diff --git a/desktop_env/controllers/setup.py b/desktop_env/controllers/setup.py
index f62a303..659db62 100644
--- a/desktop_env/controllers/setup.py
+++ b/desktop_env/controllers/setup.py
@@ -20,6 +20,7 @@ class SetupController:
     def __init__(self, vm_ip: str, cache_dir: str):
         self.vm_ip: str = vm_ip
         self.http_server: str = f"http://{vm_ip}:5000"
+        self.http_server_setup_root: str = f"http://{vm_ip}:5000/setup"
         self.cache_dir: str = cache_dir
 
     def reset_cache_dir(self, cache_dir: str):
diff --git a/desktop_env/evaluators/getters/general.py b/desktop_env/evaluators/getters/general.py
index 22a20c7..1c1ce4c 100644
--- a/desktop_env/evaluators/getters/general.py
+++ b/desktop_env/evaluators/getters/general.py
@@ -1,7 +1,5 @@
-from typing import Dict
-
 import os
-import requests
+from typing import Dict
 
 
 def get_string(env, config: Dict[str, str]) -> str:
@@ -12,12 +10,13 @@ def get_string(env, config: Dict[str, str]) -> str:
 
     return config["string"]
 
+
 def get_command_line(env, config: Dict[str, str]) -> str:
     """
     Config:
         string (str)
     """
-    
+
     f = os.popen(config["command"])
-    
-    return f.read()
\ No newline at end of file
+
+    return f.read()
diff --git a/desktop_env/evaluators/getters/misc.py b/desktop_env/evaluators/getters/misc.py
index f4c7bf2..b6b933a 100644
--- a/desktop_env/evaluators/getters/misc.py
+++ b/desktop_env/evaluators/getters/misc.py
@@ -1,6 +1,5 @@
 import logging
 from typing import TypeVar
-#from typing import Dict, List
 
 logger = logging.getLogger("desktopenv.getters.misc")
 
@@ -13,6 +12,7 @@ def get_rule(env, config: R) -> R:
     """
     return config["rules"]
 
+
 def get_accessibility_tree(env, *args) -> str:
     accessibility_tree: str = env.controller.get_accessibility_tree()
     logger.debug("AT@eval: %s", accessibility_tree)
diff --git a/desktop_env/evaluators/getters/replay.py b/desktop_env/evaluators/getters/replay.py
index a0d1d62..67ad0d3 100644
--- a/desktop_env/evaluators/getters/replay.py
+++ b/desktop_env/evaluators/getters/replay.py
@@ -1,7 +1,7 @@
 from typing import List, Dict, Any
 
-def get_replay(env, trajectory: List[Dict[str, Any]]) -> None:
 
+def get_replay(env, trajectory: List[Dict[str, Any]]) -> None:
     def parse(action):
         if action["type"] == "hotkey":
             keys = "', '".join(action["param"])
@@ -16,4 +16,4 @@ def get_replay(env, trajectory: List[Dict[str, Any]]) -> None:
             return f"pyautogui.press('{key}')"
 
     for action in trajectory:
-        env.controller.execute_python_command(parse(action))
\ No newline at end of file
+        env.controller.execute_python_command(parse(action))
diff --git a/desktop_env/evaluators/getters/vscode.py b/desktop_env/evaluators/getters/vscode.py
index 00d280a..5072ec6 100644
--- a/desktop_env/evaluators/getters/vscode.py
+++ b/desktop_env/evaluators/getters/vscode.py
@@ -1,17 +1,18 @@
-from typing import Dict
 from typing import Any
-from replay import get_replay
-from file import get_vm_file
+from typing import Dict
+
+from .file import get_vm_file
+from .replay import get_replay
+
 
 def get_vscode_config(env, config: Dict[str, Any]) -> str:
-
     trajectory = [{"type": "hotkey", "param": ["command", "shift", "p"]},
-        {"type": "typewrite", "param": "OpenProject"},
-        {"type": "press", "param": "enter"}]
+                  {"type": "typewrite", "param": "OpenProject"},
+                  {"type": "press", "param": "enter"}]
 
     get_replay(env, trajectory)
 
