mirror of
https://github.com/langchain-ai/langgraph.git
synced 2026-08-24 08:32:24 +02:00
Sydney RunkleandGitHub
fdebb1dd18
Raised by @jacoblee93, the following was raising a cryptic deprecation
warning:
```py
import pytest
from langchain_core.language_models import FakeListChatModel
from langchain.agents import create_agent
@pytest.fixture
def test_llm():
return FakeListChatModel(responses=["Hello from test agent!"])
@pytest.fixture
def simple_agent(test_llm):
return create_agent(
model=test_llm,
tools=[],
prompt="You are a helpful assistant."
)
def test_basic_agent_execution(simple_agent):
result = simple_agent.invoke({"messages": [{"role": "user", "content": "hi"}]})
assert len(result["messages"]) == 2
assert result["messages"][1].content == "Hello from test agent!"
```
```
<frozen abc>:106
<frozen abc>:106: LangGraphDeprecatedSinceV10: AgentStatePydantic has been moved to langchain.agents. Please update your import to 'from langchain.agents import AgentStatePydantic'. Deprecated in LangGraph V1.0 to be removed in V2.0.
-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
```
This fixes that issue.
I don't actually think we should re-expose these new states in
`langchain.agents` given that middleware agents don't use them. Need to
make a call on that.