Create_tool_calling_agent is not supporting short term memory

OSS Product Help LangGraph
,
1

When I tried to add a short term memory to the tool_calling_agent, im getting the below error:

ValidationError: 1 validation error for get_user_name
state
  Field required [type=missing, input_value={}, input_type=dict]
    For further information visit https://errors.pydantic.dev/2.11/v/missing

Here’s my code:

class CustomState(AgentState):
    # The user_name field in short-term state
    user_name: NotRequired[str]

@tool
def get_user_name(
    state: Annotated[CustomState, InjectedState]
) -> str:
    """Retrieve the current user-name from state."""
    # Return stored name or a default if not set
    return state.get("user_name", "Arjun")



prompt = ChatPromptTemplate.from_messages(
    [
        ("system", "You are a helpful assistant"),
        ("placeholder", "{chat_history}"),
        ("human", "{input}"),
        ("placeholder", "{agent_scratchpad}"),
    ])
tools=[get_user_name]
agent = create_tool_calling_agent(llm, tools, prompt)
agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)
response=agent_executor.invoke({"input": "what's my name?"})

It seems like the state is not getting injected properly..

When I tried to use the normal reAct agent, its working fine as expected


class CustomState(AgentState):
    # The user_name field in short-term state
    user_name: NotRequired[str]

@tool
def get_user_name(
    state: Annotated[CustomState, InjectedState]
) -> str:
    """Retrieve the current user-name from state."""
    # Return stored name or a default if not set
    return state.get("user_name", "Arjun")

# Example agent setup
agent = create_react_agent(
    model=llm,
    tools=[get_user_name],
    # state_schema=CustomState,
)

# Invocation: reads the name from state (initially empty)
agent.invoke({"messages": "what's my name?"})


response==>====

{'messages': [HumanMessage(content="what's my name?", additional_kwargs={}, response_metadata={}, id='81e9293e-f3be-4c91-aadb-ba55baaf57c9'),
  AIMessage(content='', additional_kwargs={'tool_calls': [{'id': 'chatcmpl-tool-e4
  AIMessage(content='Your name is Arjun.', additional_kwargs={}, 
==========================================================

Can someone please tell me why create_tool_calling_agent is not honouring this?

2

Hi @injusticescorpio

what is gonna happen when you change state: Annotated[CustomState, InjectedState] to state: Annotated[dict, InjectedState]? Could you check it please?

3

@pawel-twardziak
I have tried this already but getting the same error.:smiling_face_with_tear:

4

ok :slight_smile: I will try to investigate further

5

Maybe that will be useful LangGraph CLI treats InjectedState as required tool parameter and fails the validation