LangGraph Studio (mac) -> AttributeError: module 'mvgbxnshzyfjxntsvqcqbkah' has no attribute 'graph'

[acsankar]

Checked other resources

• I added a very descriptive title to this issue.
• I searched the LangGraph/LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.
• I am sure that this is a bug in LangGraph/LangChain rather than my code.
• I am sure this is better as an issue rather than a GitHub discussion, since this is a LangGraph bug and not a design question.

Example Code

def create_workflow() -> StateGraph:
    """Create the workflow graph with conditional routing"""
    
    # Initialize workflow with State type
    workflow = StateGraph(State)
    
    # Define routing function
    def router(state: State) -> str:
        """Route based on state's next value"""
        return state.get('next', 'triage')
    
    # Add nodes
    workflow.add_node("triage", triage_agent)
    workflow.add_node("data_agent", data_agent)
    workflow.add_node("access_agent", access_agent)
    workflow.add_node("human_in_loop", human_in_loop)
    
    # Set entry point
    workflow.set_entry_point("triage")
    
    # Add conditional edges from triage
    workflow.add_conditional_edges(
        "triage",
        router,
        {
            "data_agent": "data_agent",
            "access_agent": "access_agent",
            "human_in_loop": "human_in_loop"
        }
    )
    
    # Add conditional edges from human_in_loop
    workflow.add_conditional_edges(
        "human_in_loop",
        router,
        {
            "data_agent": "data_agent",
            "access_agent": "access_agent",
            "end": END
        }
    )
    
    # Add edges to END from agents
    workflow.add_conditional_edges(
        "data_agent",
        lambda x: "end",
        {"end": END}
    )
    
    workflow.add_conditional_edges(
        "access_agent",
        lambda x: "end",
        {"end": END}
    )
    
    return workflow.compile()

Error Message and Stack Trace (if applicable)

Failed to start project lang4
error | Traceback (most recent call last):
  File "/usr/local/lib/python3.11/site-packages/starlette/routing.py", line 693, in lifespan
    async with self.lifespan_context(app) as maybe_state:
  File "/usr/local/lib/python3.11/contextlib.py", line 210, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/api/langgraph_api/lifespan.py", line 30, in lifespan
  File "/api/langgraph_api/shared/graph.py", line 230, in collect_graphs_from_env
  File "/usr/local/lib/python3.11/site-packages/langchain_core/runnables/config.py", line 588, in run_in_executor
    return await asyncio.get_running_loop().run_in_executor(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/concurrent/futures/thread.py", line 58, in run
    result = self.fn(*self.args, **self.kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/langchain_core/runnables/config.py", line 579, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/api/langgraph_api/shared/graph.py", line 276, in _graph_from_spec
AttributeError: module 'mvgbxnshzyfjxntsvqcqbkah' has no attribute 'graph'

Description

I am trying to load the folder to Langstudio (mac installation), i am getting the below issue:

File "/usr/local/lib/python3.11/site-packages/langchain_core/runnables/config.py", line 579, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/api/langgraph_api/shared/graph.py", line 276, in _graph_from_spec
AttributeError: module 'mvgbxnshzyfjxntsvqcqbkah' has no attribute 'graph'

System Info

]

[vbarda]

@acsankar could you share your langgraph.json? it sounds like you don't have a variable called graph with the compiled graph in the .py file you're trying to load the graph from

[acsankar]

@acsankar could you share your langgraph.json? it sounds like you don't have a variable called graph with the compiled graph in the .py file you're trying to load the graph from

{
"dockerfile_lines": [],
"graphs": {
"agent": "./main.py:graph"
},
"env": [
"HK"
],
"python_version": "3.11",
"dependencies": [
"."
]
}

[acsankar]

below is the structure of the code setup. i am able to make it work in Streamlit but not able to loan in Langgraph studio due to the issue mentioned

[vbarda]

and just to confirm -- you have a variable like graph = create_workflow() in main.py, right?

[acsankar]

Below is the code

def create_workflow() -> Graph:
"""Create the workflow graph"""
workflow = Graph()

# Add nodes
workflow.add_node("triage", triage_agent)
workflow.add_node("data_agent", data_agent)
workflow.add_node("access_agent", access_agent)
workflow.add_node("human_in_loop", human_in_loop)

# Add edges
workflow.set_entry_point("triage")
workflow.add_edge('triage', 'data_agent')
workflow.add_edge('triage', 'access_agent')
workflow.add_edge('triage', 'human_in_loop')
workflow.add_edge('human_in_loop', 'data_agent')
workflow.add_edge('human_in_loop', 'access_agent')

return workflow.compile()

[vbarda]

Below is the code

def create_workflow() -> Graph: """Create the workflow graph""" workflow = Graph()

# Add nodes
workflow.add_node("triage", triage_agent)
workflow.add_node("data_agent", data_agent)
workflow.add_node("access_agent", access_agent)
workflow.add_node("human_in_loop", human_in_loop)

# Add edges
workflow.set_entry_point("triage")
workflow.add_edge('triage', 'data_agent')
workflow.add_edge('triage', 'access_agent')
workflow.add_edge('triage', 'human_in_loop')
workflow.add_edge('human_in_loop', 'data_agent')
workflow.add_edge('human_in_loop', 'access_agent')

return workflow.compile()

right, but do you save the output of create_workflow into a variable? if yes, then it should be named graph for your config to work. if no - you can also provide function names instead of variables like this:

"agent": "./main.py:create_workflow"

hope this helps