github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-09-11 16:01:33 +00:00
Code Issues Packages Projects Releases Wiki Activity

chore(langchain): rename create_react_agent -> create_agent (#32789)

Browse Source
This commit is contained in:
Sydney Runkle
2025-09-02 12:13:12 -04:00
committed by GitHub
parent 238ecd09e0
commit dc9f941326
8 changed files with 90 additions and 98 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
libs/langchain_v1/langchain/agents/__init__.py
View File

@@ -1,10 +1,10 @@
"""langgraph.prebuilt exposes a higher-level API for creating and executing agents and tools.""" """langgraph.prebuilt exposes a higher-level API for creating and executing agents and tools."""
from langchain.agents.react_agent import AgentState, create_react_agent from langchain.agents.react_agent import AgentState, create_agent
from langchain.agents.tool_node import ToolNode from langchain.agents.tool_node import ToolNode
__all__ = [ __all__ = [
"AgentState", "AgentState",
"ToolNode", "ToolNode",
"create_react_agent", "create_agent",
] ]

12
libs/langchain_v1/langchain/agents/react_agent.py
View File

@@ -898,7 +898,7 @@ def _supports_native_structured_output(
) )
def create_react_agent( # noqa: D417 def create_agent( # noqa: D417
model: Union[ model: Union[
str, str,
BaseChatModel, BaseChatModel,
@@ -928,7 +928,7 @@ def create_react_agent( # noqa: D417
) -> CompiledStateGraph[StateT, ContextT]: ) -> CompiledStateGraph[StateT, ContextT]:
"""Creates an agent graph that calls tools in a loop until a stopping condition is met. """Creates an agent graph that calls tools in a loop until a stopping condition is met.
For more details on using `create_react_agent`, visit [Agents](https://langchain-ai.github.io/langgraph/agents/overview/) documentation. For more details on using `create_agent`, visit [Agents](https://langchain-ai.github.io/langgraph/agents/overview/) documentation.
Args: Args:
model: The language model for the agent. Supports static and dynamic model: The language model for the agent. Supports static and dynamic
@@ -1096,13 +1096,13 @@ def create_react_agent( # noqa: D417
Example: Example:
```python ```python
from langchain.agents import create_react_agent from langchain.agents import create_agent
def check_weather(location: str) -> str: def check_weather(location: str) -> str:
'''Return the weather forecast for the specified location.''' '''Return the weather forecast for the specified location.'''
return f"It's always sunny in {location}" return f"It's always sunny in {location}"
graph = create_react_agent( graph = create_agent(
"anthropic:claude-3-7-sonnet-latest", "anthropic:claude-3-7-sonnet-latest",
tools=[check_weather], tools=[check_weather],
prompt="You are a helpful assistant", prompt="You are a helpful assistant",
@@ -1123,7 +1123,7 @@ def create_react_agent( # noqa: D417
context_schema = config_schema context_schema = config_schema
if len(deprecated_kwargs) > 0: if len(deprecated_kwargs) > 0:
msg = f"create_react_agent() got unexpected keyword arguments: {deprecated_kwargs}" msg = f"create_agent() got unexpected keyword arguments: {deprecated_kwargs}"
raise TypeError(msg) raise TypeError(msg)
if response_format and not isinstance(response_format, (ToolStrategy, ProviderStrategy)): if response_format and not isinstance(response_format, (ToolStrategy, ProviderStrategy)):
@@ -1171,5 +1171,5 @@ __all__ = [
"AgentStatePydantic", "AgentStatePydantic",
"AgentStateWithStructuredResponse", "AgentStateWithStructuredResponse",
"AgentStateWithStructuredResponsePydantic", "AgentStateWithStructuredResponsePydantic",
"create_react_agent", "create_agent",
] ]

6
libs/langchain_v1/tests/integration_tests/agents/test_response_format.py
View File

@@ -2,7 +2,7 @@ import pytest
from langchain_core.messages import HumanMessage from langchain_core.messages import HumanMessage
from pydantic import BaseModel, Field from pydantic import BaseModel, Field
from langchain.agents import create_react_agent from langchain.agents import create_agent
from langchain.agents.structured_output import ToolStrategy from langchain.agents.structured_output import ToolStrategy
@@ -24,7 +24,7 @@ def test_inference_to_native_output() -> None:
from langchain_openai import ChatOpenAI from langchain_openai import ChatOpenAI
model = ChatOpenAI(model="gpt-5") model = ChatOpenAI(model="gpt-5")
agent = create_react_agent( agent = create_agent(
model, model,
prompt=( prompt=(
"You are a helpful weather assistant. Please call the get_weather tool, " "You are a helpful weather assistant. Please call the get_weather tool, "
@@ -54,7 +54,7 @@ def test_inference_to_tool_output() -> None:
from langchain_openai import ChatOpenAI from langchain_openai import ChatOpenAI
model = ChatOpenAI(model="gpt-4") model = ChatOpenAI(model="gpt-4")
agent = create_react_agent( agent = create_agent(
model, model,
prompt=( prompt=(
"You are a helpful weather assistant. Please call the get_weather tool, " "You are a helpful weather assistant. Please call the get_weather tool, "

100
libs/langchain_v1/tests/unit_tests/agents/test_react_agent.py
View File

@@ -32,7 +32,7 @@ from typing_extensions import TypedDict
from langchain.agents import ( from langchain.agents import (
AgentState, AgentState,
ToolNode, ToolNode,
create_react_agent, create_agent,
) )
from langchain.agents.react_agent import _validate_chat_history from langchain.agents.react_agent import _validate_chat_history
from langchain.agents.tool_node import ( from langchain.agents.tool_node import (
@@ -52,7 +52,7 @@ pytestmark = pytest.mark.anyio
def test_no_prompt(sync_checkpointer: BaseCheckpointSaver) -> None: def test_no_prompt(sync_checkpointer: BaseCheckpointSaver) -> None:
model = FakeToolCallingModel() model = FakeToolCallingModel()
agent = create_react_agent( agent = create_agent(
model, model,
[], [],
checkpointer=sync_checkpointer, checkpointer=sync_checkpointer,
@@ -82,7 +82,7 @@ def test_no_prompt(sync_checkpointer: BaseCheckpointSaver) -> None:
async def test_no_prompt_async(async_checkpointer: BaseCheckpointSaver) -> None: async def test_no_prompt_async(async_checkpointer: BaseCheckpointSaver) -> None:
model = FakeToolCallingModel() model = FakeToolCallingModel()
agent = create_react_agent(model, [], checkpointer=async_checkpointer) agent = create_agent(model, [], checkpointer=async_checkpointer)
inputs = [HumanMessage("hi?")] inputs = [HumanMessage("hi?")]
