[SOLVED] LangGraph plots incorrectly the workflow #211

New Issue

Closed

opened 2026-02-20 17:31:56 -05:00 by yindo · 6 comments

yindo commented 2026-02-20 17:31:56 -05:00

Owner

Copy Link

Copy Source

Originally created by @Nachoeigu on GitHub (Aug 26, 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

import os
from dotenv import load_dotenv
import sys

load_dotenv()
WORKDIR=os.getenv("WORKDIR")
os.chdir(WORKDIR)
sys.path.append(WORKDIR)

from langchain_google_genai.chat_models import ChatGoogleGenerativeAI
from langchain_openai.chat_models import ChatOpenAI
from langgraph.graph import StateGraph
from src.utils import State, GraphInput, GraphOutput, GraphConfig, check_chapter
from src.nodes import *
from src.routers import *
from langgraph.graph import END
from src.utils import State

def should_go_to_brainstorming_writer(state: State):
    if state.get('instructor_documents', '') == '':
        return "human_feedback"
    else:
        return "brainstorming_writer"
    
def should_continue_with_critique(state: State):
    if state.get('is_plan_approved', None) is None: 
        return "brainstorming_critique"
    elif state['is_plan_approved'] == True:
        return "writer"
    else:
        return "brainstorming_critique"
    
def has_writer_ended_book(state: State):
    if state['current_chapter'] == len(state['chapters_summaries']):
        return END
    else:
        return "writer"

workflow = StateGraph(State, 
                      input = GraphInput,
                      config_schema = GraphConfig)

workflow.add_node("instructor", get_clear_instructions)
workflow.set_entry_point("instructor")
workflow.add_node("human_feedback", read_human_feedback)
workflow.add_node("brainstorming_writer", making_writer_brainstorming)
workflow.add_node("brainstorming_critique", brainstorming_critique)
workflow.add_node("writer", generate_content)
workflow.add_conditional_edges(
    "instructor",
    should_go_to_brainstorming_writer
)
workflow.add_edge("human_feedback","instructor")
workflow.add_conditional_edges(
    "brainstorming_writer",
    should_continue_with_critique
)

workflow.add_edge("brainstorming_critique","brainstorming_writer")
workflow.add_conditional_edges(
    "writer",
    has_writer_ended_book
)

app = workflow.compile(
    interrupt_before=['human_feedback']
    )

Error Message and Stack Trace (if applicable)

No response

Description

LangGraph plots my workflow in the following way, which is not correct:

The workflow should look like:

It created relationships where there are no ones.

System Info

langchain 0.2.14
langchain-community 0.2.12
langchain-core 0.2.34
langchain-google-genai 1.0.10
langchain-groq 0.1.9
langchain-openai 0.1.22
langchain-text-splitters 0.2.2
langgraph 0.2.14
langgraph-checkpoint 1.0.6
langsmith 0.1.104

Originally created by @Nachoeigu on GitHub (Aug 26, 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 import os from dotenv import load_dotenv import sys load_dotenv() WORKDIR=os.getenv("WORKDIR") os.chdir(WORKDIR) sys.path.append(WORKDIR) from langchain_google_genai.chat_models import ChatGoogleGenerativeAI from langchain_openai.chat_models import ChatOpenAI from langgraph.graph import StateGraph from src.utils import State, GraphInput, GraphOutput, GraphConfig, check_chapter from src.nodes import * from src.routers import * from langgraph.graph import END from src.utils import State def should_go_to_brainstorming_writer(state: State): if state.get('instructor_documents', '') == '': return "human_feedback" else: return "brainstorming_writer" def should_continue_with_critique(state: State): if state.get('is_plan_approved', None) is None: return "brainstorming_critique" elif state['is_plan_approved'] == True: return "writer" else: return "brainstorming_critique" def has_writer_ended_book(state: State): if state['current_chapter'] == len(state['chapters_summaries']): return END else: return "writer" workflow = StateGraph(State, input = GraphInput, config_schema = GraphConfig) workflow.add_node("instructor", get_clear_instructions) workflow.set_entry_point("instructor") workflow.add_node("human_feedback", read_human_feedback) workflow.add_node("brainstorming_writer", making_writer_brainstorming) workflow.add_node("brainstorming_critique", brainstorming_critique) workflow.add_node("writer", generate_content) workflow.add_conditional_edges( "instructor", should_go_to_brainstorming_writer ) workflow.add_edge("human_feedback","instructor") workflow.add_conditional_edges( "brainstorming_writer", should_continue_with_critique ) workflow.add_edge("brainstorming_critique","brainstorming_writer") workflow.add_conditional_edges( "writer", has_writer_ended_book ) app = workflow.compile( interrupt_before=['human_feedback'] ) ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description LangGraph plots my workflow in the following way, which is not correct: <img width="722" alt="image" src="https://github.com/user-attachments/assets/239b422e-8a74-4c83-b977-ba908d1b29e3"> The workflow should look like:  It created relationships where there are no ones. ### System Info langchain 0.2.14 langchain-community 0.2.12 langchain-core 0.2.34 langchain-google-genai 1.0.10 langchain-groq 0.1.9 langchain-openai 0.1.22 langchain-text-splitters 0.2.2 langgraph 0.2.14 langgraph-checkpoint 1.0.6 langsmith 0.1.104

yindo closed this issue 2026-02-20 17:31:57 -05:00

yindo commented 2026-02-20 17:31:59 -05:00

Author

Owner

Copy Link

Copy Source

@gbaian10 commented on GitHub (Aug 26, 2024):

• ImportError: cannot import name 'REDIS_MAX_IDLE_TIME' from 'langgraph_api.config' (#987)
• [PR #342] [MERGED] Add `then=` arg for add_conditional_edge and set_conditional_entry_point (#1394)

@gbaian10 commented on GitHub (Aug 26, 2024): - #987 - #1394

yindo commented 2026-02-20 17:32:01 -05:00

Author

Owner

Copy Link

Copy Source

@Nachoeigu commented on GitHub (Aug 26, 2024):

• add_conditional_edges does not work without optional argument path_map #987
• Visualized graph has wrong conditional edges #1394

Thank you very much, I couldn´t find them in my research.

