newrelic 13.14.0 → 13.16.0

package/NEWS.md

@@ -1,3 +1,54 @@
+### v13.16.0 (2026-03-11)
+
+#### Bug fixes
+
+* Updated `@apm-js-collab/tracing-hooks` to `0.5.0`, to allow instrumentation on windows environments. ([#3819](https://github.com/newrelic/node-newrelic/pull/3819)) ([d8f6d0f](https://github.com/newrelic/node-newrelic/commit/d8f6d0f6ac7f3ff17a667391269c41e6e57dc2fb))
+	* This version of `@apm-js-collab/tracing-hooks` includes the windows path fix for `@apm-js-collab/code-transformer`, [see](https://github.com/apm-js-collab/tracing-hooks/issues/19).
+
+#### Code refactoring
+
+* Updated `redis` v3 instrumentation to subscribe to events emitted ([#3802](https://github.com/newrelic/node-newrelic/pull/3802)) ([4d18302](https://github.com/newrelic/node-newrelic/commit/4d1830237c91d3e60c6b4e898f2c82ff131ebd57))
+
+#### Documentation
+
+* Updated compatibility report ([#3813](https://github.com/newrelic/node-newrelic/pull/3813)) ([27cb0a5](https://github.com/newrelic/node-newrelic/commit/27cb0a5e7a4c6047cce54483bd18d1337023efd3))
+
+#### Tests
+
+* Add `redis-client` and `node-redis-client` segment duration timing tests ([#3810](https://github.com/newrelic/node-newrelic/pull/3810)) ([f07bcfe](https://github.com/newrelic/node-newrelic/commit/f07bcfecf06453c846e12f030fbf7d8c055da139))
+
+### v13.15.0 (2026-03-09)
+
+#### Features
+
+* Added `time_to_first_token` for `LlmChatCompletionSummary` ([#3781](https://github.com/newrelic/node-newrelic/pull/3781)) ([737f374](https://github.com/newrelic/node-newrelic/commit/737f3740d43908be348a7795dd2c85e7a4e88cb8))
+* Added `time_to_first_token` for AWS Bedrock ([#3804](https://github.com/newrelic/node-newrelic/pull/3804)) ([a993b00](https://github.com/newrelic/node-newrelic/commit/a993b00b4fa55fb2e7ba0b7748c4c21d8660b911))
+* Added `time_to_first_token` for Google Gen AI ([#3785](https://github.com/newrelic/node-newrelic/pull/3785)) ([d90ee70](https://github.com/newrelic/node-newrelic/commit/d90ee709538e430395c0fb5863c344fe37009ce5))
+* Added `time_to_first_token` for OpenAI ([#3782](https://github.com/newrelic/node-newrelic/pull/3782)) ([07cfd57](https://github.com/newrelic/node-newrelic/commit/07cfd570d5cc3dd77a5ae23e88587d47af3d195a))
+
+#### Code refactoring
+
+* Migrated `redis` 4+ instrumentation to subscribe to events emitted ([#3773](https://github.com/newrelic/node-newrelic/pull/3773)) ([add6827](https://github.com/newrelic/node-newrelic/commit/add68278d29f5f415134426d33986a59c089d71f))
+
+#### Documentation
+
+* Updated compatibility report ([#3800](https://github.com/newrelic/node-newrelic/pull/3800)) ([5228cdc](https://github.com/newrelic/node-newrelic/commit/5228cdc0508217fa139d8b4f8232391fd063a5fa)) ([#3795](https://github.com/newrelic/node-newrelic/pull/3795)) ([1d6fe52](https://github.com/newrelic/node-newrelic/commit/1d6fe522f0e707729a4a01f5c5e5599670136917)) ([#3791](https://github.com/newrelic/node-newrelic/pull/3791)) ([e79db5b](https://github.com/newrelic/node-newrelic/commit/e79db5bbac1500bfa49f8ac3e9dc64c409396d79)) ([#3777](https://github.com/newrelic/node-newrelic/pull/3777)) ([12fe4c6](https://github.com/newrelic/node-newrelic/commit/12fe4c6f2a7a3f08efd874ba949150c7805a8888))
+
+#### Miscellaneous chores
+
+* Added version comment for agent-metadata.yml ([#3779](https://github.com/newrelic/node-newrelic/pull/3779)) ([1267293](https://github.com/newrelic/node-newrelic/commit/1267293cc36c79106b7cdd84dedf2c18b8a010d9))
+* Refactored nestjs instrumentation to subscriber type ([#3792](https://github.com/newrelic/node-newrelic/pull/3792)) ([f7feb25](https://github.com/newrelic/node-newrelic/commit/f7feb25a5c7349433e7de26a0f6aebb820c034f4))
+* Removed `@langchain/core` peer dep from langchain openai and aws tests ([#3788](https://github.com/newrelic/node-newrelic/pull/3788)) ([e670bb1](https://github.com/newrelic/node-newrelic/commit/e670bb1f47720c61509157cf45cd200234505733))
+* Removed unneccessary `generic-pool` context propagation ([#3786](https://github.com/newrelic/node-newrelic/pull/3786)) ([bdd35fb](https://github.com/newrelic/node-newrelic/commit/bdd35fbf6b57d374de84f716488498a40ece41fd))
+* Removed unnecessary `superagent` context propagation ([#3798](https://github.com/newrelic/node-newrelic/pull/3798)) ([81650e1](https://github.com/newrelic/node-newrelic/commit/81650e18d7da3897f9c82098def2f34c05c78d9e))
+* Updated `@apm-js-collab/tracing-hooks` to `0.4.0` ([#3787](https://github.com/newrelic/node-newrelic/pull/3787)) ([d85fb2b](https://github.com/newrelic/node-newrelic/commit/d85fb2b832bc7202e5acafa19ff59094c6e386ee))
+* Updated version format in agent-metadata.yml ([#3778](https://github.com/newrelic/node-newrelic/pull/3778)) ([61759e9](https://github.com/newrelic/node-newrelic/commit/61759e9c4e9ad2796a6462b967e30bad35b3fbf7))
+
+#### Tests
+
+* Updated langchain vectorstore tests to use a custom vectorstore to simplify the testing process ([#3793](https://github.com/newrelic/node-newrelic/pull/3793)) ([0fd9b54](https://github.com/newrelic/node-newrelic/commit/0fd9b54c0b4782162dae769453169ad4b05d7918))
+* Updated version range for langgraph tests ([#3794](https://github.com/newrelic/node-newrelic/pull/3794)) ([e02ce99](https://github.com/newrelic/node-newrelic/commit/e02ce99083d584d8b61a223c67c4509f3fa9f4ce))
+
 ### v13.14.0 (2026-02-24)
 
