diff --git a/rootfs/agents/Jarvis/agent.toml b/rootfs/agents/Jarvis/agent.toml
index 43fdf73..e6da761 100644
--- a/rootfs/agents/Jarvis/agent.toml
+++ b/rootfs/agents/Jarvis/agent.toml
@@ -1,5 +1,6 @@
 instance_id = "Jarvis"
 fullname = "Jarvis"
+llm_model_name = "gpt-3.5-turbo-16k-0613"
 
 [[prompt]]
 role = "system"
diff --git a/src/aios_kernel/agent.py b/src/aios_kernel/agent.py
index 3d15a7e..d673371 100644
--- a/src/aios_kernel/agent.py
+++ b/src/aios_kernel/agent.py
@@ -248,6 +248,14 @@ class AIAgent:
     async def _get_agent_prompt(self) -> AgentPrompt:
         return self.prompt
 
+    def _format_msg_by_env_value(self,prompt:AgentPrompt):
+        if self.owner_env is None:
+            return
+        
+        for msg in prompt.messages:
+            old_content = msg.get("content")
+            msg["content"] = old_content.format_map(self.owner_env)
+
     async def _process_msg(self,msg:AgentMsg) -> AgentMsg:
             from .compute_kernel import ComputeKernel
             from .bus import AIBus
@@ -269,6 +277,7 @@ class AIAgent:
             msg_prompt.messages = [{"role":"user","content":msg.body}]
             prompt.append(msg_prompt)
 
+            self._format_msg_by_env_value(prompt)
             inner_functions = self._get_inner_functions()
 
             task_result:ComputeTaskResult = await ComputeKernel.get_instance().do_llm_completion(prompt,self.llm_model_name,self.max_token_size,inner_functions)
diff --git a/src/aios_kernel/open_ai_node.py b/src/aios_kernel/open_ai_node.py
index c30fc87..857f2c7 100644
--- a/src/aios_kernel/open_ai_node.py
+++ b/src/aios_kernel/open_ai_node.py
@@ -167,8 +167,12 @@ class OpenAI_ComputeNode(ComputeNode):
 
     def is_support(self, task: ComputeTask) -> bool:
         if task.task_type == ComputeTaskType.LLM_COMPLETION: 
-            if not task.params["model_name"] or task.params["model_name"] == "gpt-4-0613":
+            if not task.params["model_name"]:
                 return True
+            model_name : str = task.params["model_name"]
+            if model_name.startswith("gpt-"):
+                return True
+            
         if task.task_type == ComputeTaskType.TEXT_EMBEDDING:
             if task.params["model_name"] == "text-embedding-ada-002":
                 return True
diff --git a/src/service/aios_shell/aios_shell.py b/src/service/aios_shell/aios_shell.py
index 6f2daff..13a5ee8 100644
--- a/src/service/aios_shell/aios_shell.py
+++ b/src/service/aios_shell/aios_shell.py
@@ -409,7 +409,7 @@ async def main():
 
     await asyncio.sleep(0.2) 
     while True:
-        user_input = await session.prompt_async(f"{shell.username}<->{shell.current_topic}@{shell.current_target}$",completer=completer,style=shell_style)
+        user_input = await session.prompt_async(f"{shell.username}<->{shell.current_topic}@{shell.current_target}$ ",completer=completer,style=shell_style)
         if len(user_input) <= 1:
             continue