BraintrustExporter

Sends Tracing data to Braintrust for eval and observability.

Constructor

new BraintrustExporter(config: BraintrustExporterConfig)

BraintrustExporterConfig

interface BraintrustExporterConfig extends BaseExporterConfig {
  apiKey?: string
  endpoint?: string
  projectName?: string
  braintrustLogger?: Logger<true>
  currentSpan?: () => Span | undefined
  tuningParameters?: Record<string, any>
}

Extends BaseExporterConfig, which includes:

• logger?: IMastraLogger - Logger instance
• logLevel?: LogLevel | 'debug' | 'info' | 'warn' | 'error' - Log level (default: INFO)

Methods

exportTracingEvent

async exportTracingEvent(event: TracingEvent): Promise<void>

Exports a tracing event to Braintrust.

export

async export(spans: ReadOnlySpan[]): Promise<void>

Batch exports spans to Braintrust.

flush

async flush(): Promise<void>

Force flushes any buffered spans to Braintrust without shutting down the exporter. Useful in serverless environments where you need to ensure spans are exported before the runtime terminates.

shutdown

async shutdown(): Promise<void>

Flushes pending data and shuts down the client.

Usage

Zero-Config (using environment variables)

import { BraintrustExporter } from '@mastra/braintrust'

// Reads from BRAINTRUST_API_KEY, BRAINTRUST_ENDPOINT
const exporter = new BraintrustExporter()

Explicit Configuration

import { BraintrustExporter } from '@mastra/braintrust'

const exporter = new BraintrustExporter({
  apiKey: process.env.BRAINTRUST_API_KEY,
  projectName: 'my-ai-project',
})

Span type mapping

Span Type	Braintrust Type
MODEL_GENERATION	llm
MODEL_CHUNK	llm
TOOL_CALL	tool
MCP_TOOL_CALL	tool
PROVIDER_TOOL_CALL	tool
WORKFLOW_CONDITIONAL_EVAL	function
WORKFLOW_WAIT_EVENT	function
All others	task