     return get_vm_file(env, {
-            "path": config["path"],
-            "dest": config["dest"]
-        })
\ No newline at end of file
+        "path": config["path"],
+        "dest": config["dest"]
+    })
diff --git a/desktop_env/evaluators/metrics/__init__.py b/desktop_env/evaluators/metrics/__init__.py
index 44c55e9..67b8925 100644
--- a/desktop_env/evaluators/metrics/__init__.py
+++ b/desktop_env/evaluators/metrics/__init__.py
@@ -13,4 +13,4 @@ from .vlc import is_vlc_playing, is_vlc_recordings_folder, is_vlc_fullscreen, co
 from .gimp import increase_saturation, decrease_brightness, check_file_exists, compare_triangle_positions
 from .general import check_csv, check_accessibility_tree, check_list, run_sqlite3
 from .thunderbird import check_thunderbird_prefs, check_thunderbird_filter
-
+from .vscode import compare_text_file, compare_config, compare_answer
diff --git a/desktop_env/evaluators/metrics/docs.py b/desktop_env/evaluators/metrics/docs.py
index 48d2c03..9f2d940 100644
--- a/desktop_env/evaluators/metrics/docs.py
+++ b/desktop_env/evaluators/metrics/docs.py
@@ -1,12 +1,14 @@
-import xml.etree.ElementTree as ET
+import logging
 import os
+import xml.etree.ElementTree as ET
 from typing import List, Dict, Any
+
 from docx import Document
 from docx.enum.text import WD_PARAGRAPH_ALIGNMENT
 
-import logging
 logger = logging.getLogger("desktopenv.metric.docs")
 
+
 def find_default_font(config_file_path, rules):
     """Find the default font in LibreOffice Writer."""
     default_font = None
diff --git a/desktop_env/evaluators/metrics/libreoffice.py b/desktop_env/evaluators/metrics/libreoffice.py
index 4ca07de..441d932 100644
--- a/desktop_env/evaluators/metrics/libreoffice.py
+++ b/desktop_env/evaluators/metrics/libreoffice.py
@@ -1,37 +1,38 @@
-import lxml.cssselect
-from lxml.etree import _Element as Element
-import lxml.etree
 import fnmatch
-
 from typing import Dict, List
 
+import lxml.cssselect
+import lxml.etree
+from lxml.etree import _Element as Element
+
 _libconf_namespaces = [("oor", "http://openoffice.org/2001/registry")]
 _libconf_ns_mapping = dict(_libconf_namespaces)
-_setup_locale_selector = lxml.cssselect.CSSSelector( 'item[oor|path$=L10N]>prop[oor|name=ooSetupSystemLocale]>value'
-                                                   , namespaces=_libconf_ns_mapping
-                                                   )
-_locale_selector = lxml.cssselect.CSSSelector( 'item[oor|path$=L10N]>prop[oor|name=ooLocale]>value'
-                                             , namespaces=_libconf_ns_mapping
-                                             )
+_setup_locale_selector = lxml.cssselect.CSSSelector('item[oor|path$=L10N]>prop[oor|name=ooSetupSystemLocale]>value',
+                                                    namespaces=_libconf_ns_mapping)
+_locale_selector = lxml.cssselect.CSSSelector('item[oor|path$=L10N]>prop[oor|name=ooLocale]>value',
+                                              namespaces=_libconf_ns_mapping)
+
+
 def check_libre_locale(config_file: str, rules: Dict[str, List[str]]) -> float:
     config: Element = lxml.etree.parse(config_file).getroot()
     setup_locale_setting: List[Element] = _setup_locale_selector(config)
     locale_setting: List[Element] = _locale_selector(config)
 
-    setup_locale_setting: str = setup_locale_setting[0].text\
-                             if len(setup_locale_setting)>0\
-                           else locale_setting[0].text
+    setup_locale_setting: str = setup_locale_setting[0].text \
+        if len(setup_locale_setting) > 0 \
+        else locale_setting[0].text
 
