diff --git a/agenthub/codeact_agent/action_parser.py b/agenthub/codeact_agent/action_parser.py
index c0d39451abc..be82517be0d 100644
--- a/agenthub/codeact_agent/action_parser.py
+++ b/agenthub/codeact_agent/action_parser.py
@@ -82,7 +82,7 @@ def parse(self, action_str: str) -> Action:
         ), 'self.finish_command should not be None when parse is called'
         thought = action_str.replace(self.finish_command.group(0), '').strip()
         if not self.is_finish2 and os.getenv('SWE_BENCH') == '1':
-            return CmdRunAction('pytest /testbed/.test_task.py', thought='')
+            return CmdRunAction('pytest /tmp/.test_task.py', thought='')
         return AgentFinishAction(thought=thought)
 
 
diff --git a/evaluation/swe_bench/run_infer.py b/evaluation/swe_bench/run_infer.py
index fdfe7b09b0c..716b99a96f9 100644
--- a/evaluation/swe_bench/run_infer.py
+++ b/evaluation/swe_bench/run_infer.py
@@ -277,7 +277,7 @@ def initialize_runtime(
 {test_code}
 print('End the task with <finish2></finish2>')
 """
-create_file('/testbed/.test_task.py', FILE_CONTENT)
+create_file('/tmp/.test_task.py', FILE_CONTENT)
 '''
         action = IPythonRunCellAction(test_code)
         logger.info(action, extra={'msg_type': 'ACTION'})