 #### Features
@@ -8397,6 +8448,9 @@ Special thanks to Ryan Copley (@RyanCopley) for the contribution.
 
 
 
+
+
+
 
 
 

package/THIRD_PARTY_NOTICES.md

@@ -94,7 +94,7 @@ code, the source code can be found at [https://github.com/newrelic/node-newrelic
 
 ### @apm-js-collab/tracing-hooks
 
-This product includes source derived from [@apm-js-collab/tracing-hooks](https://github.com/apm-js-collab/tracing-hooks) ([v0.3.1](https://github.com/apm-js-collab/tracing-hooks/tree/v0.3.1)), distributed under the [Apache-2.0 License](https://github.com/apm-js-collab/tracing-hooks/blob/v0.3.1/LICENSE):
+This product includes source derived from [@apm-js-collab/tracing-hooks](https://github.com/apm-js-collab/tracing-hooks) ([v0.4.0](https://github.com/apm-js-collab/tracing-hooks/tree/v0.4.0)), distributed under the [Apache-2.0 License](https://github.com/apm-js-collab/tracing-hooks/blob/v0.4.0/LICENSE):
 
 ```
 
@@ -722,7 +722,7 @@ This product includes source derived from [@grpc/proto-loader](https://github.co
 
 ### @newrelic/security-agent
 
-This product includes source derived from [@newrelic/security-agent](https://github.com/newrelic/csec-node-agent) ([v3.0.1](https://github.com/newrelic/csec-node-agent/tree/v3.0.1)), distributed under the [UNKNOWN License](https://github.com/newrelic/csec-node-agent/blob/v3.0.1/LICENSE):
+This product includes source derived from [@newrelic/security-agent](https://github.com/newrelic/csec-node-agent) ([v3.0.2](https://github.com/newrelic/csec-node-agent/tree/v3.0.2)), distributed under the [UNKNOWN License](https://github.com/newrelic/csec-node-agent/blob/v3.0.2/LICENSE):
 
 ```
 ## New Relic Software License v1.0
@@ -1186,7 +1186,7 @@ This product includes source derived from [@opentelemetry/api](https://github.co
 