thread = {"configurable": {"thread_id": "123"}} thread = {"configurable": {"thread_id": "123"}}
response = await agent.ainvoke({"messages": inputs}, thread, debug=True) response = await agent.ainvoke({"messages": inputs}, thread, debug=True)
@@ -107,7 +107,7 @@ async def test_no_prompt_async(async_checkpointer: BaseCheckpointSaver) -> None:
def test_system_message_prompt() -> None: def test_system_message_prompt() -> None:
prompt = SystemMessage(content="Foo") prompt = SystemMessage(content="Foo")
agent = create_react_agent(FakeToolCallingModel(), [], prompt=prompt) agent = create_agent(FakeToolCallingModel(), [], prompt=prompt)
inputs = [HumanMessage("hi?")] inputs = [HumanMessage("hi?")]
response = agent.invoke({"messages": inputs}) response = agent.invoke({"messages": inputs})
expected_response = {"messages": [*inputs, AIMessage(content="Foo-hi?", id="0", tool_calls=[])]} expected_response = {"messages": [*inputs, AIMessage(content="Foo-hi?", id="0", tool_calls=[])]}
@@ -116,7 +116,7 @@ def test_system_message_prompt() -> None:
def test_string_prompt() -> None: def test_string_prompt() -> None:
prompt = "Foo" prompt = "Foo"
agent = create_react_agent(FakeToolCallingModel(), [], prompt=prompt) agent = create_agent(FakeToolCallingModel(), [], prompt=prompt)
inputs = [HumanMessage("hi?")] inputs = [HumanMessage("hi?")]
response = agent.invoke({"messages": inputs}) response = agent.invoke({"messages": inputs})
expected_response = {"messages": [*inputs, AIMessage(content="Foo-hi?", id="0", tool_calls=[])]} expected_response = {"messages": [*inputs, AIMessage(content="Foo-hi?", id="0", tool_calls=[])]}
@@ -128,7 +128,7 @@ def test_callable_prompt() -> None:
modified_message = f"Bar {state['messages'][-1].content}" modified_message = f"Bar {state['messages'][-1].content}"
return [HumanMessage(content=modified_message)] return [HumanMessage(content=modified_message)]
agent = create_react_agent(FakeToolCallingModel(), [], prompt=prompt) agent = create_agent(FakeToolCallingModel(), [], prompt=prompt)
inputs = [HumanMessage("hi?")] inputs = [HumanMessage("hi?")]
response = agent.invoke({"messages": inputs}) response = agent.invoke({"messages": inputs})
expected_response = {"messages": [*inputs, AIMessage(content="Bar hi?", id="0")]} expected_response = {"messages": [*inputs, AIMessage(content="Bar hi?", id="0")]}
@@ -140,7 +140,7 @@ async def test_callable_prompt_async() -> None:
modified_message = f"Bar {state['messages'][-1].content}" modified_message = f"Bar {state['messages'][-1].content}"
return [HumanMessage(content=modified_message)] return [HumanMessage(content=modified_message)]
agent = create_react_agent(FakeToolCallingModel(), [], prompt=prompt) agent = create_agent(FakeToolCallingModel(), [], prompt=prompt)
inputs = [HumanMessage("hi?")] inputs = [HumanMessage("hi?")]
response = await agent.ainvoke({"messages": inputs}) response = await agent.ainvoke({"messages": inputs})
expected_response = {"messages": [*inputs, AIMessage(content="Bar hi?", id="0")]} expected_response = {"messages": [*inputs, AIMessage(content="Bar hi?", id="0")]}
@@ -152,7 +152,7 @@ def test_runnable_prompt() -> None:
lambda state: [HumanMessage(content=f"Baz {state['messages'][-1].content}")] lambda state: [HumanMessage(content=f"Baz {state['messages'][-1].content}")]
) )
agent = create_react_agent(FakeToolCallingModel(), [], prompt=prompt) agent = create_agent(FakeToolCallingModel(), [], prompt=prompt)
inputs = [HumanMessage("hi?")] inputs = [HumanMessage("hi?")]
