> ## Documentation Index
> Fetch the complete documentation index at: https://docs.bryel.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# TypeScript

> Trace a Node/TypeScript LLM app — the Vercel AI SDK, any OpenTelemetry app, or a provider SDK directly.

bryel is OpenTelemetry-native. On the server you have three ways in, all shipping
the same OpenInference traces — models, tokens, cost, tool calls, full trajectory.

<CardGroup cols={3}>
  <Card title="Vercel AI SDK" icon="triangle" href="#vercel-ai-sdk">One processor + a per-call flag.</Card>
  <Card title="Any OTel app" icon="radio" href="#any-opentelemetry-app">Ship spans you already emit.</Card>
  <Card title="Provider SDKs" icon="boxes" href="#provider-sdks">OpenAI / Anthropic / LangChain.</Card>
</CardGroup>

## Vercel AI SDK

`@bryel/vercel` maps the AI SDK's OpenTelemetry spans into OpenInference and ships
them to bryel. You never wrap your model calls.

<Steps>
  <Step title="Install">
    <CodeGroup>
      ```bash bun theme={"theme":{"light":"github-light","dark":"github-dark"}}
      bun add @bryel/vercel @vercel/otel
      ```

      ```bash npm theme={"theme":{"light":"github-light","dark":"github-dark"}}
      npm i @bryel/vercel @vercel/otel
      ```

      ```bash pnpm theme={"theme":{"light":"github-light","dark":"github-dark"}}
      pnpm add @bryel/vercel @vercel/otel
      ```

      ```bash yarn theme={"theme":{"light":"github-light","dark":"github-dark"}}
      yarn add @bryel/vercel @vercel/otel
      ```
    </CodeGroup>
  </Step>

  <Step title="Register the processor once">
    In `instrumentation.ts` — Next.js calls this on startup.

    ```ts instrumentation.ts theme={"theme":{"light":"github-light","dark":"github-dark"}}
    import { registerOTel } from "@vercel/otel";
    import { bryelVercelProcessor } from "@bryel/vercel";

    export function register() {
      registerOTel({
        serviceName: "my-app",
        spanProcessors: [bryelVercelProcessor({ apiKey: process.env.BRYEL_KEY! })],
      });
    }
    ```
  </Step>

  <Step title="Enable telemetry per call">
    ```ts theme={"theme":{"light":"github-light","dark":"github-dark"}}
    const result = streamText({
      model: openai("gpt-5"),
      prompt,
      experimental_telemetry: {
        isEnabled: true,
        metadata: { sessionId, userId },
      },
    });
    ```
  </Step>
</Steps>

<Warning>A streamed span finalizes only once the stream is fully consumed — always drain it.</Warning>

## Any OpenTelemetry app

If your app already emits [OpenInference](/concepts) spans, add
`bryelSpanProcessor` to your provider — no mapping.

<CodeGroup>
  ```bash bun theme={"theme":{"light":"github-light","dark":"github-dark"}}
  bun add @bryel/sdk-core @opentelemetry/sdk-trace-base
  ```

  ```bash npm theme={"theme":{"light":"github-light","dark":"github-dark"}}
  npm i @bryel/sdk-core @opentelemetry/sdk-trace-base
  ```

  ```bash pnpm theme={"theme":{"light":"github-light","dark":"github-dark"}}
  pnpm add @bryel/sdk-core @opentelemetry/sdk-trace-base
  ```

  ```bash yarn theme={"theme":{"light":"github-light","dark":"github-dark"}}
  yarn add @bryel/sdk-core @opentelemetry/sdk-trace-base
  ```
</CodeGroup>

```ts theme={"theme":{"light":"github-light","dark":"github-dark"}}
import { NodeSDK } from "@opentelemetry/sdk-node";
import { bryelSpanProcessor } from "@bryel/sdk-core";

new NodeSDK({
  spanProcessors: [bryelSpanProcessor({ apiKey: process.env.BRYEL_KEY })],
}).start();
```

If your spans aren't OpenInference yet, supply a `SpanMapper` (this is exactly
what `@bryel/vercel` does):

```ts theme={"theme":{"light":"github-light","dark":"github-dark"}}
import { buildSpanProcessor, type SpanMapper } from "@bryel/sdk-core";

const mapper: SpanMapper = {
  name: "my-framework",
  shouldMap: (span) => span.name.startsWith("my."),
  map: (span) => ({ /* OpenInference attributes to merge */ }),
};

const processor = buildSpanProcessor(mapper, { apiKey: process.env.BRYEL_KEY });
```

## Provider SDKs

Calling OpenAI / Anthropic / LangChain directly? Add any OpenTelemetry
instrumentation that emits GenAI spans, then ship them with `bryelSpanProcessor`
— no call-site changes:

```ts theme={"theme":{"light":"github-light","dark":"github-dark"}}
import { NodeSDK } from "@opentelemetry/sdk-node";
import { bryelSpanProcessor } from "@bryel/sdk-core";
// + the OpenTelemetry instrumentation for your provider SDK

new NodeSDK({
  instrumentations: [/* your GenAI instrumentation */],
  spanProcessors: [bryelSpanProcessor({ apiKey: process.env.BRYEL_KEY })],
}).start();
```

## Grouping and feedback

The `metadata` you pass groups and joins your data:

<ParamField path="sessionId" type="string">Groups turns into one conversation.</ParamField>
<ParamField path="userId" type="string">The end user behind the run.</ParamField>

<ParamField path="messageId" type="string">
  A per-turn id **you mint** (`crypto.randomUUID()`). bryel stamps it as
  `bryel.interaction.id` so [feedback](/guides/feedback) joins back to the turn.
</ParamField>

## API

<ParamField path="apiKey" type="string" required>Project API key (`bk_…`).</ParamField>
<ParamField path="endpoint" type="string" default="https://ingest.eu.bryel.ai/v1/traces">OTLP/HTTP endpoint. Override for self-hosting.</ParamField>
<ParamField path="headers" type="Record<string, string>">Extra headers, merged after auth.</ParamField>

`bryelVercelProcessor` (AI SDK, with mapping), `bryelSpanProcessor` (no mapping),
and `buildSpanProcessor(mapper, options)` (custom mapping) all take these options
and return an OTel `SpanProcessor`.