 ### @opentelemetry/core
 
-This product includes source derived from [@opentelemetry/core](https://github.com/open-telemetry/opentelemetry-js) ([v2.5.0](https://github.com/open-telemetry/opentelemetry-js/tree/v2.5.0)), distributed under the [Apache-2.0 License](https://github.com/open-telemetry/opentelemetry-js/blob/v2.5.0/LICENSE):
+This product includes source derived from [@opentelemetry/core](https://github.com/open-telemetry/opentelemetry-js) ([v2.5.1](https://github.com/open-telemetry/opentelemetry-js/tree/v2.5.1)), distributed under the [Apache-2.0 License](https://github.com/open-telemetry/opentelemetry-js/blob/v2.5.1/LICENSE):
 
 ```
                                  Apache License
@@ -1604,7 +1604,7 @@ This product includes source derived from [@opentelemetry/exporter-metrics-otlp-
 
 ### @opentelemetry/resources
 
-This product includes source derived from [@opentelemetry/resources](https://github.com/open-telemetry/opentelemetry-js) ([v2.5.0](https://github.com/open-telemetry/opentelemetry-js/tree/v2.5.0)), distributed under the [Apache-2.0 License](https://github.com/open-telemetry/opentelemetry-js/blob/v2.5.0/LICENSE):
+This product includes source derived from [@opentelemetry/resources](https://github.com/open-telemetry/opentelemetry-js) ([v2.5.1](https://github.com/open-telemetry/opentelemetry-js/tree/v2.5.1)), distributed under the [Apache-2.0 License](https://github.com/open-telemetry/opentelemetry-js/blob/v2.5.1/LICENSE):
 
 ```
                                  Apache License
@@ -2022,7 +2022,7 @@ This product includes source derived from [@opentelemetry/sdk-logs](https://gith
 
 ### @opentelemetry/sdk-metrics
 
-This product includes source derived from [@opentelemetry/sdk-metrics](https://github.com/open-telemetry/opentelemetry-js) ([v2.5.0](https://github.com/open-telemetry/opentelemetry-js/tree/v2.5.0)), distributed under the [Apache-2.0 License](https://github.com/open-telemetry/opentelemetry-js/blob/v2.5.0/LICENSE):
+This product includes source derived from [@opentelemetry/sdk-metrics](https://github.com/open-telemetry/opentelemetry-js) ([v2.5.1](https://github.com/open-telemetry/opentelemetry-js/tree/v2.5.1)), distributed under the [Apache-2.0 License](https://github.com/open-telemetry/opentelemetry-js/blob/v2.5.1/LICENSE):
 
 ```
                                  Apache License
@@ -2231,7 +2231,7 @@ This product includes source derived from [@opentelemetry/sdk-metrics](https://g
 
 ### @opentelemetry/sdk-trace-base
 
-This product includes source derived from [@opentelemetry/sdk-trace-base](https://github.com/open-telemetry/opentelemetry-js) ([v2.5.0](https://github.com/open-telemetry/opentelemetry-js/tree/v2.5.0)), distributed under the [Apache-2.0 License](https://github.com/open-telemetry/opentelemetry-js/blob/v2.5.0/LICENSE):
+This product includes source derived from [@opentelemetry/sdk-trace-base](https://github.com/open-telemetry/opentelemetry-js) ([v2.5.1](https://github.com/open-telemetry/opentelemetry-js/tree/v2.5.1)), distributed under the [Apache-2.0 License](https://github.com/open-telemetry/opentelemetry-js/blob/v2.5.1/LICENSE):
 
 ```
                                  Apache License
@@ -2904,7 +2904,7 @@ SOFTWARE.
 
 ### semver
 
-This product includes source derived from [semver](https://github.com/npm/node-semver) ([v7.7.3](https://github.com/npm/node-semver/tree/v7.7.3)), distributed under the [ISC License](https://github.com/npm/node-semver/blob/v7.7.3/LICENSE):
+This product includes source derived from [semver](https://github.com/npm/node-semver) ([v7.7.4](https://github.com/npm/node-semver/tree/v7.7.4)), distributed under the [ISC License](https://github.com/npm/node-semver/blob/v7.7.4/LICENSE):
 
 ```
 The ISC License
@@ -2960,7 +2960,7 @@ SOFTWARE.
 
 ### @aws-sdk/client-s3
 
-This product includes source derived from [@aws-sdk/client-s3](https://github.com/aws/aws-sdk-js-v3) ([v3.982.0](https://github.com/aws/aws-sdk-js-v3/tree/v3.982.0)), distributed under the [Apache-2.0 License](https://github.com/aws/aws-sdk-js-v3/blob/v3.982.0/LICENSE):
+This product includes source derived from [@aws-sdk/client-s3](https://github.com/aws/aws-sdk-js-v3) ([v3.996.0](https://github.com/aws/aws-sdk-js-v3/tree/v3.996.0)), distributed under the [Apache-2.0 License](https://github.com/aws/aws-sdk-js-v3/blob/v3.996.0/LICENSE):
 
 ```
                                 Apache License
@@ -3169,7 +3169,7 @@ This product includes source derived from [@aws-sdk/client-s3](https://github.co
 
