assert actual > _v
-                    logger.info("断言成功！")
+                    assert _v < actual
+                    logger.info(f"预期结果: {_v} <  实际结果: {actual}, 断言成功！")
                 elif k == "not":  # 预期结果 != 实际结果
-                    assert actual != _v
-                    logger.info("断言成功！")
+                    assert _v != actual
+                    logger.info(f"预期结果: {_v} !=  实际结果: {actual}, 断言成功！")
                 else:
-                    logger.error(f"判断关键字: {k} 错误！")
+                    logger.error(f"判断关键字: {k} 错误！, 目前仅支持如下关键字：eq, in, gt, lt, not")
             except AssertionError:
-                logger.error("断言失败")
-                raise AssertionError(f"第{index}个断言 -|- 预期结果: {_v}, {type(_v)}   {k}   实际结果: {actual},{type(actual)}")
+                logger.error(f"第{index}个响应断言失败 -|- 预期结果: {_v}, {type(_v)}   {k}   实际结果: {actual}, {type(actual)}")
+                raise AssertionError(
+                    f"第{index}个响应断言失败 -|- 预期结果: {_v}, {type(_v)}   {k}   实际结果: {actual}, {type(actual)}")
+
+
+def assert_sql(env, expected: dict):
+    """
+    数据库断言
+    :param env: 当前所处环境
+    :param expected: 预期结果，从excel中或者yaml读取、或者手动传入,格式如下：
+    {
+    'eq':
+        {'sql': 'select count(*) from users where user_id=1;', 'len': '1'},
+    'eq':
+        {'sql': 'select * from users where user_id=1;', '$.username': '${username}'},
+    }
+    """
+    if expected is None:
+        logger.info("当前用例无数据库断言！")
+        return
+    try:
+        # 拿不到数据库配置，则不进行数据库断言
+        db = db_info["test" if env.lower() == "test" else "live"]
+        db_host = db["db_host"]
+    except KeyError:
+        logger.error("当前环境无数据库配置，跳过数据库断言！")
+        return
+    for k, v in expected.items():
+        sql_result = None
+        for _k, _v in v.items():
+            if _k == "sql":
+                # 查询数据库，获取查询结果
+                sql_result = MysqlServer(**db).query_one(_v)
+                logger.info(f'数据库响应断言 -|- SQL：{_v} || 查询结果：{sql_result}')
+            try:
+                if k == "eq":  # 预期结果 = 实际结果
+                    if _k == "len":
+                        assert _v == len(sql_result)
+                        logger.info(f"预期结果: {_v} ==  实际结果: {len(sql_result)}, 断言成功！")
+                    # 如果时$.开头，则从数据库查询结果中提取相应的值作为实际结果
+                    elif _k.startswith("$."):
+                        actual = json_extractor(sql_result, _k)
+                        assert _v == actual
+                        logger.info(f"预期结果: {_v} ==  实际结果: {actual}, 断言成功！")
+            except AssertionError:
+                logger.error(f"数据库断言失败 -|- 预期结果:{_v} ==  实际结果: {len(sql_result)}")
+                raise AssertionError(f"数据库断言失败 -|-预期结果: {_v} ==  实际结果: {len(sql_result)}")
diff --git a/common_utils/mysql_handle.py b/common_utils/mysql_handle.py
new file mode 100644
index 0000000..c36f590
--- /dev/null
+++ b/common_utils/mysql_handle.py
@@ -0,0 +1,152 @@
+# -*- coding: utf-8 -*-
+# @Time    : 2023/5/12 11:04
+# @Author  : chenyinhua
+# @File    : mysql_handle.py
+# @Software: PyCharm
+# @Desc: 使用pymysql模块连接mysql数据库的公共方法
+import pymysql
+from typing import Union
+import json
+from datetime import datetime
+from sshtunnel import SSHTunnelForwarder
+from loguru import logger
+
+
+class MysqlServer:
+    """
+    初始化数据库连接(支持通过SSH隧道的方式连接)，并指定查询的结果集以字典形式返回
+    """
+
+    def __init__(self, db_host, db_port, db_user, db_pwd, db_database, ssh=False,
+                 **kwargs):
+        """
+        初始化方法中， 连接mysql数据库， 根据ssh参数决定是否走SSH隧道方式连接mysql数据库
+        """
+        self.server = None
