@sentry/node

Important Changes

• core: Instrument LangGraph Agent

Adds support for instrumenting LangGraph StateGraph operations in Node. The LangGraph integration can be configured as follows:

Sentry.init({
  dsn: '__DSN__',
  sendDefaultPii: false, // Even with PII disabled globally
  integrations: [
    Sentry.langGraphIntegration({
      recordInputs: true, // Force recording input messages
      recordOutputs: true, // Force recording response text
    }),
  ],
});

• cloudflare/vercel-edge: Add manual instrumentation for LangGraph

Instrumentation for LangGraph in Cloudflare Workers and Vercel Edge environments is supported by manually calling instrumentLangGraph:

import * as Sentry from '@sentry/cloudflare'; // or '@sentry/vercel-edge'
import { StateGraph, START, END, MessagesAnnotation } from '@langchain/langgraph';

// Create and instrument the graph
const graph = new StateGraph(MessagesAnnotation)
  .addNode('agent', agentFn)
  .addEdge(START, 'agent')
  .addEdge('agent', END);

Sentry.instrumentLangGraph(graph, {
  recordInputs: true,
  recordOutputs: true,
});

const compiled = graph.compile({ name: 'weather_assistant' });

await compiled.invoke({
  messages: [{ role: 'user', content: 'What is the weather in SF?' }],
});

• node: Add OpenAI SDK v6 support

Other Changes

• core: Support OpenAI embeddings API

• core: Support truncation for LangChain integration request messages

• metrics: Add default server.address attribute on server runtimes

• nextjs: Add URL to server-side transaction events

• node-core: Add mechanism to prevent wrapping ai providers multiple times

• replay: Bump limit for minReplayDuration

• browser: Add ok status to successful idleSpans

• core: Check fetch support with data URL

• core: Decrease number of Sentry stack frames for messages from captureConsoleIntegration

• core: Emit processed metric

• core: Ensure logs past MAX_LOG_BUFFER_SIZE are not swallowed

• core: Ensure metrics past MAX_METRIC_BUFFER_SIZE are not swallowed

• core: Fix logs and metrics flush timeout starvation with continuous logging

• core: Flatten gen_ai.request.available_tools in google-genai

• core: Stringify available tools sent from vercelai

• core/vue: Detect and skip normalizing Vue VNode objects with high normalizeDepth

• nextjs: Avoid wrapping middleware files when in standalone mode

• nextjs: Drop meta trace tags if rendered page is ISR

• nextjs: Respect PORT variable for dev error symbolication

• nextjs: Use LRU map instead of map for ISR route cache

• node: tracingChannel export missing in older node versions

• node: Fix Spotlight configuration precedence to match specification

• react: Prevent navigation span leaks for consecutive navigations

• deprecated

  react-router: Deprecate ErrorBoundary exports

Internal Changes