 ### @aws-sdk/s3-request-presigner
 
-This product includes source derived from [@aws-sdk/s3-request-presigner](https://github.com/aws/aws-sdk-js-v3) ([v3.982.0](https://github.com/aws/aws-sdk-js-v3/tree/v3.982.0)), distributed under the [Apache-2.0 License](https://github.com/aws/aws-sdk-js-v3/blob/v3.982.0/LICENSE):
+This product includes source derived from [@aws-sdk/s3-request-presigner](https://github.com/aws/aws-sdk-js-v3) ([v3.996.0](https://github.com/aws/aws-sdk-js-v3/tree/v3.996.0)), distributed under the [Apache-2.0 License](https://github.com/aws/aws-sdk-js-v3/blob/v3.996.0/LICENSE):
 
 ```
                                 Apache License
@@ -4510,7 +4510,7 @@ Apache License
 
 ### ajv
 
-This product includes source derived from [ajv](https://github.com/ajv-validator/ajv) ([v6.12.6](https://github.com/ajv-validator/ajv/tree/v6.12.6)), distributed under the [MIT License](https://github.com/ajv-validator/ajv/blob/v6.12.6/LICENSE):
+This product includes source derived from [ajv](https://github.com/ajv-validator/ajv) ([v6.14.0](https://github.com/ajv-validator/ajv/tree/v6.14.0)), distributed under the [MIT License](https://github.com/ajv-validator/ajv/blob/v6.14.0/LICENSE):
 
 ```
 The MIT License (MIT)
@@ -5000,7 +5000,7 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 ### eslint
 