+        if ssh:
+            self.server = SSHTunnelForwarder(
+                ssh_address_or_host=(kwargs.get("ssh_host"), kwargs.get("ssh_port")),  # ssh 目标服务器 ip 和 port
+                ssh_username=kwargs.get("ssh_user"),  # ssh 目标服务器用户名
+                ssh_password=kwargs.get("ssh_pwd"),  # ssh 目标服务器用户密码
+                remote_bind_address=(db_host, db_port),  # mysql 服务ip 和 part
+                local_bind_address=('127.0.0.1', 5143),  # ssh 目标服务器的用于连接 mysql 或 redis 的端口，该 ip 必须为 127.0.0.1
+            )
+            self.server.start()
+            db_host = self.server.local_bind_host  # server.local_bind_host 是 参数 local_bind_address 的 ip
+            db_port = self.server.local_bind_port  # server.local_bind_port 是 参数 local_bind_address 的 port
+        # 建立连接
+        self.conn = pymysql.connect(host=db_host,
+                                    port=db_port,
+                                    user=db_user,
+                                    password=db_pwd,
+                                    database=db_database,
+                                    charset="utf8",
+                                    cursorclass=pymysql.cursors.DictCursor  # 加上pymysql.cursors.DictCursor这个返回的就是字典
+                                    )
+        # 创建一个游标对象
+        self.cursor = self.conn.cursor()
+
+    def query_all(self, sql):
+        """
+        查询所有符合sql条件的数据
+        :param sql: 执行的sql
+        :return: 查询结果
+        """
+        try:
+            self.conn.commit()
+            self.cursor.execute(sql)
+            data = self.cursor.fetchall()
+            # 关闭数据库链接和隧道
+            self.close()
+            return self.verify(data)
+        except Exception as e:
+            logger.error(f"查询所有符合sql条件的数据报错: {e}")
+            raise e
+
+    def query_one(self, sql):
+        """
+        查询符合sql条件的数据的第一条数据
+        :param sql: 执行的sql
+        :return: 返回查询结果的第一条数据
+        """
+        try:
+            self.conn.commit()
+            self.cursor.execute(sql)
+            data = self.cursor.fetchone()
+            # 关闭数据库链接和隧道
+            self.close()
+            return self.verify(data)
+        except Exception as e:
+            logger.error(f"{sql} --> 报错: {e}")
+            raise e
+
+    def insert(self, sql):
+        """
+        插入数据
+        :param sql: 执行的sql
+        """
+        try:
+            self.cursor.execute(sql)
+            # 提交  只要数据库更新就要commit
+            self.conn.commit()
+            # 关闭数据库链接和隧道
+            self.close()
+        except Exception as e:
+            logger.error(f"{sql} --> 报错: {e}")
+            raise e
+
+    def update(self, sql):
+        """
+        更新数据
+        :param sql: 执行的sql
+        """
+        try:
+            self.cursor.execute(sql)
+            # 提交 只要数据库更新就要commit
+            self.conn.commit()
+            # 关闭数据库链接和隧道
+            self.close()
+        except Exception as e:
+            logger.error(f"{sql} --> 报错: {e}")
+            raise e
+
+    def query(self, sql, one=True):
+        """
+        根据传值决定查询一条数据还是所有
+        :param sql: 查询的SQL语句
+        :param one: 默认True. True查一条数据，否则查所有
+        :return:
+        """
+        try:
+            if one:
+                return self.query_one(sql)
+            else:
+                return self.query_all(sql)
+        except Exception as e:
+            logger.error(f"{sql} --> 报错: {e}")
+            raise e
+
+    def close(self):
+        """
+        断开游标，关闭数据库
+        如果开启了SSH隧道，也关闭
+        :return:
+        """
+        # 关闭游标
+        self.cursor.close()