response = agent.invoke({"messages": inputs}) response = agent.invoke({"messages": inputs})
expected_response = {"messages": [*inputs, AIMessage(content="Baz hi?", id="0")]} expected_response = {"messages": [*inputs, AIMessage(content="Baz hi?", id="0")]}
@@ -179,7 +179,7 @@ def test_prompt_with_store() -> None:
model = FakeToolCallingModel() model = FakeToolCallingModel()
# test state modifier that uses store works # test state modifier that uses store works
agent = create_react_agent( agent = create_agent(
model, model,
[add], [add],
prompt=prompt, prompt=prompt,
@@ -189,7 +189,7 @@ def test_prompt_with_store() -> None:
assert response["messages"][-1].content == "User name is Alice-hi" assert response["messages"][-1].content == "User name is Alice-hi"
# test state modifier that doesn't use store works # test state modifier that doesn't use store works
agent = create_react_agent( agent = create_agent(
model, model,
[add], [add],
prompt=prompt_no_store, prompt=prompt_no_store,
@@ -219,14 +219,14 @@ async def test_prompt_with_store_async() -> None:
model = FakeToolCallingModel() model = FakeToolCallingModel()
# test state modifier that uses store works # test state modifier that uses store works
agent = create_react_agent(model, [add], prompt=prompt, store=in_memory_store) agent = create_agent(model, [add], prompt=prompt, store=in_memory_store)
response = await agent.ainvoke( response = await agent.ainvoke(
{"messages": [("user", "hi")]}, {"configurable": {"user_id": "1"}} {"messages": [("user", "hi")]}, {"configurable": {"user_id": "1"}}
) )
assert response["messages"][-1].content == "User name is Alice-hi" assert response["messages"][-1].content == "User name is Alice-hi"
# test state modifier that doesn't use store works # test state modifier that doesn't use store works
agent = create_react_agent(model, [add], prompt=prompt_no_store, store=in_memory_store) agent = create_agent(model, [add], prompt=prompt_no_store, store=in_memory_store)
response = await agent.ainvoke( response = await agent.ainvoke(
{"messages": [("user", "hi")]}, {"configurable": {"user_id": "2"}} {"messages": [("user", "hi")]}, {"configurable": {"user_id": "2"}}
) )
@@ -267,7 +267,7 @@ def test_model_with_tools(tool_style: str, include_builtin: bool) -> None:
) )
# check valid agent constructor # check valid agent constructor
with pytest.raises(ValueError): with pytest.raises(ValueError):
create_react_agent( create_agent(
model.bind_tools(tools), model.bind_tools(tools),
tools, tools,
) )
@@ -414,7 +414,7 @@ def test_react_agent_with_structured_response() -> None:
model = FakeToolCallingModel[WeatherResponse]( model = FakeToolCallingModel[WeatherResponse](
tool_calls=tool_calls, structured_response=expected_structured_response tool_calls=tool_calls, structured_response=expected_structured_response
) )
agent = create_react_agent( agent = create_agent(
model, model,
[get_weather], [get_weather],
response_format=WeatherResponse, response_format=WeatherResponse,
@@ -472,7 +472,7 @@ def test_react_agent_update_state(
tool_calls = [[{"args": {}, "id": "1", "name": "get_user_name"}]] tool_calls = [[{"args": {}, "id": "1", "name": "get_user_name"}]]
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, model,
[get_user_name], [get_user_name],
state_schema=CustomState, state_schema=CustomState,
@@ -523,7 +523,7 @@ def test_react_agent_parallel_tool_calls(
[], [],
] ]
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, model,
[human_assistance, get_weather], [human_assistance, get_weather],
checkpointer=sync_checkpointer, checkpointer=sync_checkpointer,
@@ -561,7 +561,7 @@ class AgentStateExtraKey(AgentState):
foo: int foo: int
def test_create_react_agent_inject_vars() -> None: def test_create_agent_inject_vars() -> None:
"""Test that the agent can inject state and store into tool functions.""" """Test that the agent can inject state and store into tool functions."""
store = InMemoryStore() store = InMemoryStore()
namespace = ("test",) namespace = ("test",)
@@ -583,7 +583,7 @@ def test_create_react_agent_inject_vars() -> None:
"type": "tool_call", "type": "tool_call",
} }
model = FakeToolCallingModel(tool_calls=[[tool_call], []]) model = FakeToolCallingModel(tool_calls=[[tool_call], []])
agent = create_react_agent( agent = create_agent(
model, model,
ToolNode([tool1], handle_tool_errors=False), ToolNode([tool1], handle_tool_errors=False),
state_schema=AgentStateExtraKey, state_schema=AgentStateExtraKey,
@@ -623,7 +623,7 @@ async def test_return_direct() -> None:
tool_calls=first_tool_call, tool_calls=first_tool_call,
) )
model = FakeToolCallingModel(tool_calls=[first_tool_call, []]) model = FakeToolCallingModel(tool_calls=[first_tool_call, []])
agent = create_react_agent( agent = create_agent(
model, model,
[tool_return_direct, tool_normal], [tool_return_direct, tool_normal],
) )
@@ -648,7 +648,7 @@ async def test_return_direct() -> None:
), ),
] ]
model = FakeToolCallingModel(tool_calls=[second_tool_call, []]) model = FakeToolCallingModel(tool_calls=[second_tool_call, []])
agent = create_react_agent(model, [tool_return_direct, tool_normal]) agent = create_agent(model, [tool_return_direct, tool_normal])