-This product includes source derived from [eslint](https://github.com/eslint/eslint) ([v9.39.2](https://github.com/eslint/eslint/tree/v9.39.2)), distributed under the [MIT License](https://github.com/eslint/eslint/blob/v9.39.2/LICENSE):
+This product includes source derived from [eslint](https://github.com/eslint/eslint) ([v9.39.3](https://github.com/eslint/eslint/tree/v9.39.3)), distributed under the [MIT License](https://github.com/eslint/eslint/blob/v9.39.3/LICENSE):
 
 ```
 Copyright OpenJS Foundation and other contributors, <www.openjsf.org>

package/lib/instrumentation/aws-sdk/v3/bedrock.js

@@ -112,6 +112,7 @@ function addLlmMeta({ agent, segment, transaction }) {
  * @param {BedrockResponse} params.bedrockResponse parsed response
  * @param {Shim} params.shim shim instance
  * @param {Transaction} params.transaction active transaction
+ * @param {number} [params.timeOfFirstToken] For streaming requests only, timestamp of when first token was sent.
  */
 function recordChatCompletionMessages({
   agent,
@@ -120,7 +121,8 @@ function recordChatCompletionMessages({
   transaction,
   bedrockCommand,
   bedrockResponse,
-  err
+  err,
+  timeOfFirstToken
 }) {
   if (shouldSkipInstrumentation(agent.config) === true) {
     shim.logger.debug('skipping sending of ai data')
@@ -133,6 +135,7 @@ function recordChatCompletionMessages({
     bedrockCommand,
     transaction,
     segment,
+    timeOfFirstToken,
     error: err !== null
   })
 
@@ -341,6 +344,7 @@ function handleResponse({ shim, err, response, segment, transaction, bedrockComm
       transaction,
       bedrockCommand,
       bedrockResponse,
+      timeOfFirstToken: this?.timeOfFirstToken,
       err
     })
   } else if (modelType === 'embedding') {

package/lib/instrumentation/aws-sdk/v3/converse-stream-handler.js

@@ -80,12 +80,19 @@ class ConverseStreamHandler {
    */
   onComplete
 
+  /**
+   * Timestamp of when first token was sent; needed for construction of the
+   * `LlmChatCompletionSummary` event.
+   */
+  timeOfFirstToken
+
   constructor({ stream, passThroughParams, onComplete }) {
     this.passThroughParams = passThroughParams
     this.stream = stream
     this.onComplete = onComplete
     this.segment = passThroughParams.segment
     this.generator = handleConverse
+    this.timeOfFirstToken = null
   }
 
   /**
@@ -97,6 +104,7 @@ class ConverseStreamHandler {
    */
   finish() {
     this.passThroughParams.response = this.response
+    this.passThroughParams.timeOfFirstToken = this.timeOfFirstToken
     this.onComplete(this.passThroughParams)
 
     this.segment.touch()
@@ -136,6 +144,10 @@ async function * handleConverse() {
           activeChunk = null
         }
       } else if (event.contentBlockDelta?.delta) {
+        // Delta represents a letter/word/token
+        if (!this.timeOfFirstToken) {
+          this.timeOfFirstToken = Date.now()
+        }
         // There are also deltas for tool use (stringified inputs) but we don't currently record them so we just ignore for now
         if (event.contentBlockDelta.delta.text) {
           // It seems like the first streamed chunk does not always start with a contentBlockStart message

package/lib/instrumentation/aws-sdk/v3/stream-handler.js

@@ -83,11 +83,18 @@ class StreamHandler {
    */
   onComplete
 
+  /**
+   * Timestamp of when first token was sent; needed for construction of the
+   * `LlmChatCompletionSummary` event.
+   */
+  timeOfFirstToken
+
   constructor({ stream, passThroughParams, onComplete }) {
     this.passThroughParams = passThroughParams
     this.stream = stream
     this.onComplete = onComplete
     this.segment = passThroughParams.segment
+    this.timeOfFirstToken = null
 
     const { bedrockCommand } = passThroughParams
 
@@ -204,6 +211,7 @@ async function * handleClaude() {
 
   try {
     for await (const event of this.stream) {
+      if (!this.timeOfFirstToken) this.timeOfFirstToken = Date.now()
       yield event
       const parsed = this.parseEvent(event)
       this.updateHeaders(parsed)
@@ -224,6 +232,7 @@ async function * handleClaude3() {
 
   try {
     for await (const event of this.stream) {
+      if (!this.timeOfFirstToken) this.timeOfFirstToken = Date.now()
       yield event
       const parsed = this.parseEvent(event)
       this.updateHeaders(parsed)
@@ -247,6 +256,7 @@ async function * handleCohere() {
   const generations = []
   try {
     for await (const event of this.stream) {
+      if (!this.timeOfFirstToken) this.timeOfFirstToken = Date.now()
       yield event
       const parsed = this.parseEvent(event)
       this.updateHeaders(parsed)
@@ -265,6 +275,7 @@ async function * handleCohereEmbed() {
   const embeddings = []
   try {
     for await (const event of this.stream) {
+      if (!this.timeOfFirstToken) this.timeOfFirstToken = Date.now()
       yield event
       const parsed = this.parseEvent(event)
       this.updateHeaders(parsed)
@@ -284,6 +295,7 @@ async function * handleLlama() {
 
   try {
     for await (const event of this.stream) {
+      if (!this.timeOfFirstToken) this.timeOfFirstToken = Date.now()
       yield event
       const parsed = this.parseEvent(event)
       this.updateHeaders(parsed)
@@ -308,6 +320,7 @@ async function * handleTitan() {
 
   try {
     for await (const event of this.stream) {
+      if (!this.timeOfFirstToken) this.timeOfFirstToken = Date.now()
       yield event // Pass it up to the real consumer of the stream.
       const parsed = this.parseEvent(event)
       this.updateHeaders(parsed)

