diff --git a/flo_ai/flo_ai/agent/agent.py b/flo_ai/flo_ai/agent/agent.py
index d23602b6..04da762b 100644
--- a/flo_ai/flo_ai/agent/agent.py
+++ b/flo_ai/flo_ai/agent/agent.py
@@ -291,16 +291,6 @@ async def _run_with_tools(
                                 'content': assistant_message_content,
                             }
                         )
-                    else:
-                        # Use default text content extraction
-                        assistant_text = self.llm.get_message_content(response)
-                        if assistant_text:
-                            messages.append(
-                                {
-                                    'role': self.act_as,
-                                    'content': assistant_text,
-                                }
-                            )
 
                     # Execute the tool
                     try:
diff --git a/flo_ai/flo_ai/models/agent.py b/flo_ai/flo_ai/models/agent.py
index 184f20ff..d0f8bd66 100644
--- a/flo_ai/flo_ai/models/agent.py
+++ b/flo_ai/flo_ai/models/agent.py
@@ -8,6 +8,8 @@
 from typing import List, Optional, Dict, Any, Union, Literal
 from pydantic import BaseModel, Field, field_validator
 
+from flo_ai.models import MessageType
+
 
 class MetadataModel(BaseModel):
     """Metadata section for agent YAML configuration."""
@@ -187,7 +189,9 @@ class AgentConfigModel(BaseModel):
     job: Optional[str] = Field(None, description='System prompt/job description')
     prompt: Optional[str] = Field(None, description='System prompt (alias for job)')
     role: Optional[str] = Field(None, description='Agent role')
-    act_as: Optional[str] = Field(None, description='Agent act_as setting')
+    act_as: Optional[str] = Field(
+        MessageType.ASSISTANT, description='Agent act_as setting'
+    )
     model: Optional[LLMConfigModel] = Field(None, description='LLM model configuration')
     base_url: Optional[str] = Field(
         None, description='Base URL (can be at agent or model level)'