result = agent.invoke({"messages": [HumanMessage(content="Test normal", id="hum1")]}) result = agent.invoke({"messages": [HumanMessage(content="Test normal", id="hum1")]})
assert result["messages"] == [ assert result["messages"] == [
HumanMessage(content="Test normal", id="hum1"), HumanMessage(content="Test normal", id="hum1"),
@@ -675,7 +675,7 @@ async def test_return_direct() -> None:
), ),
] ]
model = FakeToolCallingModel(tool_calls=[both_tool_calls, []]) model = FakeToolCallingModel(tool_calls=[both_tool_calls, []])
agent = create_react_agent(model, [tool_return_direct, tool_normal]) agent = create_agent(model, [tool_return_direct, tool_normal])
result = agent.invoke({"messages": [HumanMessage(content="Test both", id="hum2")]}) result = agent.invoke({"messages": [HumanMessage(content="Test both", id="hum2")]})
assert result["messages"] == [ assert result["messages"] == [
HumanMessage(content="Test both", id="hum2"), HumanMessage(content="Test both", id="hum2"),
@@ -712,7 +712,7 @@ def test__get_state_args() -> None:
def test_inspect_react() -> None: def test_inspect_react() -> None:
model = FakeToolCallingModel(tool_calls=[]) model = FakeToolCallingModel(tool_calls=[])
agent = create_react_agent(model, []) agent = create_agent(model, [])
inspect.getclosurevars(agent.nodes["agent"].bound.func) inspect.getclosurevars(agent.nodes["agent"].bound.func)
@@ -766,7 +766,7 @@ def test_react_with_subgraph_tools(
] ]
) )
tool_node = ToolNode([addition, multiplication], handle_tool_errors=False) tool_node = ToolNode([addition, multiplication], handle_tool_errors=False)
agent = create_react_agent( agent = create_agent(
model, model,
tool_node, tool_node,
checkpointer=sync_checkpointer, checkpointer=sync_checkpointer,
@@ -812,7 +812,7 @@ def test_react_agent_subgraph_streaming_sync() -> None:
] ]
) )
agent = create_react_agent( agent = create_agent(
model, model,
tools=[get_weather], tools=[get_weather],
prompt="You are a helpful travel assistant.", prompt="You are a helpful travel assistant.",
@@ -899,7 +899,7 @@ async def test_react_agent_subgraph_streaming() -> None:
] ]
) )
agent = create_react_agent( agent = create_agent(
model, model,
tools=[get_weather], tools=[get_weather],
prompt="You are a helpful travel assistant.", prompt="You are a helpful travel assistant.",
@@ -994,7 +994,7 @@ def test_tool_node_node_interrupt(
] ]
) )
config = {"configurable": {"thread_id": "1"}} config = {"configurable": {"thread_id": "1"}}
agent = create_react_agent( agent = create_agent(
model, model,
[tool_interrupt, tool_normal], [tool_interrupt, tool_normal],
checkpointer=sync_checkpointer, checkpointer=sync_checkpointer,
@@ -1045,7 +1045,7 @@ def test_dynamic_model_basic() -> None:
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
agent = create_react_agent(dynamic_model, []) agent = create_agent(dynamic_model, [])
result = agent.invoke({"messages": [HumanMessage("hello")]}) result = agent.invoke({"messages": [HumanMessage("hello")]})
assert len(result["messages"]) == 2 assert len(result["messages"]) == 2
@@ -1082,7 +1082,7 @@ def test_dynamic_model_with_tools() -> None:
tool_calls=[[{"args": {"x": 1}, "id": "1", "name": "basic_tool"}], []] tool_calls=[[{"args": {"x": 1}, "id": "1", "name": "basic_tool"}], []]
) )
agent = create_react_agent(dynamic_model, [basic_tool, advanced_tool]) agent = create_agent(dynamic_model, [basic_tool, advanced_tool])
# Test basic tool usage # Test basic tool usage
result = agent.invoke({"messages": [HumanMessage("basic request")]}) result = agent.invoke({"messages": [HumanMessage("basic request")]})
@@ -1114,7 +1114,7 @@ def test_dynamic_model_with_context() -> None:
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
agent = create_react_agent(dynamic_model, [], context_schema=Context) agent = create_agent(dynamic_model, [], context_schema=Context)
# Test with basic user # Test with basic user
result = agent.invoke( result = agent.invoke(
@@ -1143,7 +1143,7 @@ def test_dynamic_model_with_state_schema() -> None:
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
agent = create_react_agent(dynamic_model, [], state_schema=CustomDynamicState) agent = create_agent(dynamic_model, [], state_schema=CustomDynamicState)
result = agent.invoke({"messages": [HumanMessage("hello")], "model_preference": "advanced"}) result = agent.invoke({"messages": [HumanMessage("hello")], "model_preference": "advanced"})
assert len(result["messages"]) == 2 assert len(result["messages"]) == 2
@@ -1157,7 +1157,7 @@ def test_dynamic_model_with_prompt() -> None:
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
# Test with string prompt # Test with string prompt
agent = create_react_agent(dynamic_model, [], prompt="system_msg") agent = create_agent(dynamic_model, [], prompt="system_msg")
result = agent.invoke({"messages": [HumanMessage("human_msg")]}) result = agent.invoke({"messages": [HumanMessage("human_msg")]})
assert result["messages"][-1].content == "system_msg-human_msg" assert result["messages"][-1].content == "system_msg-human_msg"
@@ -1166,7 +1166,7 @@ def test_dynamic_model_with_prompt() -> None:
"""Generate a dynamic system message based on state.""" """Generate a dynamic system message based on state."""
