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

Validation Errors When Passing Graph State To Tools #250

New Issue
Closed
opened 2026-02-20 17:33:40 -05:00 by yindo · 2 comments
yindo commented 2026-02-20 17:33:40 -05:00
Owner
Copy Link
Copy Source

Originally created by @fletchertyler914 on GitHub (Sep 30, 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 json
from typing import Annotated, Sequence, TypedDict

from langchain_core.tools import tool
from langchain_core.runnables import RunnableConfig
from langchain_core.messages import ToolMessage, SystemMessage, BaseMessage

from langchain_openai import ChatOpenAI
from langgraph.graph import StateGraph, END
from langgraph.prebuilt import InjectedState
from langgraph.graph.message import add_messages


class AgentState(TypedDict):
    """The state of the agent."""
    messages: Annotated[Sequence[BaseMessage], add_messages]


# Define a dummy get_weather tool with no state argument
@tool(parse_docstring=True)
def get_weather(location: str):
    """
    Call to get the weather from a specific location.

    Args:
        location: The location to get the weather from.
    """
    # This is a placeholder for the actual implementation
    # Don't let the LLM know this though 😊
    if any([city in location.lower() for city in ['sf', 'san francisco']]):
        return "It's sunny in San Francisco, but you better look out if you're a Gemini 😈."
    else:
        return f"I am not sure what the weather is in {location}"


# ONLY add the (injected) state argument to this tool to show its broken
@tool(parse_docstring=True)
def get_weather_with_state(
    location: str,
    state: Annotated[dict, InjectedState]
):
    """
    Call to get the BROKEN weather from a specific location. Use this if a user asks for BROKEN weather only, otherwise use get_weather.

    Args:
        location: The location to get the weather from.
    """
    # This is a placeholder for the actual implementation
    # Don't let the LLM know this though 😊
    print('get_weather state: ', state)

    if any([city in location.lower() for city in ['sf', 'san francisco']]):
        return "It's sunny in San Francisco, but you better look out if you're a Gemini 😈."
    else:
        return f"I am not sure what the weather is in {location}"


# Define our tools
tools = [get_weather, get_weather_with_state]
tools_by_name = {tool.name: tool for tool in tools}

# Define our model
model = ChatOpenAI(model="gpt-4o-mini")
model = model.bind_tools(tools)


# Define our tool node
def tool_node(state: AgentState):
    outputs = []
    for tool_call in state['messages'][-1].tool_calls:
        tool_result = tools_by_name[tool_call["name"]].invoke(
            tool_call["args"]
        )
        outputs.append(
            ToolMessage(
                content=json.dumps(tool_result),
                name=tool_call["name"],
                tool_call_id=tool_call["id"],
            )
        )
    return {"messages": outputs}


# Define the node that calls the model
def call_model(
    state: AgentState,
    config: RunnableConfig,
):
    # this is similar to customizing the create_react_agent with state_modifier, but is a lot more flexible
    system_prompt = SystemMessage(
        "You are a helpful AI assistant, please respond to the users query to the best of your ability!")

    response = model.invoke([system_prompt] + state['messages'], config)
    # We return a list, because this will get added to the existing list
    return {"messages": [response]}


# Define the conditional edge that determines whether to continue or not
def should_continue(state: AgentState):
    messages = state["messages"]
    last_message = messages[-1]
    # If there is no function call, then we finish
    if not last_message.tool_calls:
        return "end"
    # Otherwise if there is, we continue
    else:
        return "continue"


# Define a new graph
workflow = StateGraph(AgentState)

# Define the two nodes we will cycle between
workflow.add_node("agent", call_model)
workflow.add_node("tools", tool_node)

# Set the entrypoint as `agent`
# This means that this node is the first one called
workflow.set_entry_point("agent")

# We now add a conditional edge
workflow.add_conditional_edges(
    # First, we define the start node. We use `agent`.
    # This means these are the edges taken after the `agent` node is called.
    "agent",
    # Next, we pass in the function that will determine which node is called next.
    should_continue,
    # Finally we pass in a mapping.
    # The keys are strings, and the values are other nodes.
    # END is a special node marking that the graph should finish.
    # What will happen is we will call `should_continue`, and then the output of that
    # will be matched against the keys in this mapping.
    # Based on which one it matches, that node will then be called.
    {
        # If `tools`, then we call the tool node.
        "continue": "tools",
        # Otherwise we finish.
        "end": END,
    },
)