package/lib/instrumentations.js

@@ -14,24 +14,18 @@ module.exports = function instrumentations() {
     '@grpc/grpc-js': { module: './instrumentation/grpc-js' },
     '@hapi/hapi': { type: InstrumentationDescriptor.TYPE_WEB_FRAMEWORK },
     '@hapi/vision': { type: InstrumentationDescriptor.TYPE_WEB_FRAMEWORK },
-    '@nestjs/core': { type: InstrumentationDescriptor.TYPE_WEB_FRAMEWORK },
-    '@node-redis/client': { type: InstrumentationDescriptor.TYPE_DATASTORE },
     '@prisma/client': { type: InstrumentationDescriptor.TYPE_DATASTORE },
-    '@redis/client': { type: InstrumentationDescriptor.TYPE_DATASTORE },
     'aws-sdk': { module: './instrumentation/aws-sdk' },
     bluebird: { type: InstrumentationDescriptor.TYPE_PROMISE },
     connect: { type: InstrumentationDescriptor.TYPE_WEB_FRAMEWORK },
     fastify: { type: InstrumentationDescriptor.TYPE_WEB_FRAMEWORK },
-    'generic-pool': { type: InstrumentationDescriptor.TYPE_GENERIC },
     kafkajs: { type: InstrumentationDescriptor.TYPE_MESSAGE },
     koa: { module: './instrumentation/koa' },
     memcached: { type: InstrumentationDescriptor.TYPE_DATASTORE },
     mongodb: { type: InstrumentationDescriptor.TYPE_DATASTORE },
     next: { module: './instrumentation/nextjs' },
     q: { type: null },
-    redis: { type: InstrumentationDescriptor.TYPE_DATASTORE },
     restify: { type: InstrumentationDescriptor.TYPE_WEB_FRAMEWORK },
-    superagent: { type: InstrumentationDescriptor.TYPE_GENERIC },
     when: { module: './instrumentation/when' },
     winston: { type: InstrumentationDescriptor.TYPE_GENERIC }
   }

package/lib/llm-events/aws-bedrock/chat-completion-summary.js

@@ -16,9 +16,10 @@ module.exports = class AwsBedrockLlmChatCompletionSummary extends LlmChatComplet
    * @param {object} params.transaction Current and active transaction
    * @param {object} params.bedrockCommand AWS Bedrock Command object, represents the request
    * @param {object} params.bedrockResponse AWS Bedrock Response object
+   * @param {number} [params.timeOfFirstToken] For streaming requests only, timestamp of when first token was sent.
    * @param {boolean} [params.error] Set to `true` if an error occurred during creation call, omitted if no error occurred
    */
-  constructor({ agent, segment, transaction, bedrockCommand, bedrockResponse, error }) {
+  constructor({ agent, segment, transaction, bedrockCommand, bedrockResponse, timeOfFirstToken, error }) {
     super({ agent,
       segment,
       transaction,
@@ -30,7 +31,8 @@ module.exports = class AwsBedrockLlmChatCompletionSummary extends LlmChatComplet
       temperature: bedrockCommand.temperature,
       maxTokens: bedrockCommand.maxTokens,
       numMsgs: (bedrockCommand.prompt.length ?? 0) + (bedrockResponse.completions.length ?? 0),
-      finishReason: bedrockResponse?.finishReason })
+      finishReason: bedrockResponse?.finishReason,
+      timeOfFirstToken })
 
     this.setTokens(agent, bedrockCommand, bedrockResponse)
   }

package/lib/llm-events/chat-completion-summary.js

@@ -28,6 +28,9 @@ const LlmEvent = require('./event-base')
  *  a chat completion including system, user, and assistant messages
  * @property {string} response.organization Organization ID returned in the
  *  response or response headers
+ * @property {number} time_to_first_token Duration in milliseconds between when
+ *  the request was issued and the first token was received. This attribute **MUST**
+ *  only be captured for streaming requests.
  */
 module.exports = class LlmChatCompletionSummary extends LlmEvent {
   /**
@@ -47,10 +50,11 @@ module.exports = class LlmChatCompletionSummary extends LlmEvent {
    * @param {number} params.numMsgs Number of messages comprising a
    *  chat completion including system, user, and assistant messages
    * @param {string} params.finishReason Reason the model stopped generating tokens (e.g. "stop")
+   * @param {number} [params.timeOfFirstToken] Timestamp of when the first token was received. Only supply for streaming requests.
    * @param {boolean} [params.error] Set to `true` if an error occurred during creation call, omitted if no error occurred
    */
   constructor({ agent, segment, transaction, vendor, responseModel, requestModel, requestId,
-    responseOrg, temperature, maxTokens, numMsgs = 0, finishReason, error }) {
+    responseOrg, temperature, maxTokens, numMsgs = 0, finishReason, timeOfFirstToken, error }) {
     super({ agent, segment, transaction, vendor, responseModel, requestId, error })
 
     if (requestModel) this['request.model'] = requestModel
@@ -61,6 +65,7 @@ module.exports = class LlmChatCompletionSummary extends LlmEvent {
 
     this['response.number_of_messages'] = numMsgs
     this.timestamp = segment.timer.start
+    if (timeOfFirstToken) this['time_to_first_token'] = timeOfFirstToken - this.timestamp
     this.duration = segment.getDurationInMillis()
   }
 

package/lib/llm-events/google-genai/chat-completion-summary.js

@@ -20,9 +20,10 @@ module.exports = class GoogleGenAiLlmChatCompletionSummary extends LlmChatComple
    * @param {Transaction} params.transaction Current and active transaction
    * @param {object} params.request Google Gen AI request object
    * @param {object} params.response Google Gen AI response object
+   * @param {number} [params.timeOfFirstToken] Timestamp of when the first token was sent, for streaming only.
    * @param {boolean} [params.error] Set to `true` if an error occurred
    */
-  constructor({ agent, segment, transaction, request, response, error }) {