I think a good improvement in the framework could be that automaticaly detects the return possible values of the router and, based on it, make the relationships. What do you think?

@Nachoeigu commented on GitHub (Aug 26, 2024): > * [`add_conditional_edges` does not work without optional argument `path_map` #987](https://github.com/langchain-ai/langgraph/issues/987) > * [Visualized graph has wrong conditional edges #1394](https://github.com/langchain-ai/langgraph/issues/1394) Thank you very much, I couldn´t find them in my research. I think a good improvement in the framework could be that automaticaly detects the return possible values of the router and, based on it, make the relationships. What do you think?

yindo commented 2026-02-20 17:32:01 -05:00

Author

Owner

Copy Link

Copy Source

@gbaian10 commented on GitHub (Aug 26, 2024):

• add_conditional_edges does not work without optional argument path_map #987
• Visualized graph has wrong conditional edges #1394

Thank you very much, I couldn´t find them in my research.

I think a good improvement in the framework could be that automaticaly detects the return possible values of the router and, based on it, make the relationships. What do you think?

It seems to be mentioned here.
https://langchain-ai.github.io/langgraph/cloud/faq/studio/?h=router#why-are-extra-edges-showing-up-in-my-graph

If it could automatically detect and complete, that would be even better. After all, many people have raised the same issue.
At least I'm sure VSCode can automatically detect return types, which proves it's feasible from a programming perspective. I'm just not sure how difficult it would be to implement.

@gbaian10 commented on GitHub (Aug 26, 2024): > > * [`add_conditional_edges` does not work without optional argument `path_map` #987](https://github.com/langchain-ai/langgraph/issues/987) > > * [Visualized graph has wrong conditional edges #1394](https://github.com/langchain-ai/langgraph/issues/1394) > > Thank you very much, I couldn´t find them in my research. > > I think a good improvement in the framework could be that automaticaly detects the return possible values of the router and, based on it, make the relationships. What do you think? It seems to be mentioned here. https://langchain-ai.github.io/langgraph/cloud/faq/studio/?h=router#why-are-extra-edges-showing-up-in-my-graph If it could automatically detect and complete, that would be even better. After all, many people have raised the same issue. At least I'm sure VSCode can automatically detect return types, which proves it's feasible from a programming perspective. I'm just not sure how difficult it would be to implement.   

yindo commented 2026-02-20 17:32:04 -05:00

Author

Owner

Copy Link

Copy Source

@Nachoeigu commented on GitHub (Aug 26, 2024):

• add_conditional_edges does not work without optional argument path_map #987
• Visualized graph has wrong conditional edges #1394

Thank you very much, I couldn´t find them in my research.
I think a good improvement in the framework could be that automaticaly detects the return possible values of the router and, based on it, make the relationships. What do you think?

It seems to be mentioned here. https://langchain-ai.github.io/langgraph/cloud/faq/studio/?h=router#why-are-extra-edges-showing-up-in-my-graph

If it could automatically detect and complete, that would be even better. After all, many people have raised the same issue. At least I'm sure VSCode can automatically detect return types, which proves it's feasible from a programming perspective. I'm just not sure how difficult it would be to implement.

I think with ast something like that could be possible.

import ast
import inspect
from langgraph.graph import END
import pandas as pd

class ReturnValueExtractor(ast.NodeVisitor):
    """
    A class to extract all possible return expressions from a Python function.
    
    Attributes:
        return_expressions (set): A set of unique return expressions found in the function.
    """

    def __init__(self):
        """Initialize the ReturnValueExtractor with an empty set of return expressions."""
        self.return_expressions = set()

    def visit_Return(self, node):
        """
        Visit a return statement in the AST and extract the return expression.

        Args:
            node (ast.Return): The return node in the AST.
        """
        if node.value:
            # Convert the return expression to a string and add to the set
            self.return_expressions.add(ast.dump(node.value, annotate_fields=False))

        # Continue walking the AST
        self.generic_visit(node)

    def extract(self, func):
        """
        Extract all return expressions from the given function.

        Args:
            func (function): The function from which to extract return expressions.

        Returns:
            list: A list of unique return expressions as strings.
        """
        source = inspect.getsource(func)
        tree = ast.parse(source)
        self.visit(tree)
        
        return list(self.return_expressions)

# Example function to test
def example_function(state):
    if state.get('is_plan_approved', None) is None: 
        return "brainstorming_critique"
    elif state['is_plan_approved'] == True:
        return "writer"
    elif state['is_plan_approved'] == False:
        return "another_agent"
    else:
        return END


def complex_function(state):
    if state.get('is_ready', False):
        return END
    elif 1+1 == 3:
        return pd.DataFrame()
    else:
        return example_function
if __name__ == '__main__':
    extractor = ReturnValueExtractor()
    possible_returns = extractor.extract(example_function)
    print(possible_returns)
    print('--------------')
    possible_returns = extractor.extract(complex_function)
    print(possible_returns)