# We now add a normal edge from `tools` to `agent`.
# This means that after `tools` is called, `agent` node is called next.
workflow.add_edge("tools", "agent")

# Now we can compile and visualize our graph
graph = workflow.compile()


# Helper function for formatting the stream nicely
def print_stream(stream):
    for s in stream:
        message = s["messages"][-1]
        if isinstance(message, tuple):
            print(message)
        else:
            message.pretty_print()


# invoke the stateless (working) tool first to show it works
inputs = {"messages": [("user", "what is the weather in sf")]}
print_stream(graph.stream(inputs, stream_mode="values"))

# invoke the stateful (non-working) tool next to show its broken (pydantic validation errors)
inputs = {"messages": [("user", "what is the BROKEN weather in sf")]}
print_stream(graph.stream(inputs, stream_mode="values"))

Error Message and Stack Trace (if applicable)

---------------------------------------------------------------------------
ValidationError                           Traceback (most recent call last)
Cell In[1], line 166
    164 # invoke the stateful (non-working) tool next to show its broken (pydantic validation errors)
    165 inputs = {"messages": [("user", "what is the BROKEN weather in sf")]}
--> 166 print_stream(graph.stream(inputs, stream_mode="values"))

Cell In[1], line 152
    151 def print_stream(stream):
--> 152     for s in stream:
    153         message = s["messages"][-1]
    154         if isinstance(message, tuple):

File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langgraph/pregel/__init__.py:1298, in Pregel.stream(self, input, config, stream_mode, output_keys, interrupt_before, interrupt_after, debug, subgraphs)
   1287     # Similarly to Bulk Synchronous Parallel / Pregel model
   1288     # computation proceeds in steps, while there are channel updates
   1289     # channel updates from step N are only visible in step N+1
   1290     # channels are guaranteed to be immutable for the duration of the step,
   1291     # with channel updates applied only at the transition between steps
   1292     while loop.tick(
   1293         input_keys=self.input_channels,
   1294         interrupt_before=interrupt_before_,
   1295         interrupt_after=interrupt_after_,
   1296         manager=run_manager,
   1297     ):
-> 1298         for _ in runner.tick(
   1299             loop.tasks.values(),
   1300             timeout=self.step_timeout,
   1301             retry_policy=self.retry_policy,
   1302             get_waiter=get_waiter,
   1303         ):
   1304             # emit output
   1305             yield from output()
   1306 # emit output

File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langgraph/pregel/runner.py:56, in PregelRunner.tick(self, tasks, reraise, timeout, retry_policy, get_waiter)
     54 t = tasks[0]
     55 try:
---> 56     run_with_retry(t, retry_policy)
     57     self.commit(t, None)
     58 except Exception as exc:

File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langgraph/pregel/retry.py:29, in run_with_retry(task, retry_policy)
     27 task.writes.clear()
     28 # run the task
---> 29 task.proc.invoke(task.input, config)
     30 # if successful, end
     31 break

File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langgraph/utils/runnable.py:405, in RunnableSeq.invoke(self, input, config, **kwargs)
    403 context.run(_set_config_context, config)
    404 if i == 0:
--> 405     input = context.run(step.invoke, input, config, **kwargs)
    406 else:
    407     input = context.run(step.invoke, input, config)

File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langgraph/utils/runnable.py:181, in RunnableCallable.invoke(self, input, config, **kwargs)
    179 else:
    180     context.run(_set_config_context, config)
--> 181     ret = context.run(self.func, input, **kwargs)
    182 if isinstance(ret, Runnable) and self.recurse:
    183     return ret.invoke(input, config)

Cell In[1], line 71
     69 outputs = []
     70 for tool_call in state['messages'][-1].tool_calls:
---> 71     tool_result = tools_by_name[tool_call["name"]].invoke(
     72         tool_call["args"]
     73     )
     74     outputs.append(
     75         ToolMessage(
     76             content=json.dumps(tool_result),
   (...)
     79         )
     80     )
     81 return {"messages": outputs}

