Installation
pip install pandaprobe[langgraph]
Setup
from pandaprobe.integrations.langgraph import LangGraphCallbackHandler
handler = LangGraphCallbackHandler(
session_id="conversation-123",
user_id="user-abc",
tags=["production"],
)
Usage
Pass the handler via LangChain’s config["callbacks"]:
from langgraph.graph import StateGraph
# ... define your graph ...
result = graph.invoke(
{"messages": [{"role": "user", "content": "Hello!"}]},
config={"callbacks": [handler]},
)
Unlike other integrations, LangGraph does not use the instrument() pattern. You must pass the handler in config for each invocation.
What gets traced
| LangChain Callback | Span Kind | Description |
|---|
on_chain_start / on_chain_end | CHAIN (root) or AGENT (nested) | Root chain creates the trace boundary |
on_llm_start / on_chat_model_start / on_llm_end | LLM | Model, parameters, token usage, reasoning |
on_tool_start / on_tool_end | TOOL | Tool name, input, output |
on_retriever_start / on_retriever_end | RETRIEVER | Retrieval queries and results |
from langchain_openai import ChatOpenAI
from langgraph.prebuilt import create_react_agent
from langchain_core.tools import tool
@tool
def get_weather(city: str) -> str:
"""Get the weather for a city."""
return f"The weather in {city} is sunny, 72F."
model = ChatOpenAI(model="gpt-4o")
agent = create_react_agent(model, [get_weather])
handler = LangGraphCallbackHandler(session_id="s1")
result = agent.invoke(
{"messages": [{"role": "user", "content": "What's the weather in Paris?"}]},
config={"callbacks": [handler]},
)
CHAIN (root) → LLM (model call) → TOOL (get_weather) → LLM (final response).
Token usage
Token usage is extracted from LangChain’s usage_metadata (primary) or legacy llm_output.token_usage (fallback). The mapping is: input_tokens → prompt_tokens, output_tokens → completion_tokens. Reasoning tokens are subtracted from output_tokens when present.