LangGraph agent observability

Test the integration

Send and verify events with LangGraph agent observability

Use LangGraph agent observability where your app knows the final result. Collect only the fields you need, then verify a test event before building charts.

1
Choose the outcome
Stateful agent monitoring
2
Define the contract
run_id, graph_name, node_name, and status
3
Log the final result
Emit node-level events from a controlled wrapper or callback and one final run event after graph invocation.
4
Check the stored event
Exercise a known fixture, then inspect agent_run_completed for one correctly typed terminal row.

Before you start

@langchain/langgraph and @langchain/core
A stable run ID shared across graph events
A server-side TELEMETRY_API_KEY

Delivery setup

Install and initialize server-side

Import telemetry-sh in server-only code and initialize it once with process.env.TELEMETRY_API_KEY. Keep ingestion credentials out of browser bundles, client-visible environment variables, source control, logs, and exception messages.

langgraph-install

npm installation

bash

npm install telemetry-sh

1Create one reusable server-side client. Set its timeout and retry limit.
2Log an event when the operation succeeds, fails, retries, or times out.
3Send test events with known results and inspect the stored rows before enabling an alert.

Snippet

Start with one structured event

Add this shape where the workflow completes, fails, or retries. Then build the dashboard from real fields.

langgraph

LangGraph agent observability event

javascript

const runId = crypto.randomUUID();
const startedAt = performance.now();
const result = await graph.invoke(
  { messages },
  { configurable: { thread_id: runId } }
);

await telemetry.log("agent_run_completed", {
  run_id: runId,
  graph_name: "support_agent",
  status: "success",
  message_count: result.messages.length,
  human_handoff: false,
  duration_ms: Math.round(performance.now() - startedAt),
  release: process.env.APP_RELEASE,
});

Event schema

run_id, graph_name, node_name, and status

tool_name, attempt, latency_ms, and error_type

checkpoint_count, human_handoff, reviewer_outcome, and release

Check your setup

Checkpoint 1

Emit node-level events from a controlled wrapper or callback and one final run event after graph invocation.

Checkpoint 2

Keep run ID stable across resumed checkpoints while recording each execution attempt separately.

Checkpoint 3

Do not send graph state, messages, checkpoint contents, or tool arguments wholesale.

Verification

Prove the event arrived

Run this after exercising known success and failure cases. Replace the fallback table name if your final event contract differs from the snippet.

langgraph-verification

LangGraph agent observability verification query

sql

SELECT *
FROM agent_run_completed
ORDER BY timestamp_utc DESC
LIMIT 20;

Confirm one terminal row per logical outcome, with the expected status, identifiers, units, and UTC time.

Inspect the inferred schema and verify that retries do not change field types or generate a new logical event ID.

Search the stored fields for credentials, raw payloads, prompts, private content, and unbounded error messages.

Exercise a provider timeout, ingestion rejection, and process shutdown before treating the dashboard as complete.

Implementation references

Review the event contract, data-safety guidance, and upstream primary documentation before enabling a new production path.

Nested tool-event SQL Agent quality SQL Official LangGraph JavaScript overview

Where to log

Keep the outcome event small and recoverable

This pattern provides

Record the outcome as an event you can query with SQL.
Stable fields for dashboards, alerts, and cross-event correlation.
Test events for checking success, failure, retries, and timeouts.

This pattern does not provide

An OTLP exporter, automatic collection pipeline, or replacement for detailed traces and diagnostic logs.
Exactly-once delivery merely because the payload contains an event ID.
Permission to collect raw provider payloads, user content, credentials, or regulated data.

Delivery and idempotency Batching and shutdown Troubleshoot ingestion

Example event schemas

Check what each event records, when to send it, and which field types it needs. Review the example payload and privacy checklist before using it in production.

AI operations10 fields