Skip to main content
Automatically capture agent steps, tool calls, and LLM invocations within LangChain chains and LangGraph graphs.

​
Setup

import traceroot
from traceroot import Integration

traceroot.initialize(integrations=[Integration.LANGCHAIN])

​
Usage with LangGraph

from langchain_openai import ChatOpenAI
from langgraph.prebuilt import create_react_agent

llm = ChatOpenAI(model="gpt-4o")

tools = [search_tool, calculator_tool]
agent = create_react_agent(llm, tools=tools)

# All agent steps are automatically traced
result = agent.invoke({
    "messages": [{"role": "user", "content": "What is 2 + 2?"}]
})

​
Usage with LangChain

from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate

llm = ChatOpenAI(model="gpt-4o")
prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a helpful assistant."),
    ("user", "{input}"),
])

chain = prompt | llm

# Chain execution is automatically traced
result = chain.invoke({"input": "Hello!"})

​
What Gets Captured

AttributeDescription
Agent stepsEach agent iteration as a span
Tool callsEach tool invocation with input/output
LLM callsAll LLM calls within the chain/graph
Graph structureParent-child relationships between steps
Token usageAggregated across all LLM calls
CostTotal cost for the full chain/graph execution