@Nachoeigu commented on GitHub (Aug 26, 2024): > > > * [`add_conditional_edges` does not work without optional argument `path_map` #987](https://github.com/langchain-ai/langgraph/issues/987) > > > * [Visualized graph has wrong conditional edges #1394](https://github.com/langchain-ai/langgraph/issues/1394) > > > > > > Thank you very much, I couldn´t find them in my research. > > I think a good improvement in the framework could be that automaticaly detects the return possible values of the router and, based on it, make the relationships. What do you think? > > It seems to be mentioned here. https://langchain-ai.github.io/langgraph/cloud/faq/studio/?h=router#why-are-extra-edges-showing-up-in-my-graph > > If it could automatically detect and complete, that would be even better. After all, many people have raised the same issue. At least I'm sure VSCode can automatically detect return types, which proves it's feasible from a programming perspective. I'm just not sure how difficult it would be to implement. > >  > >  > >  I think with ast something like that could be possible. ``` import ast import inspect from langgraph.graph import END import pandas as pd class ReturnValueExtractor(ast.NodeVisitor): """ A class to extract all possible return expressions from a Python function. Attributes: return_expressions (set): A set of unique return expressions found in the function. """ def __init__(self): """Initialize the ReturnValueExtractor with an empty set of return expressions.""" self.return_expressions = set() def visit_Return(self, node): """ Visit a return statement in the AST and extract the return expression. Args: node (ast.Return): The return node in the AST. """ if node.value: # Convert the return expression to a string and add to the set self.return_expressions.add(ast.dump(node.value, annotate_fields=False)) # Continue walking the AST self.generic_visit(node) def extract(self, func): """ Extract all return expressions from the given function. Args: func (function): The function from which to extract return expressions. Returns: list: A list of unique return expressions as strings. """ source = inspect.getsource(func) tree = ast.parse(source) self.visit(tree) return list(self.return_expressions) # Example function to test def example_function(state): if state.get('is_plan_approved', None) is None: return "brainstorming_critique" elif state['is_plan_approved'] == True: return "writer" elif state['is_plan_approved'] == False: return "another_agent" else: return END def complex_function(state): if state.get('is_ready', False): return END elif 1+1 == 3: return pd.DataFrame() else: return example_function if __name__ == '__main__': extractor = ReturnValueExtractor() possible_returns = extractor.extract(example_function) print(possible_returns) print('--------------') possible_returns = extractor.extract(complex_function) print(possible_returns) ```

yindo commented 2026-02-20 17:32:05 -05:00

Author

Owner

Copy Link

Copy Source

@gbaian10 commented on GitHub (Aug 26, 2024):

If you include some situations where None is implicitly returned, the problem becomes very complex.
It will be much simpler if you only capture all explicitly returned values to check if they are Liteval, leaving the rest to the user.

def add(a: int, b: int):
    if a > b:
        raise ValueError("a should be less than b")
    if a == 1:
        return 1
    if b == 2:
        return "2"
    # return a + b

For example, it might return None, but you cannot easily complete this check just by getting the return value; more complex validation is required.

@gbaian10 commented on GitHub (Aug 26, 2024): If you include some situations where None is implicitly returned, the problem becomes very complex. It will be much simpler if you only capture all explicitly returned values to check if they are Liteval, leaving the rest to the user. ```py def add(a: int, b: int): if a > b: raise ValueError("a should be less than b") if a == 1: return 1 if b == 2: return "2" # return a + b ``` For example, it might return None, but you cannot easily complete this check just by getting the return value; more complex validation is required.

yindo commented 2026-02-20 17:32:07 -05:00

Author

Owner

Copy Link

Copy Source

@gbaian10 commented on GitHub (Aug 26, 2024):

import ast
import inspect
from collections.abc import Callable
from typing import Any


def is_literal_return(func: Callable) -> bool:
    """
    Check if the function only returns constant values.

    Args:
        func (Callable): The function to analyze.

    Returns:
        bool: True if all return statements are constant values, False otherwise.
    """
    source = inspect.getsource(func)
    tree = ast.parse(source)
    return all(
        isinstance(node.value, ast.Constant)
        for node in ast.walk(tree)
        if isinstance(node, ast.Return)
    )


def has_explicit_return(func: Callable) -> bool:
    """
    Check if all code paths in the function have an explicit return statement.

    Args:
        func (Callable): The function to analyze.

    Returns:
        bool: True if all paths have explicit return statements, False otherwise.
    """
    source = inspect.getsource(func)
    tree = ast.parse(source)
    function_defs = [
        node for node in ast.walk(tree) if isinstance(node, ast.FunctionDef)
    ]
    if not function_defs:
        return False
    return check_all_paths_return(function_defs[0].body)


def check_all_paths_return(statements: list) -> bool:
    """
    Recursively check if all paths in the given statements have a return statement.

    Args:
        statements (list): A list of AST nodes representing statements.

    Returns:
        bool: True if all paths have return statements, False otherwise.
    """
    has_return = False
    for stmt in statements:
        if isinstance(stmt, ast.Return | ast.Raise):
            has_return = True
        elif isinstance(stmt, ast.If):
            has_return = check_all_paths_return(stmt.body) and check_all_paths_return(
                stmt.orelse
            )
        elif isinstance(stmt, ast.While | ast.For):
            continue
    return has_return


def analyze_function(func: Callable) -> set[Any] | None:
    """
    Analyze the function to determine if it returns a set of constant values.

    Args:
        func (Callable): The function to analyze.

    Returns:
        Optional[Set[Any]]: A set of constant values if the function meets the criteria, None otherwise.
    """
    if not is_literal_return(func) or not has_explicit_return(func):
        return None
    source = inspect.getsource(func)
    tree = ast.parse(source)
    return {node.value.value for node in ast.walk(tree) if isinstance(node, ast.Return)}


# Test examples
def example_func() -> int:
    if True:
        return 42
    return 42


def add(a: int, b: int):
    if a > b:
        raise ValueError("a should be less than b")
    if a == 1:
        return 1
    if b == 2:
        return "2"
    # return a + b


def ok():
    return "ok"


def no_ok():
    while 1 + 1 == 3:
        return 1


def one():
    return 1


# Usage examples
print(analyze_function(example_func))  # {42}
print(analyze_function(add))  # None
print(analyze_function(ok))  # {'ok'}
print(analyze_function(no_ok))  # None
print(analyze_function(one))  # {1}

I used an LLM to generate this code and I'm not sure if it covers all scenarios. Since I rarely use AST, I haven't fully reviewed its logic to ensure it's correct.