-    return float( any( fnmatch.fnmatchcase(setup_locale_setting, ptn)\
-                   for ptn in rules["locale_set"]
+    return float(any(fnmatch.fnmatchcase(setup_locale_setting, ptn) \
+                     for ptn in rules["locale_set"]
                      )
-                )
+                 )
+
 
 if __name__ == "__main__":
     path1 = "../../任务数据/LibreOffice Calc/registrymodifications.ru.xcu"
-    print( check_libre_locale( path1, { "locale_set": [ "ru-*", "de-*", "fr-*"
-		     		                                  , "pt-*", "es-*", "it-*"
-                                                      ]
-                                      }
+    print(check_libre_locale(path1, {"locale_set": ["ru-*", "de-*", "fr-*"
+        , "pt-*", "es-*", "it-*"
+                                                    ]
+                                     }
                              )
-         )
+          )
diff --git a/desktop_env/evaluators/metrics/pdf.py b/desktop_env/evaluators/metrics/pdf.py
index 51c79f3..d607733 100644
--- a/desktop_env/evaluators/metrics/pdf.py
+++ b/desktop_env/evaluators/metrics/pdf.py
@@ -1,13 +1,11 @@
-from pypdf import PdfReader
 import operator
-
-from typing import Dict
 from typing import Any
+from typing import Dict
+
+from pypdf import PdfReader
+
 
 def check_pdf_pages(pdf_file: str, rules: Dict[str, Any]) -> float:
     reader = PdfReader(pdf_file)
     nb_pages: int = len(reader.pages)
-    return float( getattr(operator, rules["relation"])( nb_pages
-                                                      , rules["ref_value"]
-                                                      )
-                )
+    return float(getattr(operator, rules["relation"])(nb_pages, rules["ref_value"]))
diff --git a/desktop_env/evaluators/metrics/table.py b/desktop_env/evaluators/metrics/table.py
index 25b55f3..03daef5 100644
--- a/desktop_env/evaluators/metrics/table.py
+++ b/desktop_env/evaluators/metrics/table.py
@@ -1,18 +1,19 @@
-import pandas as pd
+import logging
+import operator
+from numbers import Number
+from typing import Any, Union
+from typing import Dict, List
+
 import openpyxl
+import pandas as pd
 from openpyxl import Workbook
 from openpyxl.worksheet.worksheet import Worksheet
 
 from .utils import load_charts, load_sparklines
-import operator
 
-from typing import Dict, List
-from typing import Any, Union
-from numbers import Number
-
-import logging
 logger = logging.getLogger("desktopenv.metric.table")
 
+
 def compare_table(actual: str, expected: str, **options) -> float:
     """
     Args:
@@ -44,28 +45,28 @@ def compare_table(actual: str, expected: str, **options) -> float:
         workbook1: Workbook = openpyxl.load_workbook(actual)
         workbook2: Workbook = openpyxl.load_workbook(expected)
 
-        if ftr=="sparkline":
+        if ftr == "sparkline":
             sp1 = load_sparklines(actual)
             sp2 = load_sparklines(expected)
             new_metric: bool = sp1 == sp2
             logger.debug("Sparkline Metric: {:}".format(new_metric))
-        elif ftr=="chart":
+        elif ftr == "chart":
             charts1 = load_charts(workbook1, **options)
             charts2 = load_charts(workbook2, **options)
             new_metric: bool = charts1 == charts2
             logger.debug("Chart Metric: {:}".format(new_metric))
-        elif ftr=="number_format":
-            number_formats1: List[str] = [ c.number_format.lower()\
-                                           for col in workbook1.active.iter_cols()\
-                                            for c in col\
-                                            if c.data_type=="n"
-                                         ]
-            number_formats2: List[str] = [ c.number_format.lower()\
-                                           for col in workbook2.active.iter_cols()\
-                                            for c in col\
-                                            if c.data_type=="n"
-                                         ]
-            new_metric: bool = number_formats1==number_formats2
+        elif ftr == "number_format":
+            number_formats1: List[str] = [c.number_format.lower() \
+                                          for col in workbook1.active.iter_cols() \
+                                          for c in col \
+                                          if c.data_type == "n"
+                                          ]
+            number_formats2: List[str] = [c.number_format.lower() \
+                                          for col in workbook2.active.iter_cols() \
+                                          for c in col \
+                                          if c.data_type == "n"
+                                          ]
+            new_metric: bool = number_formats1 == number_formats2
             logger.debug("Number Format Metric: {:}".format(new_metric))
         else:
             raise NotImplementedError("Unsupported xlsx feature: {:}".format(ftr))
@@ -73,6 +74,7 @@ def compare_table(actual: str, expected: str, **options) -> float:
 
     return float(metric)
 
