langchain-ai/langgraph
1
0
Fork 0
mirror of https://github.com/langchain-ai/langgraph.git synced 2026-07-15 12:35:43 -04:00
Code Issues 314 Packages Projects Releases 17 Wiki Activity

Multi-agent supervisor gives json related error #328

New Issue
Closed
opened 2026-02-20 17:36:44 -05:00 by yindo · 4 comments
yindo commented 2026-02-20 17:36:44 -05:00
Owner
Copy Link
Copy Source

Originally created by @sfallahpour on GitHub (Nov 29, 2024).

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

https://colab.research.google.com/drive/1QszbxpiFJkhWWYhBpSmDCX_3MMMeHVdd?usp=sharing#scrollTo=UiWxjTxDNNOe


    class RouteResponse(TypedDict):
        """Worker to route to next. If no workers needed, route to FINISH."""
        next: Union[Literal["FINISH"], Literal["Rag_agent"], Literal["Sql_agent"]]

Error Message and Stack Trace (if applicable)

OutputParserException: Function RouteResponse arguments:

     {
       next: "Rag_agent"
     }


     are not valid JSON. Received JSONDecodeError Expecting property name enclosed in double quotes: 
     line 2 column 3 (char 4)
     For troubleshooting, visit: 
     https://python.langchain.com/docs/troubleshooting/errors/OUTPUT_PARSING_FAILURE
     For troubleshooting, visit: 
     https://python.langchain.com/docs/troubleshooting/errors/OUTPUT_PARSING_FAILURE

Description

i used the multi agent supervisor code: https://github.com/langchain-ai/langgraph/discussions/683

and i created similar agentic workflow with two agents (rag and sql agent). My code is here:
https://colab.research.google.com/drive/1QszbxpiFJkhWWYhBpSmDCX_3MMMeHVdd?usp=sharing#scrollTo=XbyYy3HU3J--

and i get the error related to RouteResponse argument not generating in a json format.
even the

 class Router(TypedDict):
       """Worker to route to next. If no workers needed, route to FINISH."""
       next: Literal[*options]

gives me syntax error because of asterisk (*) which i had to replace it with

 class RouteResponse(TypedDict):
      """Worker to route to next. If no workers needed, route to FINISH."""
      next: Union[Literal["FINISH"], Literal["Rag_agent"], Literal["Sql_agent"]]

i replicated the github code as is and i got same error. So the router function instead of generating next action in json like

{'supervisor': {'next': 'Rag_agent'}})

It generates {'next': 'Rag_agent'} which is a non json format

image

System Info

langchain 0.3.7
langchain-anthropic 0.3.0
langchain-community 0.3.7
langchain-core 0.3.19
langchain-experimental 0.3.3
langchain-openai 0.2.9
langchain-text-splitters 0.3.2
langgraph 0.2.52
langgraph-checkpoint 2.0.4
langgraph-sdk 0.1.36
langsmith 0.1.143

Originally created by @sfallahpour on GitHub (Nov 29, 2024). ### Checked other resources - [x] I added a very descriptive title to this issue. - [X] I searched the [LangGraph](https://langchain-ai.github.io/langgraph/)/LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangGraph/LangChain rather than my code. - [X] I am sure this is better as an issue [rather than a GitHub discussion](https://github.com/langchain-ai/langgraph/discussions/new/choose), since this is a LangGraph bug and not a design question. ### Example Code ```python https://colab.research.google.com/drive/1QszbxpiFJkhWWYhBpSmDCX_3MMMeHVdd?usp=sharing#scrollTo=UiWxjTxDNNOe class RouteResponse(TypedDict): """Worker to route to next. If no workers needed, route to FINISH.""" next: Union[Literal["FINISH"], Literal["Rag_agent"], Literal["Sql_agent"]] ``` ### Error Message and Stack Trace (if applicable) ```shell OutputParserException: Function RouteResponse arguments: { next: "Rag_agent" } are not valid JSON. Received JSONDecodeError Expecting property name enclosed in double quotes: line 2 column 3 (char 4) For troubleshooting, visit: https://python.langchain.com/docs/troubleshooting/errors/OUTPUT_PARSING_FAILURE For troubleshooting, visit: https://python.langchain.com/docs/troubleshooting/errors/OUTPUT_PARSING_FAILURE ``` ### Description i used the multi agent supervisor code: https://github.com/langchain-ai/langgraph/discussions/683 and i created similar agentic workflow with two agents (rag and sql agent). My code is here: https://colab.research.google.com/drive/1QszbxpiFJkhWWYhBpSmDCX_3MMMeHVdd?usp=sharing#scrollTo=XbyYy3HU3J-- and i get the error related to RouteResponse argument not generating in a json format. even the class Router(TypedDict): """Worker to route to next. If no workers needed, route to FINISH.""" next: Literal[*options] gives me syntax error because of asterisk (*) which i had to replace it with class RouteResponse(TypedDict): """Worker to route to next. If no workers needed, route to FINISH.""" next: Union[Literal["FINISH"], Literal["Rag_agent"], Literal["Sql_agent"]] i replicated the github code as is and i got same error. So the router function instead of generating next action in json like {'supervisor': {'next': 'Rag_agent'}}) It generates {'next': 'Rag_agent'} which is a non json format ![image](https://github.com/user-attachments/assets/43b3b852-86b3-44b8-b49b-e0418a30d6a1) ### System Info langchain 0.3.7 langchain-anthropic 0.3.0 langchain-community 0.3.7 langchain-core 0.3.19 langchain-experimental 0.3.3 langchain-openai 0.2.9 langchain-text-splitters 0.3.2 langgraph 0.2.52 langgraph-checkpoint 2.0.4 langgraph-sdk 0.1.36 langsmith 0.1.143
yindo closed this issue 2026-02-20 17:36:44 -05:00
yindo commented 2026-02-20 17:36:44 -05:00
Author
Owner
Copy Link
Copy Source

