LangSmith Observability

Traces are a series of steps that your agent takes to go from input to output. Each of these individual steps is represented by a run. You can use LangSmith to visualize these execution steps. To use it, enable tracing for your agent. This enables you to do the following:

Prerequisites

Before you begin, ensure you have the following:

A LangSmith account: Sign up (for free) or log in at smith.langchain.com.
A LangSmith API key: Follow the Create an API key guide.

Enable tracing

Note

In an agent-based workspace, create an agent once before sending traces. Use agent addressing to select its identifier and environment. If an integration requires project_name, use the selected environment's tracer_session_name as a compatibility value.

To enable tracing for your agent, set the following environment variables:

export LANGSMITH_AGENT_ID="<your-agent-id>"
export LANGSMITH_AGENT_ENVIRONMENT="production"
export LANGSMITH_TRACING=true
export LANGSMITH_API_KEY=<your-api-key>

Traces appear in the destination configured during setup.

For more information, see Trace with LangGraph.

Trace selectively

You may opt to trace specific invocations or parts of your agent using LangSmith's tracing_context context manager:

import langsmith as ls

# This WILL be traced
with ls.tracing_context(enabled=True):
    agent.invoke({"messages": [{"role": "user", "content": "Send a test email to alice@example.com"}]})

# This will NOT be traced (if LANGSMITH_TRACING is not set)
agent.invoke({"messages": [{"role": "user", "content": "Send another email"}]})

Select an agent and environment

Use LANGSMITH_AGENT_ID and LANGSMITH_AGENT_ENVIRONMENT to select the registered agent and environment. For per-request routing, pass an agent address in code. See Log traces to an agent.

Add metadata to traces

You can annotate your traces with custom metadata and tags:

response = agent.invoke(
    {"messages": [{"role": "user", "content": "Send a welcome email"}]},
    config={
        "tags": ["production", "email-assistant", "v1.0"],
        "metadata": {
            "user_id": "user_123",
            "session_id": "session_456",
            "environment": "production"
        }
    }
)

tracing_context also accepts tags and metadata for fine-grained control:

with ls.tracing_context(
    enabled=True,
    tags=["production", "email-assistant", "v1.0"],
    metadata={"user_id": "user_123", "session_id": "session_456", "environment": "production"}):
    response = agent.invoke(
        {"messages": [{"role": "user", "content": "Send a welcome email"}]}
    )

This custom metadata and tags will be attached to the trace in LangSmith.

Tip

To learn more about how to use traces to debug, evaluate, and monitor your agents, see the LangSmith documentation.

Use anonymizers to prevent logging of sensitive data in traces

You may want to mask sensitive data to prevent it from being logged to LangSmith. You can create anonymizers and apply them to your graph using configuration. This example will redact anything matching the Social Security Number format XXX-XX-XXXX from traces sent to LangSmith.

from langchain_core.tracers.langchain import LangChainTracer
from langgraph.graph import StateGraph, MessagesState
from langsmith import Client
from langsmith.anonymizer import create_anonymizer

anonymizer = create_anonymizer([
    # Matches SSNs
    { "pattern": r"\b\d{3}-?\d{2}-?\d{4}\b", "replace": "<ssn>" }
])

tracer_client = Client(anonymizer=anonymizer)
tracer = LangChainTracer(client=tracer_client)
# Define the graph
graph = (
    StateGraph(MessagesState)
    ...
    .compile()
    .with_config({'callbacks': [tracer]})
)