tidy

agent/graph.py

@@ -9,7 +9,6 @@ from agent.utils.state import AgentState
 from agent.utils.nodes import (
     call_model,
     tool_node,
-    read_memory,
     write_memory,
     should_continue
 )
@@ -22,7 +21,6 @@ def create_graph_builder():
     # Add nodes
     builder.add_node("agent", call_model)
     builder.add_node("action", tool_node)
-    builder.add_node("read_memory", read_memory)
     builder.add_node("write_memory", write_memory)
 
     # Set entry point
@@ -37,7 +35,6 @@ def create_graph_builder():
         should_continue,
         {
             "action": "action",
-            "read_memory": "read_memory",
             "write_memory": "write_memory",
             END: END
         }
@@ -46,9 +43,6 @@ def create_graph_builder():
     # Connect action back to agent
     builder.add_edge("action", "agent")
     
-    # Memory operations should end after completion
-    builder.add_edge("read_memory", "agent")
-    
     return builder
 
 def create_agent_graph_without_memory():

agent/utils/nodes.py

@@ -100,7 +100,7 @@ def update_memory(state: AgentState, config: RunnableConfig, store: BaseStore):
     store.put(namespace, "user_memory", {"memory": new_memory.content})
     return state
 
-def should_continue(state: AgentState) -> Literal["action", "read_memory", "write_memory", END]:
+def should_continue(state: AgentState) -> Literal["action", "write_memory", END]:
     """Determine the next node in the graph."""
     if not state["messages"]:
         return END
@@ -121,25 +121,9 @@ def should_continue(state: AgentState) -> Literal["action", "read_memory", "writ
         # Write memory for longer messages that might contain personal information
         if len(last_human_message.content.split()) > 3:
             return "write_memory"
-        # Read memory for short queries to ensure personalized responses
-        else:
-            return "read_memory"
             
     return END
 
-def read_memory(state: AgentState, config: RunnableConfig, store: BaseStore):
-    """Read and apply memory context without updating it."""
-    user_id = config["configurable"].get("session_id", "default")
-    namespace = ("memory", user_id)
-    existing_memory = store.get(namespace, "user_memory")
-    
-    if existing_memory:
-        memory_content = existing_memory.value.get('memory')
-        # Add memory context to state for next model call
-        state["memory_context"] = memory_content
-    
-    return state
-
 # Define the memory creation prompt
 MEMORY_CREATION_PROMPT = """"You are collecting information about the user to personalize your responses.
 
@@ -195,34 +179,6 @@ async def write_memory(state: AgentState, config: RunnableConfig, store: BaseSto
 # Initialize tool node
 tool_node = ToolNode(tool_belt)
 
-def should_call_memory(state: AgentState) -> Literal["update_memory", "end"]:
-    """
-    Determine if we should update memory based on the conversation state.
-    
-    Rules for updating memory:
-    1. Only update after human messages (not tool responses)
-    2. Update if the message might contain personal information
-    3. Don't update for simple queries or acknowledgments
-    """
-    if not state["messages"]:
-        return "end"
-        
-    last_message = state["messages"][-1]
-    
-    # Skip memory update for tool calls
-    if hasattr(last_message, "additional_kwargs") and last_message.additional_kwargs.get("tool_calls"):
-        return "agent"
-    
-    # Skip memory update for very short messages (likely acknowledgments)
-    if isinstance(last_message, HumanMessage) and len(last_message.content.split()) <= 3:
-        return "agent"
-    
-    # Update memory for human messages that might contain personal information
-    if isinstance(last_message, HumanMessage):
-        return "update_memory"
-    
-    return "agent"
-
 # def route_message(state: MessagesState, config: RunnableConfig, store: BaseStore) -> Literal[END, "update_todos", "update_instructions", "update_profile"]:
     
 #     """Reflect on the memories and chat history to decide whether to update the memory collection."""