Observability for BeeAI (TypeScript) with Opik

BeeAI is an agent framework designed to simplify the development of AI agents with a focus on simplicity and performance. It provides a clean API for building agents with built-in support for tool usage, conversation management, and extensible architecture.

BeeAI's primary advantage is its lightweight design that makes it easy to create and deploy AI agents without unnecessary complexity, while maintaining powerful capabilities for production use.

Getting started

To use the BeeAI integration with Opik, you will need to have BeeAI and the required OpenTelemetry packages installed.

Installation

Option 1: Using npm

Bash

npm install beeai-framework@0.1.13 @ai-sdk/openai @arizeai/openinference-instrumentation-beeai @opentelemetry/sdk-node dotenv

Option 2: Using yarn

Bash

yarn add beeai-framework@0.1.13 @ai-sdk/openai @arizeai/openinference-instrumentation-beeai @opentelemetry/sdk-node dotenv

Requirements

Node.js ≥ 18
BeeAI Framework (beeai-framework)
OpenInference Instrumentation for BeeAI (@arizeai/openinference-instrumentation-beeai)
OpenTelemetry SDK for Node.js (@opentelemetry/sdk-node)

Environment configuration

Configure your environment variables based on your Opik deployment:

Intent: Route BeeAI OpenInference spans to Opik through OTLP. No breaking changes: existing BeeAI setups continue to work; these settings are only required to send traces to Opik.

Applies when: You run BeeAI (TypeScript) with @arizeai/openinference-instrumentation-beeai.

Required fields:

OTEL_EXPORTER_OTLP_ENDPOINT
OTEL_EXPORTER_OTLP_HEADERS (Authorization, Comet-Workspace for Cloud/Enterprise)

Optional fields:

projectName in OTEL_EXPORTER_OTLP_HEADERS (recommended)
additional OpenTelemetry env vars for sampling/resource labels

Minimal valid input:

endpoint for your deployment mode
matching headers for auth and routing

wordWrap

# Your LLM API key
export OPENAI_API_KEY="your-openai-api-key"

# Opik configuration
export OTEL_EXPORTER_OTLP_ENDPOINT=https://www.comet.com/opik/api/v1/private/otel
export OTEL_EXPORTER_OTLP_HEADERS='Authorization=<your-api-key>,Comet-Workspace=default'

To log the traces to a specific project, you can add the projectName parameter to the OTEL_EXPORTER_OTLP_HEADERS environment variable:

wordWrap

export OTEL_EXPORTER_OTLP_HEADERS='Authorization=<your-api-key>,Comet-Workspace=default,projectName=<your-project-name>'

You can also update the Comet-Workspace parameter to a different value if you would like to log the data to a different workspace.

wordWrap

# Your LLM API key
export OPENAI_API_KEY="your-openai-api-key"

# Opik configuration
export OTEL_EXPORTER_OTLP_ENDPOINT=https://<comet-deployment-url>/opik/api/v1/private/otel
export OTEL_EXPORTER_OTLP_HEADERS='Authorization=<your-api-key>,Comet-Workspace=default'

To log the traces to a specific project, you can add the projectName parameter to the OTEL_EXPORTER_OTLP_HEADERS environment variable:

wordWrap

export OTEL_EXPORTER_OTLP_HEADERS='Authorization=<your-api-key>,Comet-Workspace=default,projectName=<your-project-name>'

You can also update the Comet-Workspace parameter to a different value if you would like to log the data to a different workspace.

Bash

# Your LLM API key
export OPENAI_API_KEY="your-openai-api-key"

# Opik configuration
export OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:5173/api/v1/private/otel
export OTEL_EXPORTER_OTLP_HEADERS='projectName=<your-project-name>'

Using Opik with BeeAI

Set up OpenTelemetry instrumentation for BeeAI:

TypeScript

import "dotenv/config";
import { NodeSDK } from "@opentelemetry/sdk-node";
import { OTLPTraceExporter } from "@opentelemetry/exporter-trace-otlp-http";
import { BeeAIInstrumentation } from "@arizeai/openinference-instrumentation-beeai";
import * as beeaiFramework from "beeai-framework";

// Initialize BeeAI Instrumentation
const beeAIInstrumentation = new BeeAIInstrumentation();

// Configure and start the OpenTelemetry SDK
const sdk = new NodeSDK({
  traceExporter: new OTLPTraceExporter(),
  instrumentations: [beeAIInstrumentation],
});
sdk.start();

// Manually patch BeeAI framework (required for trace collection)
beeAIInstrumentation.manuallyInstrument(beeaiFramework);

// Now you can use BeeAI as normal
import { ReActAgent } from "beeai-framework/agents/react";
import { OpenAIChatModel } from "beeai-framework/adapters/openai/backend/chat";
import { WikipediaTool } from "beeai-framework/tools/search/wikipedia";
import { OpenMeteoTool } from "beeai-framework/tools/weather/openmeteo";
import { TokenMemory } from "beeai-framework/memory";

// Initialize the OpenAI language model
const llm = new OpenAIChatModel("gpt-5-nano", {
  temperature: 0.7,
});

// Create tools for the agent
const tools = [
  new WikipediaTool(),
  new OpenMeteoTool(),
];

// Create a ReAct agent with memory
const agent = new ReActAgent({
  llm,
  tools,
  memory: new TokenMemory({ llm }),
});

// Run the agent
async function main() {
  const response = await agent.run({
    prompt: "I'm planning a trip to Barcelona, Spain. Can you research key attractions and landmarks I should visit, and also tell me what the current weather conditions are like there?",
    execution: {
      maxRetriesPerStep: 3,
      totalMaxRetries: 10,
      maxIterations: 5,
    },
  });
  
  console.log("Agent Response:", response.result.text);
  return response;
}

// Run the example
main();

Validation

Start the app and run one agent request.
Confirm OTLP export succeeds.
Verify the trace in Opik under the expected workspace/project.

Source references

Further improvements

If you have any questions or suggestions for improving the BeeAI integration, please open an issue on our GitHub repository.