Missing conditional edge when displaying graph

[fabio-barbieri]

Hi, I was trying to build a simple workflow in langgraph to gain familiarity with the framework. My problem basically is that I'm adding two conditional edges in my graph, but when displaying it, it presents only a single conditional edge, while the other is missing.
I'm working on Google Colaboratory at the moment, using version 0.3.34 of langgraph. I'm leaving the code down below:

import os
import getpass
from langchain_google_genai import ChatGoogleGenerativeAI
from langchain_core.messages import HumanMessage
from langgraph.graph import StateGraph, START, END
from langgraph.graph import MessagesState
from IPython.display import Image, display
from langgraph.types import interrupt

lite_llm = ChatGoogleGenerativeAI(
    model="gemini-1.5-flash",
    temperature=0.1
)

class State(MessagesState):
    risk_profile_accepted: bool
    
def user_profiling_agent(state: State) -> State:
    messages = [
        ("system", USER_PROFILER_SYSTEM_PROMPT),
        ("human", state["messages"])
    ]
    response = lite_llm.invoke(messages)
    return state

def user_risk_level_acceptance(state: State) -> State:
    ai_message = state["messages"][-1]["content"]
    user_response = interrupt({
        "prompt": f"{ai_message}. Do you want to proceed? (y/n)"
    })["prompt"]
    state["approved"] = user_response.strip().lower() == "y"
    return state

def user_risk_acceptance_check(state: State):
    if state["risk_profile_accepted"] == True:
        return END
    return "user_risk_level_acceptance"

builder = StateGraph(State)

builder.add_node("user_profiling_agent", user_profiling_agent)
builder.add_node("user_risk_level_acceptance", user_risk_level_acceptance)

builder.add_edge(START, "user_profiling_agent")
builder.add_conditional_edges(
    "user_profiling_agent",
    user_risk_acceptance_check,
    {END: END, "user_risk_level_acceptance": "user_risk_level_acceptance"}
)
builder.add_edge("user_risk_level_acceptance", END)

graph = builder.compile()

display(Image(graph.get_graph().draw_mermaid_png()))

Here's the graph that I get:

Trying to understand the problem, I noticed that if I substitute the edge from node "user_risk_level_acceptance" to END, with an edge that goes from "user_risk_level_acceptance" back to "user_profiling_agent" I get a consistent graph. So, in the code, substituting builder.add_edge("user_risk_level_acceptance", END) with builder.add_edge("user_risk_level_acceptance", "user_profiling_agent"), I get the expected graph, as shown below:

Have I missed something in the first version of the code? Any help or clarification will be really appreciated! Thanks!

[aidiary]

I am also experiencing the same issue. When I try to render the following graph using Mermaid, the arrow pointing from the "conversation" node to "END" is not displayed.

My langgraph version is 0.3.34.

from IPython.display import Image, display
from langgraph.graph import StateGraph, START, END
from typing import Literal
from langgraph.graph import MessagesState


class State(MessagesState):
    summary: str


def call_model(state: State):
    response = "dummy response"
    return {"messages": response}


def summarize_conversation(state: State):
    return {"summary": "dummy summary", "messages": []}


# Determine whether to end or summarize the conversation
def should_continue(state: State) -> Literal["summarize_conversation", "__end__"]:
    """Return the next node to execute."""

    messages = state["messages"]

    # If there are more than six messages, then we summarize the conversation
    if len(messages) > 6:
        return "summarize_conversation"

    # Otherwise we can just end
    return END


# Define a new graph
workflow = StateGraph(State)
workflow.add_node("conversation", call_model)
workflow.add_node(summarize_conversation)

# Set the entrypoint as conversation
workflow.add_edge(START, "conversation")
workflow.add_conditional_edges("conversation", should_continue)
workflow.add_edge("summarize_conversation", END)

# Compile
graph = workflow.compile()
display(Image(graph.get_graph().draw_mermaid_png()))

[shruti9619]

@fabio-barbieri @aidiary I am also facing the same issue. Also, I think there's one more approach to defining a decision node and with that approach the decision node also shows up in the graph here: https://youtu.be/jGg_1h0qzaM?si=LMf2Q5iyZlS7REfd
I am not sure why this should happen.