return [{"role": "system", "content": "system_msg"}, *list(state["messages"])] return [{"role": "system", "content": "system_msg"}, *list(state["messages"])]
agent = create_react_agent(dynamic_model, [], prompt=dynamic_prompt) agent = create_agent(dynamic_model, [], prompt=dynamic_prompt)
result = agent.invoke({"messages": [HumanMessage("human_msg")]}) result = agent.invoke({"messages": [HumanMessage("human_msg")]})
assert result["messages"][-1].content == "system_msg-human_msg" assert result["messages"][-1].content == "system_msg-human_msg"
@@ -1177,7 +1177,7 @@ async def test_dynamic_model_async() -> None:
def dynamic_model(state: AgentState, runtime: Runtime) -> BaseChatModel: def dynamic_model(state: AgentState, runtime: Runtime) -> BaseChatModel:
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
agent = create_react_agent(dynamic_model, []) agent = create_agent(dynamic_model, [])
result = await agent.ainvoke({"messages": [HumanMessage("hello async")]}) result = await agent.ainvoke({"messages": [HumanMessage("hello async")]})
assert len(result["messages"]) == 2 assert len(result["messages"]) == 2
@@ -1205,7 +1205,7 @@ def test_dynamic_model_with_structured_response() -> None:
], ],
) )
agent = create_react_agent(dynamic_model, [], response_format=TestResponse) agent = create_agent(dynamic_model, [], response_format=TestResponse)
result = agent.invoke({"messages": [HumanMessage("hello")]}) result = agent.invoke({"messages": [HumanMessage("hello")]})
assert "structured_response" in result assert "structured_response" in result
@@ -1229,7 +1229,7 @@ def test_dynamic_model_with_checkpointer(sync_checkpointer) -> None:
index=call_count, index=call_count,
) )
agent = create_react_agent(dynamic_model, [], checkpointer=sync_checkpointer) agent = create_agent(dynamic_model, [], checkpointer=sync_checkpointer)
config = {"configurable": {"thread_id": "test_dynamic"}} config = {"configurable": {"thread_id": "test_dynamic"}}
# First call # First call
@@ -1267,7 +1267,7 @@ def test_dynamic_model_state_dependent_tools() -> None:
tool_calls=[[{"args": {"x": 1}, "id": "1", "name": "tool_a"}], []] tool_calls=[[{"args": {"x": 1}, "id": "1", "name": "tool_a"}], []]
) )
agent = create_react_agent(dynamic_model, [tool_a, tool_b]) agent = create_agent(dynamic_model, [tool_a, tool_b])
# Ask to use tool B # Ask to use tool B
result = agent.invoke({"messages": [HumanMessage("use_b please")]}) result = agent.invoke({"messages": [HumanMessage("use_b please")]})
@@ -1291,7 +1291,7 @@ def test_dynamic_model_error_handling() -> None:
raise ValueError(msg) raise ValueError(msg)
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
agent = create_react_agent(failing_dynamic_model, []) agent = create_agent(failing_dynamic_model, [])
# Normal operation should work # Normal operation should work
result = agent.invoke({"messages": [HumanMessage("hello")]}) result = agent.invoke({"messages": [HumanMessage("hello")]})
@@ -1306,13 +1306,13 @@ def test_dynamic_model_vs_static_model_behavior() -> None:
"""Test that dynamic and static models produce equivalent results when configured the same.""" """Test that dynamic and static models produce equivalent results when configured the same."""
# Static model # Static model
static_model = FakeToolCallingModel(tool_calls=[]) static_model = FakeToolCallingModel(tool_calls=[])
static_agent = create_react_agent(static_model, []) static_agent = create_agent(static_model, [])
# Dynamic model returning the same model # Dynamic model returning the same model
def dynamic_model(state, runtime: Runtime): def dynamic_model(state, runtime: Runtime):
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
dynamic_agent = create_react_agent(dynamic_model, []) dynamic_agent = create_agent(dynamic_model, [])
input_msg = {"messages": [HumanMessage("test message")]} input_msg = {"messages": [HumanMessage("test message")]}
@@ -1337,7 +1337,7 @@ def test_dynamic_model_receives_correct_state() -> None:
received_states.append(state) received_states.append(state)
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
agent = create_react_agent(dynamic_model, [], state_schema=CustomAgentState) agent = create_agent(dynamic_model, [], state_schema=CustomAgentState)
# Test with initial state # Test with initial state
input_state = {"messages": [HumanMessage("hello")], "custom_field": "test_value"} input_state = {"messages": [HumanMessage("hello")], "custom_field": "test_value"}
@@ -1368,7 +1368,7 @@ async def test_dynamic_model_receives_correct_state_async() -> None:
received_states.append(state) received_states.append(state)
return FakeToolCallingModel(tool_calls=[]) return FakeToolCallingModel(tool_calls=[])
agent = create_react_agent(dynamic_model, [], state_schema=CustomAgentStateAsync) agent = create_agent(dynamic_model, [], state_schema=CustomAgentStateAsync)