+        # 关闭数据库链接
+        self.conn.close()
+        # 如果开启了SSH隧道，则关闭
+        if self.server:
+            self.server.close()
+
+    def verify(self, result: dict) -> Union[dict, None]:
+        """验证结果能否被json.dumps序列化"""
+        # 尝试变成字符串，解决datetime 无法被json 序列化问题
+        try:
+            json.dumps(result)
+        except TypeError:  # TypeError: Object of type datetime is not JSON serializable
+            for k, v in result.items():
+                if isinstance(v, datetime):
+                    result[k] = str(v)
+        return result
diff --git a/config/case_template.txt b/config/case_template.txt
index b9e1417..9f3bf28 100644
--- a/config/case_template.txt
+++ b/config/case_template.txt
@@ -1,8 +1,8 @@
 import pytest
-from case_utils.assert_handle import assert_result
-from case_utils.requests_handle import BaseRequest
+from case_utils.assert_handle import assert_response, assert_sql
+from common_utils.base_request import BaseRequest
 from loguru import logger
-from case_utils.request_data_handle import RequestPreDataHandle
+from case_utils.request_data_handle import RequestPreDataHandle, after_extract, case_data_replace,eval_data_process
 from pytest_html import extras  # 往pytest-html报告中填写额外的内容
 from common_utils.func_handle import add_docstring
 
@@ -16,28 +16,38 @@ def case(request):
 
 @pytest.mark.${func_title}
 @pytest.mark.auto
-def ${func_title}_auto(case, extra):
-    """
-
-    """
+def ${func_title}_auto(case, extra, request):
     logger.info("-----------------------------START-开始执行用例-----------------------------")
     logger.debug(f"当前执行的用例数据:{case}")
-    # 给当前测试方法添加文档注释
-    add_docstring(case.get("title", ""))(${func_title}_auto)
-    if case.get("run", None):
-        # 获取处理完成的用例数据
-        case_data = RequestPreDataHandle(case).request_data_handle()
-        # 将用例数据显示在pytest-html报告中
-        extra.append(extras.text(str(case_data), name="用例数据"))
-        # 发送请求
-        response = BaseRequest.send_request(case_data)
-        # 将响应数据显示在pytest-html报告中
-        extra.append(extras.text(str(response.text), name="响应数据"))
-        # 进行断言
-        assert_result(response, case_data["expected"])
-    else:
-        reason = f"标记了该用例为false，不执行\\n"
-        logger.warning(f"{reason}")
-        pytest.skip(reason)
-    logger.info("------------------------------------------用例执行结束------------------------------------------")
+    try:
+        # 获取命令行参数，判断当前处于哪个环境
+        env = request.config.getoption("--env")
+        # 给当前测试方法添加文档注释
+        add_docstring(case.get("title", ""))(${func_title}_auto)
+        if case.get("run", None):
+             # 处理请求前的用例数据
+            case_data = RequestPreDataHandle(case).request_data_handle()
+            # 将用例数据显示在pytest-html报告中
+            extra.append(extras.text(str(case_data), name="用例数据"))
+            # 发送请求
+            response = BaseRequest.send_request(case_data)
+            # 将响应数据显示在pytest-html报告中
+            extra.append(extras.text(str(response.text), name="响应数据"))
+            # 请求后，提取后置参数作为全局变量
+            after_extract(response, case_data["extract"])
+            # 从全局变量中获取最新值，替换数据库断言中的参数
+            case_data["assert_sql"] = eval_data_process(case_data_replace(case_data["assert_sql"]))
+            # 进行响应断言
+            assert_response(response, case_data["assert_response"])
+             # 进行数据库断言
+            assert_sql(env, case_data["assert_sql"])
+        else:
+            reason = f"标记了该用例为false，不执行\\n"
+            logger.warning(f"{reason}")
+            pytest.skip(reason)