I expect the following order of rules:

1. If path_map is included, use the path provided by path_map as a priority.
2. If path_map is not included, read the return type hint of the path function.
3. If neither is included, execute AST to check if the return value always returns a Liteval result in all possible cases. If so, use the set or list of that Liteval result.
4. If it returns None, or if AST checking is disabled, issue a warning indicating that this point will lead to all points.

Three and four are new, but three would be a breaking change, so it's not suitable for a minor release unless the default is set to False.

@gbaian10 commented on GitHub (Aug 26, 2024): ```py import ast import inspect from collections.abc import Callable from typing import Any def is_literal_return(func: Callable) -> bool: """ Check if the function only returns constant values. Args: func (Callable): The function to analyze. Returns: bool: True if all return statements are constant values, False otherwise. """ source = inspect.getsource(func) tree = ast.parse(source) return all( isinstance(node.value, ast.Constant) for node in ast.walk(tree) if isinstance(node, ast.Return) ) def has_explicit_return(func: Callable) -> bool: """ Check if all code paths in the function have an explicit return statement. Args: func (Callable): The function to analyze. Returns: bool: True if all paths have explicit return statements, False otherwise. """ source = inspect.getsource(func) tree = ast.parse(source) function_defs = [ node for node in ast.walk(tree) if isinstance(node, ast.FunctionDef) ] if not function_defs: return False return check_all_paths_return(function_defs[0].body) def check_all_paths_return(statements: list) -> bool: """ Recursively check if all paths in the given statements have a return statement. Args: statements (list): A list of AST nodes representing statements. Returns: bool: True if all paths have return statements, False otherwise. """ has_return = False for stmt in statements: if isinstance(stmt, ast.Return | ast.Raise): has_return = True elif isinstance(stmt, ast.If): has_return = check_all_paths_return(stmt.body) and check_all_paths_return( stmt.orelse ) elif isinstance(stmt, ast.While | ast.For): continue return has_return def analyze_function(func: Callable) -> set[Any] | None: """ Analyze the function to determine if it returns a set of constant values. Args: func (Callable): The function to analyze. Returns: Optional[Set[Any]]: A set of constant values if the function meets the criteria, None otherwise. """ if not is_literal_return(func) or not has_explicit_return(func): return None source = inspect.getsource(func) tree = ast.parse(source) return {node.value.value for node in ast.walk(tree) if isinstance(node, ast.Return)} # Test examples def example_func() -> int: if True: return 42 return 42 def add(a: int, b: int): if a > b: raise ValueError("a should be less than b") if a == 1: return 1 if b == 2: return "2" # return a + b def ok(): return "ok" def no_ok(): while 1 + 1 == 3: return 1 def one(): return 1 # Usage examples print(analyze_function(example_func)) # {42} print(analyze_function(add)) # None print(analyze_function(ok)) # {'ok'} print(analyze_function(no_ok)) # None print(analyze_function(one)) # {1} ``` I used an LLM to generate this code and I'm not sure if it covers all scenarios. Since I rarely use AST, I haven't fully reviewed its logic to ensure it's correct. I expect the following order of rules: 1. If path_map is included, use the path provided by path_map as a priority. 1. If path_map is not included, read the return type hint of the path function. 1. If neither is included, execute AST to check if the return value always returns a Liteval result in all possible cases. If so, use the set or list of that Liteval result. 1. If it returns None, or if AST checking is disabled, issue a warning indicating that this point will lead to all points. Three and four are new, but three would be a breaking change, so it's not suitable for a minor release unless the default is set to False.

yindo referenced this issue 2026-02-20 17:44:24 -05:00

[PR #211] [MERGED] Support updating/replacing existing messages in messagegraph #1307

Sign in to join this conversation.