File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langchain_core/tools/base.py:485, in BaseTool.invoke(self, input, config, **kwargs)
    478 def invoke(
    479     self,
    480     input: Union[str, dict, ToolCall],
    481     config: Optional[RunnableConfig] = None,
    482     **kwargs: Any,
    483 ) -> Any:
    484     tool_input, kwargs = _prep_run_args(input, config, **kwargs)
--> 485     return self.run(tool_input, **kwargs)

File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langchain_core/tools/base.py:688, in BaseTool.run(self, tool_input, verbose, start_color, color, callbacks, tags, metadata, run_name, run_id, config, tool_call_id, **kwargs)
    686 if error_to_raise:
    687     run_manager.on_tool_error(error_to_raise)
--> 688     raise error_to_raise
    689 output = _format_output(content, artifact, tool_call_id, self.name, status)
    690 run_manager.on_tool_end(output, color=color, name=self.name, **kwargs)

File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langchain_core/tools/base.py:651, in BaseTool.run(self, tool_input, verbose, start_color, color, callbacks, tags, metadata, run_name, run_id, config, tool_call_id, **kwargs)
    649 context = copy_context()
    650 context.run(_set_config_context, child_config)
--> 651 tool_args, tool_kwargs = self._to_args_and_kwargs(tool_input)
    652 if signature(self._run).parameters.get("run_manager"):
    653     tool_kwargs["run_manager"] = run_manager

File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langchain_core/tools/base.py:574, in BaseTool._to_args_and_kwargs(self, tool_input)
    573 def _to_args_and_kwargs(self, tool_input: Union[str, dict]) -> tuple[tuple, dict]:
--> 574     tool_input = self._parse_input(tool_input)
    575     # For backwards compatibility, if run_input is a string,
    576     # pass as a positional argument.
    577     if isinstance(tool_input, str):

File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langchain_core/tools/base.py:516, in BaseTool._parse_input(self, tool_input)
    514 if input_args is not None:
    515     if issubclass(input_args, BaseModel):
--> 516         result = input_args.model_validate(tool_input)
    517         result_dict = result.model_dump()
    518     elif issubclass(input_args, BaseModelV1):

File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/pydantic/main.py:596, in BaseModel.model_validate(cls, obj, strict, from_attributes, context)
    594 # `__tracebackhide__` tells pytest and some other tools to omit this function from tracebacks
    595 __tracebackhide__ = True
--> 596 return cls.__pydantic_validator__.validate_python(
    597     obj, strict=strict, from_attributes=from_attributes, context=context
    598 )

ValidationError: 1 validation error for get_weather_with_state
state
  Field required [type=missing, input_value={'location': 'San Francisco'}, input_type=dict]
    For further information visit https://errors.pydantic.dev/2.9/v/missing

Description

When using state: Annotated[dict, InjectedState] in a tool argument and omitting it from the docstring, the state still throws a validation exception.

I copied the example code directly from the create react agent from scratch example, and added a 2nd tool which injects the state. There are no other differences in the tools, and you can verify the schema's with get_input_schema and tool_call_schema properly omit the state key as expected.
Screenshot 2024-09-30 at 14 45 35

However, when the model calls the tool, it throws a pydantic validation exception saying the state key is missing.

System Info

System Information

OS: Darwin
OS Version: Darwin Kernel Version 23.5.0: Wed May 1 20:14:38 PDT 2024; root:xnu-10063.121.3~5/RELEASE_ARM64_T6020
Python Version: 3.11.7 (main, May 6 2024, 13:40:41) [Clang 15.0.0 (clang-1500.3.9.4)]

Package Information

langchain_core: 0.3.6
langsmith: 0.1.129
langchain_openai: 0.2.1
langgraph: 0.2.29

Optional packages not installed

langserve

Other Dependencies

httpx: 0.27.2
jsonpatch: 1.33
langgraph-checkpoint: 1.0.13
openai: 1.50.2
orjson: 3.10.7
packaging: 24.1
pydantic: 2.9.2
PyYAML: 6.0.2
requests: 2.32.3
tenacity: 8.5.0
tiktoken: 0.7.0
typing-extensions: 4.12.2