+
 def check_sheet_list(result: str, rules: List[Dict[str, Any]]) -> float:
     if result is None:
         return 0.
@@ -114,6 +116,7 @@ def check_sheet_list(result: str, rules: List[Dict[str, Any]]) -> float:
 
     return float(passes)
 
+
 def check_xlsx_freeze(result: str, rules: Dict[str, str]) -> float:
     if result is None:
         return 0.
@@ -121,16 +124,18 @@ def check_xlsx_freeze(result: str, rules: Dict[str, str]) -> float:
     worksheet: Worksheet = openpyxl.load_workbook(filename=result).active
     return float(worksheet.freeze_panes == rules["position"])
 
+
 def check_xlsx_zoom(result: str, rules: Dict[str, Union[str, Number]]) -> float:
     if result is None:
         return 0.
 
     worksheet = openpyxl.load_workbook(filename=result).active
     zoom_scale: Number = worksheet.sheet_view.zoomScale or 100.
-    return float( getattr(operator, rules["relation"])( zoom_scale
+    return float(getattr(operator, rules["relation"])(zoom_scale
                                                       , rules["ref_value"]
                                                       )
-                )
+                 )
+
 
 if __name__ == '__main__':
     # path1 = ""
@@ -168,51 +173,51 @@ if __name__ == '__main__':
     # ]
     # print(check_sheet_list(path1, rule))
 
-    #path1 = "../../任务数据/LibreOffice Calc/Create_column_charts_using_statistics_gold.xlsx"
-    #path2 = "../../任务数据/LibreOffice Calc/Create_column_charts_using_statistics_gold2.xlsx"
-    #print(compare_table(path1, path2, features=["chart"], chart_props=["type", "direction"]))
+    # path1 = "../../任务数据/LibreOffice Calc/Create_column_charts_using_statistics_gold.xlsx"
+    # path2 = "../../任务数据/LibreOffice Calc/Create_column_charts_using_statistics_gold2.xlsx"
+    # print(compare_table(path1, path2, features=["chart"], chart_props=["type", "direction"]))
 
-    #path1 = "../../任务数据/LibreOffice Calc/Represent_in_millions_billions_gold.xlsx"
-    #path2 = "../../任务数据/LibreOffice Calc/Represent_in_millions_billions_gold3.xlsx"
-    #path1 = "../../任务数据/LibreOffice Calc/Set_Decimal_Separator_Dot.xlsx"
-    #path2 = "../../任务数据/LibreOffice Calc/Set_Decimal_Separator_Dot_gold.xlsx"
-    #workbook1: Workbook = openpyxl.load_workbook(filename=path1)
-    #worksheet1: Worksheet = workbook1.active
-    #import itertools
-    #for col, r in itertools.product( ['A', 'B']
-                                   #, range(1, 20)
-                                   #):
-        #position: str = "{:}{:d}".format(col, r)
-        #print(worksheet1[position])
-        #print(worksheet1[position].value)
-        #print(worksheet1[position].number_format)
-    #workbook2: Workbook = openpyxl.load_workbook(filename=path2)
-    #worksheet2: Worksheet = workbook2.active
-    #for col, r in itertools.product( ['A', 'B']
-                                   #, range(1, 20)
-                                   #):
-        #position: str = "{:}{:d}".format(col, r)
-        #print(worksheet2[position])
-        #print(worksheet2[position].value)
-        #print(worksheet2[position].number_format)
-    #print(compare_table(path1, path2, features=["number_format"]))
+    # path1 = "../../任务数据/LibreOffice Calc/Represent_in_millions_billions_gold.xlsx"
+    # path2 = "../../任务数据/LibreOffice Calc/Represent_in_millions_billions_gold3.xlsx"
+    # path1 = "../../任务数据/LibreOffice Calc/Set_Decimal_Separator_Dot.xlsx"
+    # path2 = "../../任务数据/LibreOffice Calc/Set_Decimal_Separator_Dot_gold.xlsx"
+    # workbook1: Workbook = openpyxl.load_workbook(filename=path1)
+    # worksheet1: Worksheet = workbook1.active
+    # import itertools
+    # for col, r in itertools.product( ['A', 'B']
+    # , range(1, 20)
+    # ):
+    # position: str = "{:}{:d}".format(col, r)
+    # print(worksheet1[position])
+    # print(worksheet1[position].value)
+    # print(worksheet1[position].number_format)
+    # workbook2: Workbook = openpyxl.load_workbook(filename=path2)
+    # worksheet2: Worksheet = workbook2.active
+    # for col, r in itertools.product( ['A', 'B']
+    # , range(1, 20)
+    # ):
+    # position: str = "{:}{:d}".format(col, r)
+    # print(worksheet2[position])
+    # print(worksheet2[position].value)
+    # print(worksheet2[position].number_format)
+    # print(compare_table(path1, path2, features=["number_format"]))
 