# Test with initial state # Test with initial state
input_state = { input_state = {
@@ -1397,7 +1397,7 @@ def test_pre_model_hook() -> None:
def pre_model_hook(state: AgentState): def pre_model_hook(state: AgentState):
return {"llm_input_messages": [HumanMessage("Hello!")]} return {"llm_input_messages": [HumanMessage("Hello!")]}
agent = create_react_agent(model, [], pre_model_hook=pre_model_hook) agent = create_agent(model, [], pre_model_hook=pre_model_hook)
assert "pre_model_hook" in agent.nodes assert "pre_model_hook" in agent.nodes
result = agent.invoke({"messages": [HumanMessage("hi?")]}) result = agent.invoke({"messages": [HumanMessage("hi?")]})
assert result == { assert result == {
@@ -1411,7 +1411,7 @@ def test_pre_model_hook() -> None:
def pre_model_hook(state: AgentState): def pre_model_hook(state: AgentState):
return {"messages": [RemoveMessage(id=REMOVE_ALL_MESSAGES), HumanMessage("Hello!")]} return {"messages": [RemoveMessage(id=REMOVE_ALL_MESSAGES), HumanMessage("Hello!")]}
agent = create_react_agent(model, [], pre_model_hook=pre_model_hook) agent = create_agent(model, [], pre_model_hook=pre_model_hook)
result = agent.invoke({"messages": [HumanMessage("hi?")]}) result = agent.invoke({"messages": [HumanMessage("hi?")]})
assert result == { assert result == {
"messages": [ "messages": [
@@ -1430,9 +1430,7 @@ def test_post_model_hook() -> None:
def post_model_hook(state: FlagState) -> dict[str, bool]: def post_model_hook(state: FlagState) -> dict[str, bool]:
return {"flag": True} return {"flag": True}
pmh_agent = create_react_agent( pmh_agent = create_agent(model, [], post_model_hook=post_model_hook, state_schema=FlagState)
model, [], post_model_hook=post_model_hook, state_schema=FlagState
)
assert "post_model_hook" in pmh_agent.nodes assert "post_model_hook" in pmh_agent.nodes
@@ -1480,7 +1478,7 @@ def test_post_model_hook_with_structured_output() -> None:
return {"flag": True} return {"flag": True}
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, model,
[get_weather], [get_weather],
response_format=WeatherResponse, response_format=WeatherResponse,
@@ -1496,7 +1494,7 @@ def test_post_model_hook_with_structured_output() -> None:
# Reset the state of the model # Reset the state of the model
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, model,
[get_weather], [get_weather],
response_format=WeatherResponse, response_format=WeatherResponse,
@@ -1568,7 +1566,7 @@ def test_post_model_hook_with_structured_output() -> None:
] ]
def test_create_react_agent_inject_vars_with_post_model_hook() -> None: def test_create_agent_inject_vars_with_post_model_hook() -> None:
store = InMemoryStore() store = InMemoryStore()
namespace = ("test",) namespace = ("test",)
store.put(namespace, "test_key", {"bar": 3}) store.put(namespace, "test_key", {"bar": 3})
@@ -1594,7 +1592,7 @@ def test_create_react_agent_inject_vars_with_post_model_hook() -> None:
return {"foo": 2} return {"foo": 2}
model = FakeToolCallingModel(tool_calls=[[tool_call], []]) model = FakeToolCallingModel(tool_calls=[[tool_call], []])
agent = create_react_agent( agent = create_agent(
model, model,
ToolNode([tool1], handle_tool_errors=False), ToolNode([tool1], handle_tool_errors=False),
state_schema=AgentStateExtraKey, state_schema=AgentStateExtraKey,
@@ -1629,7 +1627,7 @@ def test_response_format_using_tool_choice() -> None:
expected_structured_response = WeatherResponse(temperature=75) expected_structured_response = WeatherResponse(temperature=75)
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, model,
[get_weather], [get_weather],
response_format=WeatherResponse, response_format=WeatherResponse,

4
libs/langchain_v1/tests/unit_tests/agents/test_react_agent_graph.py
View File

@@ -5,7 +5,7 @@ import pytest
from pydantic import BaseModel from pydantic import BaseModel
from syrupy.assertion import SnapshotAssertion from syrupy.assertion import SnapshotAssertion
from langchain.agents import create_react_agent from langchain.agents import create_agent
from .model import FakeToolCallingModel from .model import FakeToolCallingModel
@@ -39,7 +39,7 @@ def test_react_agent_graph_structure(
pre_model_hook: Union[Callable, None], pre_model_hook: Union[Callable, None],
post_model_hook: Union[Callable, None], post_model_hook: Union[Callable, None],
) -> None: ) -> None:
agent = create_react_agent( agent = create_agent(
model, model,
tools=tools, tools=tools,
pre_model_hook=pre_model_hook, pre_model_hook=pre_model_hook,

56
libs/langchain_v1/tests/unit_tests/agents/test_response_format.py
View File

@@ -1,4 +1,4 @@
"""Test suite for create_react_agent with structured output response_format permutations.""" """Test suite for create_agent with structured output response_format permutations."""