No Branch/Tag Specified

Branches Tags

main

jdr/add-session-name

deps/uv-lock-upgrade

fix/root-messages-inbox-bound

open-swe/fix-thread-actions-auth

fix/sdk-integration-ci-secrets

corridor/fix-e588a881-1783566949463

long/update-state-delta-fix-2

long/delta-replay-fix

open-swe/convert-to-ty

hunter/persist-interrupt-writes

longquanzheng-patch-1

test-json-overwrite-sentinel

fix-delta-overwrite-binop-semantics

sydney-runkle/langgraph/delta-channel-checkpointer-warning

open-swe/langsmith-multi-project-replicas

sydney/langgraph/overwrite-json-erasure

release/cli-0.4.28

mdrxy/store-docstring-pass

chore/adopt-mypy-2.1.0

cb/fix-set-node-defaults-cache-policy-opt-out

cb/fix-delta-channel-migration-writes

nh/sdk-py-sync-subgraph-stream-fix

cb/fix-delta-channel-overwrite-replay-divergence

jdr/uv-check

nh/sdk-py-v3-streaming-overview-rfc

worktree-refactor-error-handler-as-policy

parker/time-travel

feat/hook-chaining

nh/namespace-keyed-subagents

long/delta-prune-comment

long/delta-exit-mode

sr/messages-delta-reducer-uuid-assignment

open-swe/c802e0ea-c9e5-c8c6-824a-c065dd2bfd6f

long/delta-chan-combine

sr/delta-channel-no-streaming

nh/streaming-for-alpha-release

v1.0.13

delta-channel-writes-based

nh/langgraph-api-test-branch

cb/remote-graph-update

sr/snapshot-freq-bench

delta-channel-writes-based-qlong-test-sends

release/langgraph-1.1.10

cb/new-streaming

nh/subgraph-lifecycle

sr/even-better-writes-idea

sr/deepagents-perf-delta-plus-addmsgs

sr/deepagents-perf-combo

optimize/add-messages-fast-path

perf/get-function-nonlocals

diff-channel-incremental-checkpointing

wfh/diff_channel_opt_2

sr/deferred-imports

nh/tools-channel-streaming

jacob/metadatamerging

nh/streaming-with-event-log

open-swe/873b84e2-cbfb-bfdd-5e41-5f638d107a55

nh/streaming-v2

wfh/bp

open-swe/a2e2088a-1fd1-eb4d-a880-5469f1ebb7b5

wfh/conformance_format

wfh/gocli

wfh/cli_validate

wfh/releases/cli

cc/stream_mode_events

cb/lg-readme-update

hunter/interrupt-callback-lg

wfh/rename-langgraph-json-to-langsmith-json

fix/deploy-name-underscore-validation

sr/another-time-travel-fix

cb/multiple-http-app-support-a79b

wfh/build_fail_check

long/advanced-graph-rust-core-published

wfh/swr_cache

cb/protocol-improvements

wfh/releases/langgraph/1.1.4

long/advanced-graph-rust-core-published-po3

long/advanced-graph-rust-core-published-libc

long/advanced-graph-rust-core

wfh/releases/checkpoitners/4.0.2

long/advanced-graph

open-swe/feb956aa-a579-4d04-6b03-2bd214f26c7a

sr/remote-graph-generics

jk/optimizations

jk/optimizations-rebased

jk/cache-optimizations

david/lg-deploy-remote-build

long/dr-0309-version

update-llmstxt

an/09mar/cli-list-deployments

an/09mar/lg-list-deployments

sr/rippling-asks

worktree-explore-subgraph-replay

sr/third-attempt

sr/alt-fix

sr/potential_fix

sr/new-time-travel-tests

sr/1.1-v2-default

sr/logos

mdrxy/cross-process-test

sr/simplify-invoke-interrupts

wfh/cli/bazelbaby

wfh/agent-server-config

sr/streaming-hints

wfh/postgres-extended-methods

wfh/build-checkpointer-skill

wfh/respect_defaults

jk/js-yaml

wfh/testpypi_pyproject

wfh/_/native_cli

david/cache-optimized-dockerfile-v2

jdr/add-append-only-mode

claude/durable-retry-policy-hohXq

fix/add-assistant-id-to-auth-types

sr/tool-node-dynamic

wfh/topology

wfh/fix_hasalive

wfh/allowlist

caspar/msgpack_allowlist

wfh/honor_traceable

wfh/prune

mdrxy/pyodide-wasm-support

wfh/astream_events

wfh/verify

wfh/remote_graph_wait

wfh/listen

long/versions-seen-learning

wfh/flip_default_base_cache

eugene/patch

sr/response-schema-mwahaha

wfh/release_sdk

wfh/increase_default

wfh/fix_ci

wfh/sdk_py/release

update-openapi-spec-v0.5.26

update-openapi-spec-v0.5.25

update-openapi-spec-v0.5.24

update-openapi-spec-v0.5.23

update-openapi-spec-v0.5.20

sr/tool-request

update-openapi-spec-v0.5.17

update-openapi-spec-v0.5.16

update-openapi-spec-v0.5.14

sr/dep-warning-fixes

sr/linting-fixes

mdrxy/add-return-types

issue_6409

sr/new-fix

wfh/cli_int_update