@danielmartinscabanas commented on GitHub (Dec 3, 2024):

I think you also need to pass the arguments of the state object in the returning object of your supervisor_node, try this version below:

def supervisor_node(state: AgentState) -> AgentState:
    messages = [
        {"role": "system", "content": system_prompt},
    ] + state["messages"]
    response = llm.with_structured_output(RouteResponse).invoke(messages)
    next_ = response["next"]
    if next_ == "FINISH":
        next_ = END

    return {**state, "next": next_} # Added the `**state` here
@danielmartinscabanas commented on GitHub (Dec 3, 2024): I think you also need to pass the arguments of the state object in the returning object of your `supervisor_node`, try this version below: ``` def supervisor_node(state: AgentState) -> AgentState: messages = [ {"role": "system", "content": system_prompt}, ] + state["messages"] response = llm.with_structured_output(RouteResponse).invoke(messages) next_ = response["next"] if next_ == "FINISH": next_ = END return {**state, "next": next_} # Added the `**state` here ```
yindo commented 2026-02-20 17:36:44 -05:00
Author
Owner
Copy Link
Copy Source

@vbarda commented on GitHub (Dec 7, 2024):

@sfallahpour try replacing

class RouteResponse(TypedDict):
      """Worker to route to next. If no workers needed, route to FINISH."""
      next: Union[Literal["FINISH"], Literal["Rag_agent"], Literal["Sql_agent"]]

with

 class RouteResponse(TypedDict):
      """Worker to route to next. If no workers needed, route to FINISH."""
      next: Literal["FINISH", "Rag_agent", "Sql_agent"]

think the schema might not be created correctly with unions here

@vbarda commented on GitHub (Dec 7, 2024): @sfallahpour try replacing ```python class RouteResponse(TypedDict): """Worker to route to next. If no workers needed, route to FINISH.""" next: Union[Literal["FINISH"], Literal["Rag_agent"], Literal["Sql_agent"]] ``` with ```python class RouteResponse(TypedDict): """Worker to route to next. If no workers needed, route to FINISH.""" next: Literal["FINISH", "Rag_agent", "Sql_agent"] ``` think the schema might not be created correctly with unions here
yindo commented 2026-02-20 17:36:44 -05:00
Author
Owner
Copy Link
Copy Source

@vbarda commented on GitHub (Dec 7, 2024):

@danielmartinscabanas this is actually not necessary -- returns from the node functions are the state updates, so you don't need to manually add current state to the returned update

@vbarda commented on GitHub (Dec 7, 2024): @danielmartinscabanas this is actually not necessary -- returns from the node functions are the state updates, so you don't need to manually add current state to the returned update
yindo commented 2026-02-20 17:36:44 -05:00
Author
Owner
Copy Link
Copy Source

@vbarda commented on GitHub (Dec 17, 2024):

Will close this due to inactivity, but please let me know if this is still an issue

@vbarda commented on GitHub (Dec 17, 2024): Will close this due to inactivity, but please let me know if this is still an issue
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
agents
awaiting author revision
bug
change
chat models
checkpointer
ci
code quality
code quality
dependencies
documentation
drawing
duplicate
enhancement
good first issue
help wanted
invalid
maintainer
meta
missing information
not maintained by langchain
open-swe-auto
open-swe-max
open-swe-max-auto
pending
performance
pull-request
Mirrored from GitHub Pull Request
 question
requires design
sdk
server
stale
studio
tests
type checking
usage question
v0.5
v1.1
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langchain-ai/langgraph#328
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?