Originally created by @fletchertyler914 on GitHub (Sep 30, 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 json from typing import Annotated, Sequence, TypedDict from langchain_core.tools import tool from langchain_core.runnables import RunnableConfig from langchain_core.messages import ToolMessage, SystemMessage, BaseMessage from langchain_openai import ChatOpenAI from langgraph.graph import StateGraph, END from langgraph.prebuilt import InjectedState from langgraph.graph.message import add_messages class AgentState(TypedDict): """The state of the agent.""" messages: Annotated[Sequence[BaseMessage], add_messages] # Define a dummy get_weather tool with no state argument @tool(parse_docstring=True) def get_weather(location: str): """ Call to get the weather from a specific location. Args: location: The location to get the weather from. """ # This is a placeholder for the actual implementation # Don't let the LLM know this though 😊 if any([city in location.lower() for city in ['sf', 'san francisco']]): return "It's sunny in San Francisco, but you better look out if you're a Gemini 😈." else: return f"I am not sure what the weather is in {location}" # ONLY add the (injected) state argument to this tool to show its broken @tool(parse_docstring=True) def get_weather_with_state( location: str, state: Annotated[dict, InjectedState] ): """ Call to get the BROKEN weather from a specific location. Use this if a user asks for BROKEN weather only, otherwise use get_weather. Args: location: The location to get the weather from. """ # This is a placeholder for the actual implementation # Don't let the LLM know this though 😊 print('get_weather state: ', state) if any([city in location.lower() for city in ['sf', 'san francisco']]): return "It's sunny in San Francisco, but you better look out if you're a Gemini 😈." else: return f"I am not sure what the weather is in {location}" # Define our tools tools = [get_weather, get_weather_with_state] tools_by_name = {tool.name: tool for tool in tools} # Define our model model = ChatOpenAI(model="gpt-4o-mini") model = model.bind_tools(tools) # Define our tool node def tool_node(state: AgentState): outputs = [] for tool_call in state['messages'][-1].tool_calls: tool_result = tools_by_name[tool_call["name"]].invoke( tool_call["args"] ) outputs.append( ToolMessage( content=json.dumps(tool_result), name=tool_call["name"], tool_call_id=tool_call["id"], ) ) return {"messages": outputs} # Define the node that calls the model def call_model( state: AgentState, config: RunnableConfig, ): # this is similar to customizing the create_react_agent with state_modifier, but is a lot more flexible system_prompt = SystemMessage( "You are a helpful AI assistant, please respond to the users query to the best of your ability!") response = model.invoke([system_prompt] + state['messages'], config) # We return a list, because this will get added to the existing list return {"messages": [response]} # Define the conditional edge that determines whether to continue or not def should_continue(state: AgentState): messages = state["messages"] last_message = messages[-1] # If there is no function call, then we finish if not last_message.tool_calls: return "end" # Otherwise if there is, we continue else: return "continue" # Define a new graph workflow = StateGraph(AgentState) # Define the two nodes we will cycle between workflow.add_node("agent", call_model) workflow.add_node("tools", tool_node) # Set the entrypoint as `agent` # This means that this node is the first one called workflow.set_entry_point("agent") # We now add a conditional edge workflow.add_conditional_edges( # First, we define the start node. We use `agent`. # This means these are the edges taken after the `agent` node is called. "agent", # Next, we pass in the function that will determine which node is called next. should_continue, # Finally we pass in a mapping. # The keys are strings, and the values are other nodes. # END is a special node marking that the graph should finish. # What will happen is we will call `should_continue`, and then the output of that # will be matched against the keys in this mapping. # Based on which one it matches, that node will then be called. { # If `tools`, then we call the tool node. "continue": "tools", # Otherwise we finish. "end": END, }, ) # We now add a normal edge from `tools` to `agent`. # This means that after `tools` is called, `agent` node is called