wfh/trigger_changelog

copilot/sub-pr-6316

0.6

caspar/interrupt_scheduling_optimization_fix

david/10-07/cache-optimized-dockerfile

update-openapi-spec-v0.4.39

harrison/overwrite

mdrxy/lift-core-dep

caspar/fix_subgraph_state_replay

mdrxy/lc-ref-consolidate

randall-aaa899d9-8505-4622-9aa0-5cb797c2db63

joaquin/test-push-protection-datadog-revoked-key

open-swe/01d3b562-feb8-4631-b7de-fc3d9970c700

wfh/omit_more2

caspar/functional-api-mypy-typing

long-test-push

sr/what-if-we-did-typed-dict

sr/model-request-as-node

nc/25aug/agent

wfh/update_type

sr/fix-tool-node

update-openapi-spec-v0.4.4

update-openapi-spec-v0.4.3

update-openapi-spec-v0.4.2

eugene/update_agent_implementation

update-openapi-spec-v0.3.4

open-swe/89802a32-3c61-420c-990b-e6bcebee5459

open-swe/239867ad-46ba-4f8c-ba65-0483e8bf54ac

sr/drop-3.9

david/08-21/update-nat-ips

wfh/0.6.6

update-openapi-spec-v0.2.135

eugene/inject_runtime_in_tools

eugene/add_structured_response

sr/structured-outputs-draft

open-swe/c85f411c-dfc5-4894-ab9a-124661d9ab43

open-swe/bc2ef87c-6f38-432d-b4c3-ac46c27feb34

wfh/accept_extra_args

wfh/bump_cli

sr/new-msg-support

eugene/test_multiple_interrupt_with_send

open-swe/13bd038a-d872-4620-8474-04453d65376e

open-swe/6567da56-b47e-4b9c-b9c8-0c15eb3573a6

eugene/allow_default_name

open-swe/ae08c4ba-71fe-4b12-bd45-dd7cdf82ebe8

changelog-bot-update-d303e041

sr/fix-dep-warning

open-swe/e61ca3a1-8559-4dcc-a859-476b55a12775

open-swe/1d37d949-de60-4999-85ec-b835dd3fa7f8

eugene/document_dynamic_models

remove-lgp-content-mkdocs

sr/fix-context-coercion

hunter/js-preview

open-swe/d532baf6-ca9e-4246-b3f0-3ca77d4ce3c8

sr/experimental-refactor

open-swe/e5610d6c-992d-4e1f-89c1-452f9f0cfd37

open-swe/97c901db-9ad0-4dc8-bad3-d64bed36d9b2

eugene/script_to_replace_cross_refs

copilot/fix-5692

eugene/getting-started

eugene/update_cross_refs

docs/js

open-swe/f801d704-025e-4441-950c-12ffb2484849

eugene/refactor_create_react_agent

sr/fixes-for-v6

open-swe/2f1220bd-1fac-476d-8346-9cd31ef4baeb

open-swe/773f034b-9148-4456-9365-059d120fc606

open-swe/654a2f9f-75b5-472d-8991-b817dd33f0a1

open-swe/a84821fc-711e-456f-baa1-5ee6451d91b2

hunter/translation-docs-review

eugene/add_dynamic_model

open-swe/27f6f9f5-5fcf-4761-b6b6-3e4e18b5ce17

open-swe/d0aec071-9bbd-41f2-9a9c-ae02e5cdd4e5

eugene/docs_customer_custom_workflow

dqbd/translation-ref

changelog-bot-update-5265f6ce

infra/remove-api-key-v0

infra/remove-license-key-v0

changelog-bot-update-01d2ca7b

sam_testing_bot_from_version_85

open-swe/f9241737-35f7-450d-8486-d3b69266be57

changelog-bot-update-19ff701b

changelog-bot-update-8fd71780

changelog-bot-update-fc671586

changelog-bot-update-96f0bf39

changelog-bot-update-6140158c

changelog-bot-update-0c2eec86

add-copy-page-button

sr/fix-3.11

wfh/update_doc

sr/min-deps-simplified

sr/min-deps-testing

sr/perms-for-pr

sr/remove-message-graph

eugene/nav_reshuffle_2

eugene/navigation_links

eugene/add_more_cross_links

0.4

eugene/tools

sr/invokable-state-node

arjun/docs_data_storage

arjun/docs_studio_evals

eugene/conditional-rendering-global

eugene/conditional-rendering

wfh/tavilyp2

open-swe/3817d5e5-d827-4dd8-9622-db5b5da74e0d

rlm/local-testing

wfh/golang

sr/run-against-checkpoint-min

arjun/consolidate_installation

sr/update-pydantic-in-lockfile

wfh/inlineforpeoplewhocantread

wfh/header_handling2

wfh/header_handling

wfh/anymessage

wfh/add_test_generics

wfh/validate_again

wfh/_/validate_native_time

wfh/mount_prefix_sdk

wfh/pg_example

wfh/memcheck

wfh/readme_nits

wfh/set_store

wfh/cli_from_module

wfh/cli_release_studio

wfh/gc_test_all

wfh/automaton

brian/typo

nc/1mar/java

nc/1mar/lite

dqbd/sdk-js-0.0.46

wfh/cli/http_with_faux

wfh/sdk/router

ben/ts-exec

rlm/agent_fxn_api

wfh/update_dev_msg

wfh/queue_metrics

wfh/release

wfh/_/illbeback

wfh/clarify_installation

wfh/scaling

wfh/simple_scaling

wfh/docs/memstoreconcept

wfh/_/ann

wfh/_/annp0

wfh/store/avs

wfh/store/avs_backup

wfh/config/store_config

wfh/_/pgsearch

wfh/winpaths