+    except Exception as e:
+        logger.error(f"用例执行过程中报错：{e}")
+        raise e
+    finally:
+        logger.info("-----------------------------END-用例执行结束-----------------------------")
 
diff --git a/config/settings.py b/config/settings.py
index 6fa4ae2..e0a6e12 100644
--- a/config/settings.py
+++ b/config/settings.py
@@ -68,4 +68,26 @@ ding_talk = {
 # ------------------------------------ 企业微信相关配置 ----------------------------------------------------#
 wechat = {
     "webhook_url": "https://qyapi.weixin.qq.com/cgi-bin/webhook/send?key=********",
+}
+
+# ------------------------------------ 数据库相关配置 ----------------------------------------------------#
+db_info = {
+    "test": {
+         "db_host": "xx.xx.xx.xx",
+        "db_port": 3306,
+        "db_user": "root",
+        "db_pwd": "**********",
+        "db_database": "test**********",
+        "ssh": True,
+        "ssh_host": "xx.xx.xx.xx",
+        "ssh_port": 3306,
+        "ssh_user": "root",
+        "ssh_pwd": "**********"
+
+    },
+    "live": {
+
+
+    }
+
 }
\ No newline at end of file
diff --git a/data/test_login.yaml b/data/test_login_demo.yaml
similarity index 87%
rename from data/test_login.yaml
rename to data/test_login_demo.yaml
index 77d0c5c..cb8b5a9 100644
--- a/data/test_login.yaml
+++ b/data/test_login_demo.yaml
@@ -24,6 +24,10 @@
       $.user_id: 84953
     not:
       $.user_id: 85390
+  assert_sql:
+    eq:
+      sql: select count(*) from tokens where user_id=${user_id};
+      len: 1
 - feature: 登录
   title: 用户名正确，密码错误，登录失败
   url: /api/accounts/login.json
@@ -38,4 +42,5 @@
   assert_response:
     eq:
       http_code: 200
-      $.status: -2
\ No newline at end of file
+      $.status: -2
+  assert_sql:
\ No newline at end of file
diff --git a/data/test_login_excel.xlsx b/data/test_login_excel_demo.xlsx
similarity index 59%
rename from data/test_login_excel.xlsx
rename to data/test_login_excel_demo.xlsx
index f29e5a8..b6613f4 100644
Binary files a/data/test_login_excel.xlsx and b/data/test_login_excel_demo.xlsx differ
diff --git a/data/test_new_project.yaml b/data/test_new_project_demo.yaml
similarity index 68%
rename from data/test_new_project.yaml
rename to data/test_new_project_demo.yaml
index e095a09..2b14b2c 100644
--- a/data/test_new_project.yaml
+++ b/data/test_new_project_demo.yaml
@@ -18,4 +18,10 @@
   assert_response:
     eq:
       http_code: 200
-      $.login: ${login}
\ No newline at end of file
+      $.login: ${login}
+  assert_sql:
+    eq:
+      sql: select id, `name`, identifier from projects where user_id=${user_id} ORDER BY created_on DESC;
+      $.id: ${project_id}
+      $.name: ${project_name}
+      $.identifier: ${project_identifier}
\ No newline at end of file
diff --git a/test_case/test_manual_case/test_login.py b/test_case/test_manual_case/test_login.py
deleted file mode 100644
index 4750a90..0000000
--- a/test_case/test_manual_case/test_login.py
+++ /dev/null
@@ -1,47 +0,0 @@
-# -*- coding: utf-8 -*-
-# @Version: Python 3.9
-# @Time    : 2023/1/9 16:41
-# @Author  : chenyinhua
-# @File    : test_login.py
-# @Software: PyCharm
-# @Desc: python脚本编写的测试用例文件
-
-
-import pytest
-import os
-from common_utils.yaml_handle import HandleYaml
-from config.project_path import DATA_DIR
-from case_utils.requests_handle import BaseRequest
-from case_utils.assert_handle import assert_result
-from loguru import logger
-from case_utils.request_data_handle import RequestPreDataHandle
-from config.global_vars import GLOBAL_VARS
-
-# 读取用例数据
-cases = HandleYaml(filename=os.path.join(DATA_DIR, "test_login.yaml")).read_yaml