-    #path1 = "../../任务数据/LibreOffice Calc/Zoom_Out_Oversized_Cells_gold.xlsx"
-    #path2 = "../../任务数据/LibreOffice Calc/Zoom_Out_Oversized_Cells.xlsx"
-    #workbook1: Workbook = openpyxl.load_workbook(filename=path1)
-    #worksheet1: Worksheet = workbook1.active
-    #print(worksheet1.sheet_view.zoomScale)
-    #print(type(worksheet1.sheet_view.zoomScale))
-#
-    #import os
-    #import os.path
-    #for wb in filter( lambda f: f.endswith(".xlsx")
-                    #, os.listdir("../../任务数据/LibreOffice Calc/")
-                    #):
-        #path = os.path.join("../../任务数据/LibreOffice Calc/", wb)
-        #print(wb, openpyxl.load_workbook(filename=path).active.sheet_view.zoomScale)
-    #print(check_zoom(path1, {"relation": "lt", "ref_value": 100}))
-    #print(check_zoom(path2, {"relation": "lt", "ref_value": 100}))
+    # path1 = "../../任务数据/LibreOffice Calc/Zoom_Out_Oversized_Cells_gold.xlsx"
+    # path2 = "../../任务数据/LibreOffice Calc/Zoom_Out_Oversized_Cells.xlsx"
+    # workbook1: Workbook = openpyxl.load_workbook(filename=path1)
+    # worksheet1: Worksheet = workbook1.active
+    # print(worksheet1.sheet_view.zoomScale)
+    # print(type(worksheet1.sheet_view.zoomScale))
+    #
+    # import os
+    # import os.path
+    # for wb in filter( lambda f: f.endswith(".xlsx")
+    # , os.listdir("../../任务数据/LibreOffice Calc/")
+    # ):
+    # path = os.path.join("../../任务数据/LibreOffice Calc/", wb)
+    # print(wb, openpyxl.load_workbook(filename=path).active.sheet_view.zoomScale)
+    # print(check_zoom(path1, {"relation": "lt", "ref_value": 100}))
+    # print(check_zoom(path2, {"relation": "lt", "ref_value": 100}))
 
     path1 = "../../任务数据/LibreOffice Calc/Padding_Decimals_In_Formular_gold.xlsx"
     data_frame: pd.DataFrame = pd.read_excel(path1)
diff --git a/desktop_env/evaluators/metrics/vscode.py b/desktop_env/evaluators/metrics/vscode.py
index a76910b..053bc75 100644
--- a/desktop_env/evaluators/metrics/vscode.py
+++ b/desktop_env/evaluators/metrics/vscode.py
@@ -1,3 +1,6 @@
+from typing import Dict
+
+
 def compare_text_file(actual: str, expected: str, **options) -> float:
     """
     Args:
@@ -7,7 +10,7 @@ def compare_text_file(actual: str, expected: str, **options) -> float:
     Return:
         float: the score
     """
-    
+
     with open(actual) as f1:
         actual_text = f1.read()
     with open(expected) as f2:
@@ -17,24 +20,17 @@ def compare_text_file(actual: str, expected: str, **options) -> float:
         return 1.0
     return 0.0
 