wfh/cli/inmem_update

erick/langgraph-get-rid-of-beta-uvloop-version

wfh/inmemory

wfh/update-stae-modifier

rlm/chat-langgraph

wfh/stateinjection

jacoblee93-patch-1

cli==0.4.31

1.2.9

1.2.8

1.2.7

1.2.6

cli==0.4.30

1.2.5

cli==0.4.29

cli==0.4.28

1.2.4

1.2.3

sdk==0.4.2

sdk==0.4.1

cli==0.4.27

sdk==0.4.0

1.2.2

checkpoint==4.1.1

sdk==0.3.15

1.2.1

prebuilt==1.1.0

checkpointpostgres==3.1.0

checkpoint==4.1.0

1.2.0

checkpointsqlite==3.1.0

cli==0.4.26

cli==0.4.25

checkpointsqlite==3.1.0a1

sdk==0.3.14

1.2.0a7

checkpointpostgres==3.1.0a4

checkpoint==4.1.0a4

1.2.0a6

1.2.0a5

prebuilt==1.1.0a2

prebuilt==1.1.0a1

1.2.0a4

1.2.0a3

checkpointpostgres==3.1.0a3

checkpoint==4.1.0a3

checkpoint==4.1.0a2

1.2.0a2

checkpointpostgres==3.1.0a1

prebuilt==1.0.13

1.2.0a1

checkpoint==4.1.0a1

1.1.10

prebuilt==1.0.12

checkpoint==4.0.3

prebuilt==1.0.11

cli==0.4.24

1.1.9

1.1.8

prebuilt==1.0.10

cli==0.4.23

1.1.7

cli==0.4.22

checkpoint==4.0.2

1.1.7a2

1.1.7a1

cli==0.4.21

cli==0.4.20

sdk==0.3.13

1.1.6

prebuilt==1.0.9

1.1.5

1.1.4

cli==0.4.19

1.1.3

sdk==0.3.12

checkpointpostgres==3.0.5

cli==0.4.18

cli==0.4.17

cli==0.4.16

1.1.2

1.1.1

sdk==0.3.11

1.1.0

cli==0.4.15

sdk==0.3.10

cli==0.4.14

1.0.10

checkpoint==4.0.1

checkpoint==4.0.1rc3

checkpoint==4.0.1rc2

1.0.10rc1

checkpoint==4.0.1rc1

sdk==0.3.9

sdk==0.3.8

prebuilt==1.0.8

1.0.9

sdk==0.3.7

sdk==0.3.6

sdk==0.3.5

1.0.8

sdk==0.3.4

checkpointpostgres==3.0.4

cli==0.4.12

prebuilt==1.0.7

1.0.7

checkpointsqlite==3.0.3

sdk==0.3.3

checkpoint==4.0.0

1.0.6

checkpointpostgres==3.0.3

checkpointsqlite==3.0.2

prebuilt==1.0.6

sdk==0.3.2

sdk==0.3.1

cli==0.4.11

cli==0.4.10

1.0.5

sdk==0.3.0

checkpointpostgres==3.0.2

checkpointsqlite==3.0.1

sdk==0.2.15

cli==0.4.9

cli==0.4.8

sdk==0.2.14

sdk==0.2.12

1.0.4

sdk==0.2.10

prebuilt==1.0.5

prebuilt==1.0.4

prebuilt==1.0.3

1.0.3

checkpointpostgres==3.0.1

checkpoint==3.0.1

cli==0.4.7

cli==0.4.6

cli==0.4.5

1.0.2

prebuilt==1.0.2

prebuilt==0.6.5

0.6.11

checkpointpostgres==3.0.0

prebuilt==1.0.1

checkpointsqlite==3.0.0

1.0.1

checkpoint==3.0.0

1.0.0

prebuilt==1.0.0

prebuilt==0.7.0rc1

cli==0.4.4

1.0.0rc1

0.6.10

cli==0.4.3

0.6.9

checkpointpostgres==2.0.25

checkpoint==2.1.2

checkpointpostgres==2.0.24

1.0.0a4

0.6.8

sdk==0.2.9

sdk==0.2.8

sdk==0.2.7

1.0.0a3

0.6.7

cli==0.4.2

cli==0.4.1

sdk==0.2.6

sdk==0.2.5

1.0.0a2

prebuilt==0.7.0a2

sdk==0.2.4

1.0.0a1

prebuilt==0.7.0a1

cli==0.4.0

sdk==0.2.3

cli==0.3.8

cli==0.3.7

0.6.6

sdk==0.2.2

sdk==0.2.1

0.6.5

prebuilt==0.6.4

0.6.4

0.6.3

prebuilt==0.6.3

0.6.2

prebuilt==0.6.2

0.6.1

prebuilt==0.6.1

0.6.0

prebuilt==0.6.0

0.6.0a2

prebuilt==0.6.0a1

checkpointsqlite==2.0.11

cli==0.3.6

sdk==0.2.0

sdk==0.2.0a1

0.6.0a1

0.5.4

sdk==0.1.74

cli==0.3.5

checkpoint==2.1.1

checkpointpostgres==2.0.23

sdk==0.1.73

0.5.3

checkpointpostgres==2.0.22

0.5.2

cli==0.3.4

0.5.1

prebuilt==0.5.2

prebuilt==0.5.1

sdk==0.1.72

0.5.0

prebuilt==0.5.0

sdk==0.1.71

prebuilt==0.2.3

0.4.10

0.4.9

0.5.0rc1

prebuilt==0.5.0rc0

0.5.0rc0

checkpoint==2.1.0

cli==0.3.3

cli==0.3.2

cli==0.3.1

cli==0.2.12

cli==0.2.11

0.4.8

prebuilt==0.2.2

0.4.7

0.4.6

prebuilt==0.2.1

prebuilt==0.2.0

sdk==0.1.70

checkpointsqlite==2.0.10

checkpointsqlite==2.0.9

checkpointsqlite==2.0.8

0.4.5

checkpoint==2.0.26

0.4.4

sdk==0.1.69

cli==0.2.10

cli==0.2.9

0.4.3

0.4.2

checkpointsqlite==2.0.7

cli==0.2.8

sdk==0.1.66

sdk==0.1.65

0.4.1

sdk==0.1.64

0.4.0

checkpoint==2.0.25

0.3.34

0.3.32

cli==0.2.7

cli==0.2.6

sdk==0.1.63

sdk==0.1.62

checkpointpostgres==2.0.21

checkpointpostgres==2.0.20

cli==0.2.5

0.3.31

cli==0.2.4

0.3.30

0.3.29

cli==0.2.3

0.3.28

cli==0.2.2

cli==0.2.1

0.3.27

0.3.26

cli==0.1.89

0.3.25

prebuilt==0.1.8

cli==0.1.84

sdk==0.1.61

0.3.24

checkpoint==2.0.24

0.3.23

cli==0.1.83

0.3.22

cli==0.1.82

cli==0.1.81

sdk==0.1.60

0.3.21

prebuilt==0.1.7

cli==0.1.80

sdk==0.1.59

prebuilt==0.1.6

prebuilt==0.1.5

0.3.20

checkpoint==2.0.23