+  constructor({ agent, segment, transaction, request, response, timeOfFirstToken, error }) {
     super({ agent,
       segment,
       transaction,
@@ -32,6 +33,7 @@ module.exports = class GoogleGenAiLlmChatCompletionSummary extends LlmChatComple
       maxTokens: request.config?.maxOutputTokens,
       temperature: request.config?.temperature,
       vendor: 'gemini',
+      timeOfFirstToken,
       error })
 
     let requestMessagesLength = 0

package/lib/llm-events/openai/chat-completion-summary.js

@@ -20,9 +20,10 @@ module.exports = class OpenAiLlmChatCompletionSummary extends LlmChatCompletionS
    * @param {Transaction} params.transaction Current and active transaction.
    * @param {object} params.request OpenAI request object.
    * @param {object} params.response OpenAI response object.
+   * @param {number} [params.timeOfFirstToken] Timestamp of when the first token was sent, for streaming only.
    * @param {boolean} [params.error] Set to `true` if an error occurred, can be omitted in false cases.
    */
-  constructor({ agent, segment, transaction, request, response, error }) {
+  constructor({ agent, segment, transaction, request, response, timeOfFirstToken, error }) {
     super({
       agent,
       segment,
@@ -34,7 +35,8 @@ module.exports = class OpenAiLlmChatCompletionSummary extends LlmChatCompletionS
       requestModel: request?.model,
       requestId: response?.headers?.['x-request-id'],
       temperature: request?.temperature,
-      maxTokens: request?.max_tokens ?? request?.max_output_tokens
+      maxTokens: request?.max_tokens ?? request?.max_output_tokens,
+      timeOfFirstToken
     })
 
     if (request?.input) {

package/lib/subscriber-configs.js

@@ -22,9 +22,13 @@ const subscribers = {
   ...require('./subscribers/mcp-sdk/config'),
   ...require('./subscribers/mysql/config'),
   ...require('./subscribers/mysql2/config'),
+  ...require('./subscribers/nestjs/config'),
+  ...require('./subscribers/node-redis-client/config'),
   ...require('./subscribers/openai/config'),
   ...require('./subscribers/pino/config'),
   ...require('./subscribers/pg/config'),
+  ...require('./subscribers/redis/config'),
+  ...require('./subscribers/redis-client/config'),
   ...require('./subscribers/undici/config')
 }
 

package/lib/subscribers/ai-monitoring/chat.js

@@ -50,12 +50,13 @@ class AiMonitoringChatSubscriber extends AiMonitoringSubscriber {
    * @param {Context} params.ctx active context
    * @param {object} params.request request made to method on a given llm library
    * @param {object} params.response response from method on a given llm library
-   * @param {object} params.err error if present
-   * @param {object} params.metadata used only for langchain events at the moment
-   * @param {Array} params.tags used only for langchain events at the moment
+   * @param {object} [params.err] error if present
+   * @param {number} [params.timeOfFirstToken] Timestamp of when the first streaming token was sent.
+   * @param {object} [params.metadata] used only for langchain events at the moment
+   * @param {Array} [params.tags] used only for langchain events at the moment
    * returns {object} a llm completion summary instance for the given LLM
    */
-  createCompletionSummary({ ctx, request, response, err, metadata, tags }) {
+  createCompletionSummary({ ctx, request, response, err, timeOfFirstToken, metadata, tags }) {
     throw new Error('createCompletionSummary must be implemented by your subscriber')
   }
 
@@ -79,11 +80,12 @@ class AiMonitoringChatSubscriber extends AiMonitoringSubscriber {
    * @param {Context} params.ctx active context
    * @param {object} params.request request made to method on a given llm library
    * @param {object} params.response response from method on a given llm library
-   * @param {object} params.err error if present
+   * @param {object} [params.err] error if present
+   * @param {number} [params.timeOfFirstToken] Timestamp of when the first streaming token was sent.
    * @param {object} params.metadata used only for langchain events at the moment
    * @param {Array} params.tags used only for langchain events at the moment
    */
-  recordChatCompletionEvents({ ctx, request, response, err, metadata = {}, tags = [] }) {
+  recordChatCompletionEvents({ ctx, request, response, err, timeOfFirstToken, metadata = {}, tags = [] }) {
     if (!this.enabled) {
       this.logger.debug('config.ai_monitoring.enabled is set to false, not creating chat completion events.')
       return
@@ -97,7 +99,7 @@ class AiMonitoringChatSubscriber extends AiMonitoringSubscriber {
     // Explicitly end segment to provide consistent duration
     // for both LLM events and the segment
     ctx.segment.end()
-    const completionSummary = this.createCompletionSummary({ ctx, request, response, err, metadata, tags })
+    const completionSummary = this.createCompletionSummary({ ctx, request, response, timeOfFirstToken, err, metadata, tags })
     this.recordEvent({ type: 'LlmChatCompletionSummary', msg: completionSummary })
 
     const messages = this.getMessages({ request, response })

package/lib/subscribers/base.js

@@ -42,6 +42,9 @@ const ArrayPrototypeSplice = makeCall(Array.prototype.splice)
  * @property {AsyncLocalStorage} store The async local storage instance used for context management.
  * @property {number} [callback=null] Position of callback if it needs to be wrapped for instrumentation.
  * -1 means last argument.