import pytest import pytest
@@ -6,7 +6,7 @@ from dataclasses import dataclass
from typing import Union from typing import Union
from langchain_core.messages import HumanMessage from langchain_core.messages import HumanMessage
from langchain.agents import create_react_agent from langchain.agents import create_agent
from langchain.agents.structured_output import ( from langchain.agents.structured_output import (
MultipleStructuredOutputsError, MultipleStructuredOutputsError,
ProviderStrategy, ProviderStrategy,
@@ -114,7 +114,7 @@ class TestResponseFormatAsModel:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent(model, [get_weather], response_format=WeatherBaseModel) agent = create_agent(model, [get_weather], response_format=WeatherBaseModel)
response = agent.invoke({"messages": [HumanMessage("What's the weather?")]}) response = agent.invoke({"messages": [HumanMessage("What's the weather?")]})
assert response["structured_response"] == EXPECTED_WEATHER_PYDANTIC assert response["structured_response"] == EXPECTED_WEATHER_PYDANTIC
@@ -135,7 +135,7 @@ class TestResponseFormatAsModel:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent(model, [get_weather], response_format=WeatherDataclass) agent = create_agent(model, [get_weather], response_format=WeatherDataclass)
response = agent.invoke({"messages": [HumanMessage("What's the weather?")]}) response = agent.invoke({"messages": [HumanMessage("What's the weather?")]})
assert response["structured_response"] == EXPECTED_WEATHER_DATACLASS assert response["structured_response"] == EXPECTED_WEATHER_DATACLASS
@@ -156,7 +156,7 @@ class TestResponseFormatAsModel:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent(model, [get_weather], response_format=WeatherTypedDict) agent = create_agent(model, [get_weather], response_format=WeatherTypedDict)
response = agent.invoke({"messages": [HumanMessage("What's the weather?")]}) response = agent.invoke({"messages": [HumanMessage("What's the weather?")]})
assert response["structured_response"] == EXPECTED_WEATHER_DICT assert response["structured_response"] == EXPECTED_WEATHER_DICT
@@ -177,7 +177,7 @@ class TestResponseFormatAsModel:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent(model, [get_weather], response_format=weather_json_schema) agent = create_agent(model, [get_weather], response_format=weather_json_schema)
response = agent.invoke({"messages": [HumanMessage("What's the weather?")]}) response = agent.invoke({"messages": [HumanMessage("What's the weather?")]})
assert response["structured_response"] == EXPECTED_WEATHER_DICT assert response["structured_response"] == EXPECTED_WEATHER_DICT
@@ -200,9 +200,7 @@ class TestResponseFormatAsToolStrategy:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(model, [get_weather], response_format=ToolStrategy(WeatherBaseModel))
model, [get_weather], response_format=ToolStrategy(WeatherBaseModel)
)
response = agent.invoke({"messages": [HumanMessage("What's the weather?")]}) response = agent.invoke({"messages": [HumanMessage("What's the weather?")]})
assert response["structured_response"] == EXPECTED_WEATHER_PYDANTIC assert response["structured_response"] == EXPECTED_WEATHER_PYDANTIC
@@ -223,9 +221,7 @@ class TestResponseFormatAsToolStrategy:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(model, [get_weather], response_format=ToolStrategy(WeatherDataclass))
model, [get_weather], response_format=ToolStrategy(WeatherDataclass)
)
response = agent.invoke({"messages": [HumanMessage("What's the weather?")]}) response = agent.invoke({"messages": [HumanMessage("What's the weather?")]})
assert response["structured_response"] == EXPECTED_WEATHER_DATACLASS assert response["structured_response"] == EXPECTED_WEATHER_DATACLASS
@@ -246,9 +242,7 @@ class TestResponseFormatAsToolStrategy:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(model, [get_weather], response_format=ToolStrategy(WeatherTypedDict))
model, [get_weather], response_format=ToolStrategy(WeatherTypedDict)
)
response = agent.invoke({"messages": [HumanMessage("What's the weather?")]}) response = agent.invoke({"messages": [HumanMessage("What's the weather?")]})
assert response["structured_response"] == EXPECTED_WEATHER_DICT assert response["structured_response"] == EXPECTED_WEATHER_DICT
@@ -269,7 +263,7 @@ class TestResponseFormatAsToolStrategy:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, [get_weather], response_format=ToolStrategy(weather_json_schema) model, [get_weather], response_format=ToolStrategy(weather_json_schema)
) )
response = agent.invoke({"messages": [HumanMessage("What's the weather?")]}) response = agent.invoke({"messages": [HumanMessage("What's the weather?")]})
@@ -292,7 +286,7 @@ class TestResponseFormatAsToolStrategy:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, model,
[get_weather, get_location], [get_weather, get_location],
response_format=ToolStrategy({"oneOf": [weather_json_schema, location_json_schema]}), response_format=ToolStrategy({"oneOf": [weather_json_schema, location_json_schema]}),
@@ -316,7 +310,7 @@ class TestResponseFormatAsToolStrategy:
model_location = FakeToolCallingModel(tool_calls=tool_calls_location) model_location = FakeToolCallingModel(tool_calls=tool_calls_location)
agent_location = create_react_agent( agent_location = create_agent(
model_location, model_location,
[get_weather, get_location], [get_weather, get_location],
response_format=ToolStrategy({"oneOf": [weather_json_schema, location_json_schema]}), response_format=ToolStrategy({"oneOf": [weather_json_schema, location_json_schema]}),
@@ -344,7 +338,7 @@ class TestResponseFormatAsToolStrategy:
tool_calls=tool_calls tool_calls=tool_calls
) )
agent = create_react_agent( agent = create_agent(
model, model,
