Laminar sessionId and userId metadata keys
Traces in Laminar can be grouped into sessions or by user ID using reserved metadata keys sessionId and userId. Example: telemetry: { metadata: { sessionId: 'session-123', userId: 'user-123' } }
AI SDK · Providers · all subjects
laminar/options
10 notes, read out of this brain and free to use. Each one was extracted from a source and is re-checked against its exam.
LaminarAiSdkTelemetry constructor options
LaminarAiSdkTelemetry constructor accepts options including: laminarOptions (with projectApiKey, baseUrl, httpPort, grpcPort for self-hosting), recordInputs (boolean, default true), recordOutputs (boolean, default true), and createStepSpan (boolean, default false).
Laminar self-hosting options code example
registerTelemetry(new LaminarAiSdkTelemetry({ laminarOptions: { projectApiKey: process.env.LMNR_PROJECT_API_KEY, baseUrl: "http://localhost", httpPort: 8000, grpcPort: 8001, }, }));
Disable Laminar input/output recording
Pass recordInputs: false and/or recordOutputs: false to LaminarAiSdkTelemetry constructor to disable recording of inputs or outputs. Both default to true.
Laminar createStepSpan option
AI SDK telemetry integrations emit step spans for every agent step. By default, Laminar ignores these spans. Set createStepSpan: true in LaminarAiSdkTelemetry constructor options to create a span for every agent step.
Override Laminar span name with functionId
To override the default span name in Laminar, set the functionId inside the telemetry option when calling AI SDK functions. Example: telemetry: { functionId: 'poem-writer' }
Laminar observe wrapper for nested spans
Use the observe() wrapper from @lmnr-ai/lmnr to trace custom functions and create nested spans. The observe function takes a configuration object with name, an async function, and optional arguments that will be traced as inputs. The return value is traced as the span's output.
Laminar observe wrapper code example
import { observe } from '@lmnr-ai/lmnr'; const result = await observe( { name: 'poem writer' }, async (topic: string, mood: string) => { const { text } = await generateText({ model: openai('gpt-5.4-mini'), prompt: `Write a poem about ${topic} in ${mood} mood.`, }); return text; }, 'Laminar flow', 'happy', );
Laminar metadata in telemetry option
Set metadata at the trace level using the telemetry option. Metadata contains key-value pairs and can be used to filter traces. Example: telemetry: { metadata: { 'my-key': 'my-value', 'another-key': 'another-value' } }
Laminar tags metadata key
Tags is a reserved metadata key in Laminar that can be used to add tags to a span. Tags can subsequently be used to filter traces in Laminar. Example: telemetry: { metadata: { tags: ['fallback-model', 'api-handler'] } }