@mastra/laminar

@mastra/laminar

Laminar observability exporter for Mastra applications.

Exports Mastra spans to Laminar via OTLP/HTTP (protobuf) and supports sending scorer results to Laminar Evaluators.

Installation

npm install @mastra/laminar

Usage

Zero-Config Setup

The exporter automatically reads credentials from environment variables:

# Required
LMNR_PROJECT_API_KEY=lmnr_...

# Optional
LMNR_BASE_URL=https://api.lmnr.ai
LAMINAR_ENDPOINT=https://api.lmnr.ai/v1/traces

import { LaminarExporter } from '@mastra/laminar';

const mastra = new Mastra({
  ...,
  observability: {
    configs: {
      laminar: {
        serviceName: 'my-service',
        exporters: [new LaminarExporter()],
      },
    },
  },
});

Explicit Configuration

import { LaminarExporter } from '@mastra/laminar';

const mastra = new Mastra({
  ...,
  observability: {
    configs: {
      laminar: {
        serviceName: 'my-service',
        exporters: [
          new LaminarExporter({
            apiKey: 'lmnr_...',
            baseUrl: 'https://api.lmnr.ai',
            endpoint: 'https://api.lmnr.ai/v1/traces', // Optional
            realtime: false, // Optional
          }),
        ],
      },
    },
  },
});

Configuration Options

Option	Type	Description
apiKey	string	Laminar project API key. Defaults to LMNR_PROJECT_API_KEY env var
baseUrl	string	Laminar base URL. Defaults to LMNR_BASE_URL env var or https://api.lmnr.ai
endpoint	string	OTLP/HTTP traces endpoint. Defaults to LAMINAR_ENDPOINT env var or /v1/traces
headers	Record<string,string>	Additional OTLP headers
realtime	boolean	Flush after each span for immediate visibility. Defaults to false
disableBatch	boolean	Disable batching (SimpleSpanProcessor). Defaults to false
batchSize	number	Max spans per batch (BatchSpanProcessor). Defaults to 512
timeoutMillis	number	OTLP export timeout (ms). Defaults to 30000

Notes

• The exporter sets Laminar-specific attributes (lmnr.span.*, lmnr.association.properties.*) so traces render correctly in Laminar.