-
-
-@pytest.mark.test_login
-class TestLogin:
-    """
-    登录模块
-    """
-
-    @pytest.mark.parametrize("case", cases)
-    def test_login(self, case):
-        logger.info("-----------------------------START-开始执行用例-----------------------------")
-        logger.debug(f"当前执行的用例数据:{case}")
-        # 将用例标题title作为全局变量，方便后续写入到测试报告report.description中
-        GLOBAL_VARS["title"] = case.get("title", "")
-        if case.get("run", None):
-            # 获取处理完成的用例数据
-            case_data = RequestPreDataHandle(case).request_data_handle()
-            # 发送请求
-            response = BaseRequest.send_request(case_data)
-            # 进行断言
-            assert_result(response, case_data["expected"])
-        else:
-            reason = f"标记了该用例为false，不执行\\n"
-            logger.warning(f"{reason}")
-            pytest.skip(reason)
-        logger.info("-----------------------------END-用例执行结束-----------------------------")
diff --git a/test_case/test_manual_case/test_login_demo.py b/test_case/test_manual_case/test_login_demo.py
new file mode 100644
index 0000000..6d1a30f
--- /dev/null
+++ b/test_case/test_manual_case/test_login_demo.py
@@ -0,0 +1,60 @@
+# -*- coding: utf-8 -*-
+# @Version: Python 3.9
+# @Time    : 2023/1/9 16:41
+# @Author  : chenyinhua
+# @File    : test_login_demo.py
+# @Software: PyCharm
+# @Desc: python脚本编写的测试用例文件
+
+
+import pytest
+import os
+from common_utils.yaml_handle import HandleYaml
+from config.settings import DATA_DIR
+from common_utils.base_request import BaseRequest
+from case_utils.assert_handle import assert_response, assert_sql
+from loguru import logger
+from case_utils.request_data_handle import RequestPreDataHandle, after_extract, case_data_replace, eval_data_process
+from pytest_html import extras  # 往pytest-html报告中填写额外的内容
+from common_utils.func_handle import add_docstring
+
+# 读取用例数据
+cases = HandleYaml(filename=os.path.join(DATA_DIR, "test_login_demo.yaml")).read_yaml
+
+
+@pytest.mark.test_login_demo
+@pytest.mark.parametrize("case", cases)
+def test_login_demo(case, extra, request):
+    logger.info("-----------------------------START-开始执行用例-----------------------------")
+    logger.debug(f"当前执行的用例数据:{case}")
+    try:
+        # 获取命令行参数，判断当前处于哪个环境
+        env = request.config.getoption("--env")
+        # 给当前测试方法添加文档注释
+        add_docstring(case.get("title", ""))(test_login)
+        if case.get("run", None):
+            # 处理请求前的用例数据
+            case_data = RequestPreDataHandle(case).request_data_handle()
+            # 将用例数据显示在pytest-html报告中
+            extra.append(extras.text(str(case_data), name="用例数据"))
+            # 发送请求
+            response = BaseRequest.send_request(case_data)
+            # 将响应数据显示在pytest-html报告中
+            extra.append(extras.text(str(response.text), name="响应数据"))
+            # 请求后，提取后置参数作为全局变量
+            after_extract(response, case_data["extract"])
+            # 从全局变量中获取最新值，替换数据库断言中的参数
+            case_data["assert_sql"] = eval_data_process(case_data_replace(case_data["assert_sql"]))
+            # 进行响应断言
+            assert_response(response, case_data["assert_response"])
+            # 进行数据库断言
+            assert_sql(env, case_data["assert_sql"])
+        else:
+            reason = f"标记了该用例为false，不执行\\n"
+            logger.warning(f"{reason}")
+            pytest.skip(reason)
+    except Exception as e:
+        logger.error(f"用例执行过程中报错：{e}")
+        raise e
+    finally:
+        logger.info("-----------------------------END-用例执行结束-----------------------------")