[get_weather, get_location], [get_weather, get_location],
response_format=ToolStrategy(Union[WeatherBaseModel, LocationResponse]), response_format=ToolStrategy(Union[WeatherBaseModel, LocationResponse]),
@@ -368,7 +362,7 @@ class TestResponseFormatAsToolStrategy:
model_location = FakeToolCallingModel(tool_calls=tool_calls_location) model_location = FakeToolCallingModel(tool_calls=tool_calls_location)
agent_location = create_react_agent( agent_location = create_agent(
model_location, model_location,
[get_weather, get_location], [get_weather, get_location],
response_format=ToolStrategy(Union[WeatherBaseModel, LocationResponse]), response_format=ToolStrategy(Union[WeatherBaseModel, LocationResponse]),
@@ -397,7 +391,7 @@ class TestResponseFormatAsToolStrategy:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, model,
[], [],
response_format=ToolStrategy( response_format=ToolStrategy(
@@ -438,7 +432,7 @@ class TestResponseFormatAsToolStrategy:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, model,
[], [],
response_format=ToolStrategy( response_format=ToolStrategy(
@@ -467,7 +461,7 @@ class TestResponseFormatAsToolStrategy:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, model,
[], [],
response_format=ToolStrategy( response_format=ToolStrategy(
@@ -503,7 +497,7 @@ class TestResponseFormatAsToolStrategy:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, model,
[], [],
response_format=ToolStrategy( response_format=ToolStrategy(
@@ -549,7 +543,7 @@ class TestResponseFormatAsToolStrategy:
return "Custom error: Multiple outputs not allowed" return "Custom error: Multiple outputs not allowed"
return "Custom error" return "Custom error"
agent = create_react_agent( agent = create_agent(
model, model,
[], [],
response_format=ToolStrategy( response_format=ToolStrategy(
@@ -587,7 +581,7 @@ class TestResponseFormatAsToolStrategy:
model = FakeToolCallingModel(tool_calls=tool_calls) model = FakeToolCallingModel(tool_calls=tool_calls)
agent = create_react_agent( agent = create_agent(
model, model,
[], [],
response_format=ToolStrategy( response_format=ToolStrategy(
@@ -617,7 +611,7 @@ class TestResponseFormatAsProviderStrategy:
tool_calls=tool_calls, structured_response=EXPECTED_WEATHER_PYDANTIC tool_calls=tool_calls, structured_response=EXPECTED_WEATHER_PYDANTIC
) )
agent = create_react_agent( agent = create_agent(
model, [get_weather], response_format=ProviderStrategy(WeatherBaseModel) model, [get_weather], response_format=ProviderStrategy(WeatherBaseModel)
) )
response = agent.invoke({"messages": [HumanMessage("What's the weather?")]}) response = agent.invoke({"messages": [HumanMessage("What's the weather?")]})
@@ -635,7 +629,7 @@ class TestResponseFormatAsProviderStrategy:
tool_calls=tool_calls, structured_response=EXPECTED_WEATHER_DATACLASS tool_calls=tool_calls, structured_response=EXPECTED_WEATHER_DATACLASS
) )
agent = create_react_agent( agent = create_agent(
model, [get_weather], response_format=ProviderStrategy(WeatherDataclass) model, [get_weather], response_format=ProviderStrategy(WeatherDataclass)
) )
response = agent.invoke( response = agent.invoke(
@@ -655,7 +649,7 @@ class TestResponseFormatAsProviderStrategy:
tool_calls=tool_calls, structured_response=EXPECTED_WEATHER_DICT tool_calls=tool_calls, structured_response=EXPECTED_WEATHER_DICT
) )
agent = create_react_agent( agent = create_agent(
model, [get_weather], response_format=ProviderStrategy(WeatherTypedDict) model, [get_weather], response_format=ProviderStrategy(WeatherTypedDict)
) )
response = agent.invoke({"messages": [HumanMessage("What's the weather?")]}) response = agent.invoke({"messages": [HumanMessage("What's the weather?")]})
@@ -673,7 +667,7 @@ class TestResponseFormatAsProviderStrategy:
tool_calls=tool_calls, structured_response=EXPECTED_WEATHER_DICT tool_calls=tool_calls, structured_response=EXPECTED_WEATHER_DICT
) )
agent = create_react_agent( agent = create_agent(
model, [get_weather], response_format=ProviderStrategy(weather_json_schema) model, [get_weather], response_format=ProviderStrategy(weather_json_schema)
) )
response = agent.invoke({"messages": [HumanMessage("What's the weather?")]}) response = agent.invoke({"messages": [HumanMessage("What's the weather?")]})
@@ -699,7 +693,7 @@ def test_union_of_types() -> None:
tool_calls=tool_calls, structured_response=EXPECTED_WEATHER_PYDANTIC tool_calls=tool_calls, structured_response=EXPECTED_WEATHER_PYDANTIC
) )
agent = create_react_agent( agent = create_agent(
model, model,
[get_weather, get_location], [get_weather, get_location],
response_format=ToolStrategy(Union[WeatherBaseModel, LocationResponse]), response_format=ToolStrategy(Union[WeatherBaseModel, LocationResponse]),

2
libs/langchain_v1/tests/unit_tests/agents/test_responses_spec.py
View File

@@ -119,7 +119,7 @@ def test_responses_integration_matrix(case: TestCase) -> None:
http_client=http_client, http_client=http_client,
) )
agent = create_react_agent( agent = create_agent(
model, model,
tools=[role_tool["tool"], dept_tool["tool"]], tools=[role_tool["tool"], dept_tool["tool"]],
prompt=AGENT_PROMPT, prompt=AGENT_PROMPT,

4
libs/langchain_v1/tests/unit_tests/agents/test_return_direct_spec.py
View File

@@ -71,14 +71,14 @@ def test_return_direct_integration_matrix(case: TestCase) -> None:
) )
if case.response_format: if case.response_format:
agent = create_react_agent( agent = create_agent(
model, model,
tools=[poll_tool["tool"]], tools=[poll_tool["tool"]],
prompt=AGENT_PROMPT, prompt=AGENT_PROMPT,
response_format=ToolStrategy(case.response_format), response_format=ToolStrategy(case.response_format),
) )
else: else:
agent = create_react_agent( agent = create_agent(
model, model,
tools=[poll_tool["tool"]], tools=[poll_tool["tool"]],
prompt=AGENT_PROMPT, prompt=AGENT_PROMPT,