next. workflow.add_edge("tools", "agent") # Now we can compile and visualize our graph graph = workflow.compile() # Helper function for formatting the stream nicely def print_stream(stream): for s in stream: message = s["messages"][-1] if isinstance(message, tuple): print(message) else: message.pretty_print() # invoke the stateless (working) tool first to show it works inputs = {"messages": [("user", "what is the weather in sf")]} print_stream(graph.stream(inputs, stream_mode="values")) # invoke the stateful (non-working) tool next to show its broken (pydantic validation errors) inputs = {"messages": [("user", "what is the BROKEN weather in sf")]} print_stream(graph.stream(inputs, stream_mode="values")) ``` ### Error Message and Stack Trace (if applicable) ```shell --------------------------------------------------------------------------- ValidationError Traceback (most recent call last) Cell In[1], line 166 164 # invoke the stateful (non-working) tool next to show its broken (pydantic validation errors) 165 inputs = {"messages": [("user", "what is the BROKEN weather in sf")]} --> 166 print_stream(graph.stream(inputs, stream_mode="values")) Cell In[1], line 152 151 def print_stream(stream): --> 152 for s in stream: 153 message = s["messages"][-1] 154 if isinstance(message, tuple): File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langgraph/pregel/__init__.py:1298, in Pregel.stream(self, input, config, stream_mode, output_keys, interrupt_before, interrupt_after, debug, subgraphs) 1287 # Similarly to Bulk Synchronous Parallel / Pregel model 1288 # computation proceeds in steps, while there are channel updates 1289 # channel updates from step N are only visible in step N+1 1290 # channels are guaranteed to be immutable for the duration of the step, 1291 # with channel updates applied only at the transition between steps 1292 while loop.tick( 1293 input_keys=self.input_channels, 1294 interrupt_before=interrupt_before_, 1295 interrupt_after=interrupt_after_, 1296 manager=run_manager, 1297 ): -> 1298 for _ in runner.tick( 1299 loop.tasks.values(), 1300 timeout=self.step_timeout, 1301 retry_policy=self.retry_policy, 1302 get_waiter=get_waiter, 1303 ): 1304 # emit output 1305 yield from output() 1306 # emit output File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langgraph/pregel/runner.py:56, in PregelRunner.tick(self, tasks, reraise, timeout, retry_policy, get_waiter) 54 t = tasks[0] 55 try: ---> 56 run_with_retry(t, retry_policy) 57 self.commit(t, None) 58 except Exception as exc: File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langgraph/pregel/retry.py:29, in run_with_retry(task, retry_policy) 27 task.writes.clear() 28 # run the task ---> 29 task.proc.invoke(task.input, config) 30 # if successful, end 31 break File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langgraph/utils/runnable.py:405, in RunnableSeq.invoke(self, input, config, **kwargs) 403 context.run(_set_config_context, config) 404 if i == 0: --> 405 input = context.run(step.invoke, input, config, **kwargs) 406 else: 407 input = context.run(step.invoke, input, config) File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langgraph/utils/runnable.py:181, in RunnableCallable.invoke(self, input, config, **kwargs) 179 else: 180 context.run(_set_config_context, config) --> 181 ret = context.run(self.func, input, **kwargs) 182 if isinstance(ret, Runnable) and self.recurse: 183 return ret.invoke(input, config) Cell In[1], line 71 69 outputs = [] 70 for tool_call in state['messages'][-1].tool_calls: ---> 71 tool_result = tools_by_name[tool_call["name"]].invoke( 72 tool_call["args"] 73 ) 74 outputs.append( 75 ToolMessage( 76 content=json.dumps(tool_result), (...) 79 ) 80 ) 81 return {"messages": outputs} File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langchain_core/tools/base.py:485, in BaseTool.invoke(self, input, config, **kwargs) 478 def invoke( 479 self, 480 input: Union[str, dict, ToolCall], 481 config: Optional[RunnableConfig] = None, 482 **kwargs: Any, 483 ) -> Any: 484 tool_input, kwargs = _prep_run_args(input, config, **kwargs) --> 485 return self.run(tool_input, **kwargs) File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langchain_core/tools/base.py:688, in BaseTool.run(self, tool_input, verbose, start_color, color, callbacks, tags, metadata, run_name, run_id, config, tool_call_id, **kwargs) 686 if error_to_raise: 687 run_manager.on_tool_error(error_to_raise) --> 688 raise error_to_raise 