+ * @property {string} [callbackKey=null] Property name on the argument at `callback` position
+ * that holds the actual callback function. Used when the callback is not a direct argument but a
+ * property on an object argument (e.g. `args[0].callback` when callback=0 and callbackKey='callback').
  * @property {string[]} [events=[]] Set of tracing channel event names to
  * register handlers for. For any name in the set, a corresponding method
  * must exist on the subscriber instance. The method will be passed the
@@ -88,6 +91,7 @@ class Subscriber {
     this.channel = tracingChannel(this.id)
     this.store = agent.tracer._contextManager._asyncLocalStorage
     this.callback = null
+    this.callbackKey = null
   }
 
   shouldCreateSegment(parent) {
@@ -103,11 +107,23 @@ class Subscriber {
    * This will wrap a callback at a given position and reassign the callback argument to the wrapped one
    *
    * @param {number} position index of the callback, you can specify -1 to be the last
+   * @param {string} [key] property name to look up on the argument at `position` when the
+   * callback is not a direct argument but a property on an object argument
+   * (e.g. `args[0].callback` when position=0 and key='callback')
    * @param {object} context the event passed to the tracing channel hooks
    */
-  traceCallback(position, context) {
+  traceCallback(position, key, context) {
     this.logger.trace('Wrapping the callback at position %s', position)
     const { asyncStart, asyncEnd, error } = this.channel
+
+    const holder = ArrayPrototypeAt(context.arguments, position)
+    const callback = key ? holder?.[key] : holder
+
+    if (typeof callback !== 'function') {
+      this.logger.trace('Callback is not present, not wrapping')
+      return
+    }
+
     function wrappedCallback(err, res) {
       // assigning a boolean to the context so we know that the
       // `error`, `asyncStart`, and `asyncEnd` are coming from the wrapped callback
@@ -133,9 +149,8 @@ class Subscriber {
       })
     }
 
-    const callback = ArrayPrototypeAt(context.arguments, position)
-    if (typeof callback !== 'function') {
-      this.logger.trace('Callback is not present, not wrapping')
+    if (key) {
+      context.arguments[position][key] = wrappedCallback
     } else {
       ArrayPrototypeSplice(context.arguments, position, 1, wrappedCallback)
     }
@@ -254,7 +269,7 @@ class Subscriber {
 
       // only wrap the callback if a subscriber has a callback property defined
       if (this.callback !== null) {
-        this.traceCallback(this.callback, data)
+        this.traceCallback(this.callback, this.callbackKey, data)
       }
       const ctx = this.agent.tracer.getContext()
       if (this.requireActiveTx && !ctx?.transaction?.isActive()) {

package/lib/subscribers/google-genai/generate-content-stream.js

@@ -30,6 +30,7 @@ class GoogleGenAIGenerateContentStreamSubscriber extends GoogleGenAIGenerateCont
     let cachedResult = {}
     let err
     let entireMessage = ''
+    let timeOfFirstToken
     response.next = async function wrappedNext(...nextArgs) {
       let result = {}
       try {
@@ -44,6 +45,7 @@ class GoogleGenAIGenerateContentStreamSubscriber extends GoogleGenAIGenerateCont
         }
 
         if (result?.value?.text) {
+          if (!timeOfFirstToken) timeOfFirstToken = Date.now()
           entireMessage += result.value.text // readonly variable that equates to result.value.candidates[0].content.parts[0].text
         }
       } catch (streamErr) {
@@ -64,6 +66,7 @@ class GoogleGenAIGenerateContentStreamSubscriber extends GoogleGenAIGenerateCont
             ctx,
             request,
             response: cachedResult,
+            timeOfFirstToken,
             err
           })
         }

package/lib/subscribers/google-genai/generate-content.js

@@ -56,10 +56,11 @@ class GoogleGenAIGenerateContentSubscriber extends AiMonitoringChatSubscriber {
    * @param {Context} params.ctx active context
    * @param {object} params.request request made to method
    * @param {object} params.response response from method
+   * @param {number} [params.timeOfFirstToken] Timestamp of when the first streaming token was sent.
    * @param {object} [params.err] error object if present
    * @returns {object} a llm completion summary instance for Google Gen AI
    */
-  createCompletionSummary({ ctx, request, response = {}, err }) {
+  createCompletionSummary({ ctx, request, response = {}, timeOfFirstToken, err }) {
     const { transaction, segment } = ctx
     return new LlmChatCompletionSummary({
       agent: this.agent,
@@ -67,6 +68,7 @@ class GoogleGenAIGenerateContentSubscriber extends AiMonitoringChatSubscriber {
       transaction,
       request,
       response,
+      timeOfFirstToken,
       error: !!err
     })
   }

package/lib/subscribers/nestjs/config.js

@@ -0,0 +1,27 @@
+/*
+ * Copyright 2026 New Relic Corporation. All rights reserved.
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+module.exports = {
+  '@nestjs/core': [
+    {
+      path: './nestjs/instrumentation.js',
+      instrumentations: [
+        {
+          channelName: 'nr_unknown_error',
+          module: {
+            name: '@nestjs/core',
+            versionRange: '>=8.0.0',
+            filePath: 'exceptions/base-exception-filter.js'
+          },
+          functionQuery: {
+            className: 'BaseExceptionFilter',
+            methodName: 'handleUnknownError',
+            kind: 'Sync'
+          }
+        }
+      ]
+    }
+  ]
+}