-def compare_config(actual: str, expected: str, **options) -> float:
-    """
-    Args:
-        actual (str): path to result text file
-        expected (str): gold string
 
-    Return:
-        float: the score
-    """
-    
+def compare_config(actual: str, rules: Dict, **options) -> float:
     with open(actual) as f1:
         actual_text = f1.read()
 
-    if actual_text == expected:
+    if actual_text == rules['expect']:
         return 1.0
     return 0.0
 
-def compare_answer(actual: str, expected: str, **options) -> float:
+
+def compare_answer(actual: str, rules: Dict, **options) -> float:
     """
     Args:
         actual (str): result string
@@ -44,11 +40,8 @@ def compare_answer(actual: str, expected: str, **options) -> float:
         float: the score
     """
 
-    if actual == expected:
+    if actual == rules['expect']:
         return 1.0
-    
+
     # TODO: can use text embedding to get non-zero return
     return 0.0
-
-if __name__ == '__main__':
-    print(compare_text_file("README.md", "README.md"))
\ No newline at end of file
diff --git a/evaluation_examples/examples/vs_code/0ed39f63-6049-43d4-ba4d-5fa2fe04a951.json b/evaluation_examples/examples/vs_code/0ed39f63-6049-43d4-ba4d-5fa2fe04a951.json
index c61b8bd..cbbfd16 100644
--- a/evaluation_examples/examples/vs_code/0ed39f63-6049-43d4-ba4d-5fa2fe04a951.json
+++ b/evaluation_examples/examples/vs_code/0ed39f63-6049-43d4-ba4d-5fa2fe04a951.json
@@ -16,9 +16,9 @@
       }
     },
     {
-      "type": "open",
+      "type": "launch",
       "parameters": {
-        "path": "Desktop/vscode_replace_text.txt"
+        "command": ["code", "Desktop/vscode_replace_text.txt"]
       }
     }
   ],
diff --git a/evaluation_examples/examples/vs_code/53ad5833-3455-407b-bbc6-45b4c79ab8fb.json b/evaluation_examples/examples/vs_code/53ad5833-3455-407b-bbc6-45b4c79ab8fb.json
index 030a2ac..fb0d892 100644
--- a/evaluation_examples/examples/vs_code/53ad5833-3455-407b-bbc6-45b4c79ab8fb.json
+++ b/evaluation_examples/examples/vs_code/53ad5833-3455-407b-bbc6-45b4c79ab8fb.json
@@ -4,6 +4,12 @@
   "instruction": "Could you help me open the project at /home/user/project?",
   "source": "https://www.youtube.com/watch?v=VqCgcpAypFQ",
   "config": [
+    {
+      "type": "launch",
+      "parameters": {
+        "command": ["code"]
+      }
+    },
     {
       "type": "command",
       "parameters": {
@@ -18,8 +24,10 @@
   "evaluator": {
     "func": "compare_config",
     "expected": {
-      "type": "string",
-      "string": "project"
+      "type": "rule",
+      "rules": {
+        "expect": "project"
+      }
     },
     "result": {
       "type": "vscode_config",
diff --git a/evaluation_examples/examples/vs_code/59ed65c7-e9a6-43db-833f-76d6730c0004.json b/evaluation_examples/examples/vs_code/59ed65c7-e9a6-43db-833f-76d6730c0004.json
index 7975f75..b3201d8 100644
--- a/evaluation_examples/examples/vs_code/59ed65c7-e9a6-43db-833f-76d6730c0004.json
+++ b/evaluation_examples/examples/vs_code/59ed65c7-e9a6-43db-833f-76d6730c0004.json
@@ -29,8 +29,10 @@
   "evaluator": {
     "func": "compare_config",
     "expected": {
-      "type": "string",
-      "string": "100"
+      "type": "rule",
+      "rules": {
+        "expect": "100"
+      }
     },
     "result": {
       "type": "vscode_config",