checkpoint==2.0.22

0.3.19

cli==0.1.79

prebuilt==0.1.4

cli==0.1.78

checkpointpostgres==2.0.19

0.3.18

sdk==0.1.58

0.3.17

0.3.16

0.3.15

0.3.14

0.3.13

0.3.12

checkpoint==2.0.21

checkpointpostgres==2.0.18

0.3.11

checkpointpostgres==2.0.17

cli==0.1.77

checkpoint==2.0.20

0.3.10

sdk==0.1.57

prebuilt==0.1.3

0.3.8

cli==0.1.76

sdk==0.1.56

checkpoint==2.0.19

0.3.7

0.3.6

checkpointsqlite==2.0.6

checkpointpostgres==2.0.16

checkpoint==2.0.18

sdk==0.1.55

checkpoint==2.0.17

cli==0.1.75

prebuilt==0.1.2

0.3.5

0.3.4

0.3.3

0.3.2

cli==0.1.74

0.3.1

prebuilt==0.1.1

prebuilt==0.1.0

0.3.0

0.2.76

0.2.75

cli==0.1.73

sdk==0.1.53

cli==0.1.72

0.2.74

checkpoint==2.0.16

0.2.73

checkpointsqlite==2.0.5

checkpointpostgres==2.0.15

checkpoint==2.0.15

checkpoint==2.0.14

0.2.72

checkpoint==2.0.13

checkpointpostgres==2.0.14

checkpointsqlite==2.0.4

0.2.71

checkpoint==2.0.12

cli==0.1.71

0.2.70

0.2.69

cli==0.1.70

0.2.68

cli==0.1.69

0.2.67

0.2.66

0.2.65

cli==0.1.68

0.2.64

0.2.63

checkpointpostgres==2.0.13

checkpointpostgres==2.0.12

checkpointsqlite==2.0.3

checkpoint==2.0.10

checkpointduckdb==2.0.2

checkpointpostgres==2.0.11

checkpointsqlite==2.0.2

cli==0.1.67

cli==0.1.66

0.2.62

checkpointpostgres==2.0.10

sdk==0.1.51

sdk==0.1.50

0.2.61

checkpointpostgres==2.0.9

cli==0.1.65

sdk==0.1.48

0.2.60

sdk==0.1.47

cli==0.1.64

sdk==0.1.46

cli==0.1.63

cli==0.1.62

sdk==0.1.45

checkpoint==2.0.9

sdk==0.1.44

0.2.59

0.2.58

0.2.57

checkpointpostgres==2.0.8

sdk==0.1.43

0.2.56

0.2.55

0.2.54

sdk==0.1.42

checkpoint==2.0.8

checkpointpostgres==2.0.7

checkpointpostgres==2.0.6

cli==0.1.61

cli==0.1.60

checkpointpostgres==2.0.5

sdk==0.1.40

checkpoint==2.0.7

sdk==0.1.39

sdk==0.1.38

sdk==0.1.37

checkpointpostgres==2.0.4

checkpoint==2.0.6

cli==0.1.59

cli==0.1.58

cli==0.1.57

cli==0.1.56

0.2.53

cli==0.1.55

cli==0.1.55rc1

0.2.52

0.2.51

checkpointpostgres==2.0.3

checkpoint==2.0.5

0.2.50

0.2.49

cli==0.1.54

0.2.48

checkpoint==2.0.4

sdk==0.1.36

0.2.47

checkpoint==2.0.3

0.2.46

0.2.45

cli==0.1.53

0.2.44

0.2.43

0.2.42

0.2.41

0.2.40

sdk==0.1.35

sdk==0.1.34

checkpointsqlite==2.0.1

checkpointpostgres==2.0.2

checkpointduckdb==2.0.1

checkpoint==2.0.2

checkpointduckdb==2.0.0

0.2.39

0.2.38

0.2.37

0.2.36

sdk==0.1.33

0.2.35

checkpointpostgres==2.0.1

checkpoint==2.0.1

0.2.34

0.2.33

checkpointpostgres==2.0.0

0.2.32

checkpointsqlite==2.0.0

checkpoint==2.0.0

sdk==0.1.32

checkpointpostgres==1.0.11

0.2.31

checkpointpostgres==1.0.10

0.2.30

checkpoint==1.0.14

0.2.29

checkpoint==1.0.13

checkpointpostgres==1.0.9

checkpoint==1.0.12

0.2.28

0.2.27

0.2.26

checkpointsqlite==1.0.4

checkpointpostgres==1.0.8

checkpoint==1.0.11

0.2.25

sdk==0.1.31

0.2.24

0.2.23

checkpointpostgres==1.0.7

checkpoint==1.0.10

0.2.22

0.2.21

0.2.20

0.2.19

0.2.18

checkpoint==1.0.9

0.2.17

checkpointsqlite==1.0.3

checkpointsqlite==1.0.2

checkpointpostgres==1.0.6

0.2.16

checkpoint==1.0.8

0.2.15

checkpointsqlite==1.0.1

checkpoint==1.0.7

checkpointpostgres==1.0.5

sdk==0.1.30

cli==0.1.52

checkpointpostgres==1.0.4

sdk==0.1.29

0.2.13

cli==0.1.51

0.2.12

checkpoint==1.0.4

0.2.11

0.2.10

0.2.9

sdk==0.1.28

0.2.7

0.2.6

0.2.5

0.2.4

checkpoint==1.0.3

checkpointpostgres==1.0.3

checkpointpostgres==1.0.2

0.2.3

checkpointpostgres==1.0.1

0.2.2

0.2.1

checkpoint==1.0.2

0.2.0

checkpointpostgres==1.0.0

checkpointsqlite==1.0.0

checkpoint==1.0.1

sdk==0.1.27

checkpoint==1.0.0

0.1.17

0.1.16

0.1.15

0.1.14

0.1.13

0.1.12

0.1.11

0.1.10

cli==0.1.50

0.1.9

cli==0.1.49

sdk==0.1.26

0.1.8

0.1.7

0.1.6

cli==0.1.48

0.1.5

cli==0.1.47

0.1.4

0.1.3

cli==0.1.46

cli==0.1.45

cli==0.1.45a1

0.1.2

sdk==0.1.25

sdk==0.1.24

cli==0.1.45a0

cli==0.1.44

cli==0.1.42

cli==0.1.41

cli==0.1.40

sdk==0.1.23

langgraph-cli==0.1.39

v0.0.8

v0.0.7

v0.0.6

v0.0.5

v0.0.4

v0.0.3

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

Subscribe

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: langchain-ai/langgraph#211