689 output = _format_output(content, artifact, tool_call_id, self.name, status) 690 run_manager.on_tool_end(output, color=color, name=self.name, **kwargs) File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langchain_core/tools/base.py:651, in BaseTool.run(self, tool_input, verbose, start_color, color, callbacks, tags, metadata, run_name, run_id, config, tool_call_id, **kwargs) 649 context = copy_context() 650 context.run(_set_config_context, child_config) --> 651 tool_args, tool_kwargs = self._to_args_and_kwargs(tool_input) 652 if signature(self._run).parameters.get("run_manager"): 653 tool_kwargs["run_manager"] = run_manager File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langchain_core/tools/base.py:574, in BaseTool._to_args_and_kwargs(self, tool_input) 573 def _to_args_and_kwargs(self, tool_input: Union[str, dict]) -> tuple[tuple, dict]: --> 574 tool_input = self._parse_input(tool_input) 575 # For backwards compatibility, if run_input is a string, 576 # pass as a positional argument. 577 if isinstance(tool_input, str): File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/langchain_core/tools/base.py:516, in BaseTool._parse_input(self, tool_input) 514 if input_args is not None: 515 if issubclass(input_args, BaseModel): --> 516 result = input_args.model_validate(tool_input) 517 result_dict = result.model_dump() 518 elif issubclass(input_args, BaseModelV1): File ~/Library/Caches/pypoetry/virtualenvs/ask-harmony-FAOMPRZw-py3.11/lib/python3.11/site-packages/pydantic/main.py:596, in BaseModel.model_validate(cls, obj, strict, from_attributes, context) 594 # `__tracebackhide__` tells pytest and some other tools to omit this function from tracebacks 595 __tracebackhide__ = True --> 596 return cls.__pydantic_validator__.validate_python( 597 obj, strict=strict, from_attributes=from_attributes, context=context 598 ) ValidationError: 1 validation error for get_weather_with_state state Field required [type=missing, input_value={'location': 'San Francisco'}, input_type=dict] For further information visit https://errors.pydantic.dev/2.9/v/missing ``` ### Description When using `state: Annotated[dict, InjectedState]` in a tool argument and omitting it from the docstring, the `state` still throws a validation exception. I copied the example code directly from the [create react agent from scratch](https://langchain-ai.github.io/langgraph/how-tos/react-agent-from-scratch) example, and added a 2nd tool which injects the state. There are no other differences in the tools, and you can verify the schema's with `get_input_schema` and `tool_call_schema` properly omit the `state` key as expected. <img width="1168" alt="Screenshot 2024-09-30 at 14 45 35" src="https://github.com/user-attachments/assets/931f0ca2-20ed-4a12-a018-2f74bbd6628b"> However, when the model calls the tool, it throws a pydantic validation exception saying the `state` key is missing. ### System Info System Information ------------------ > OS: Darwin > OS Version: Darwin Kernel Version 23.5.0: Wed May 1 20:14:38 PDT 2024; root:xnu-10063.121.3~5/RELEASE_ARM64_T6020 > Python Version: 3.11.7 (main, May 6 2024, 13:40:41) [Clang 15.0.0 (clang-1500.3.9.4)] Package Information ------------------- > langchain_core: 0.3.6 > langsmith: 0.1.129 > langchain_openai: 0.2.1 > langgraph: 0.2.29 Optional packages not installed ------------------------------- > langserve Other Dependencies ------------------ > httpx: 0.27.2 > jsonpatch: 1.33 > langgraph-checkpoint: 1.0.13 > openai: 1.50.2 > orjson: 3.10.7 > packaging: 24.1 > pydantic: 2.9.2 > PyYAML: 6.0.2 > requests: 2.32.3 > tenacity: 8.5.0 > tiktoken: 0.7.0 > typing-extensions: 4.12.2
yindo closed this issue 2026-02-20 17:33:43 -05:00
yindo commented 2026-02-20 17:33:44 -05:00
Author
Owner
Copy Link
Copy Source

@fletchertyler914 commented on GitHub (Sep 30, 2024):

Ah, this was my mistake. I didn't realize the prebuilt ToolNode was doing this behind the scenes!

@fletchertyler914 commented on GitHub (Sep 30, 2024): Ah, this was my mistake. I didn't realize the prebuilt `ToolNode` was doing this behind the scenes!
yindo commented 2026-02-20 17:33:46 -05:00
Author
Owner
Copy Link
Copy Source

@mezeirobert0 commented on GitHub (Nov 5, 2025):

So how did you solve the problem?

@mezeirobert0 commented on GitHub (Nov 5, 2025): So how did you solve the problem?
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#250
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?