feat(arize): add Browserbase × Arize AX integration

[Nancy-Chauhan]

Adds a Browserbase × Arize AX integration example under examples/integrations/arize/.

Use Browserbase as a load_page tool for a tool-calling agent, traced end to end with Arize AX via OpenTelemetry + OpenInference. The model calls the Browserbase tool to browse a page and answer; the run is exported to AX as a CHAIN → LLM → TOOL trace with tokens and cost.

This makes Browserbase a tool an agent can call.

Files

• src/agent.ts — tool-calling agent with manual CHAIN/TOOL spans
• src/instrumentation.ts — Arize AX / OTel setup + OpenAI auto-instrumentation
• src/register-prompt.ts — optional: publish the load_page tool to the AX Prompt Hub via @arizeai/ax-client

Verification — verified end to end against live Arize AX, Browserbase, and OpenAI (gpt-5.4): correct CHAIN → LLM → TOOL trace on the happy path, ERROR-marked spans on the failure path, and idempotent prompt registration.

Type of Change

• Bug fix (non-breaking change that fixes an issue)
• New feature (non-breaking change that adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Performance improvement
• Refactoring (no functional changes)

---

Note

Low Risk
Additive documentation and example code only; no changes to shared packages or production services. Demo uses external API keys and sends traces to Arize when configured.

Overview
Adds a new Arize AX example under examples/integrations/arize/ and lists it in the root monorepo README.

The sample is a Node/TypeScript tool-calling agent that exposes Browserbase Fetch API as a load_page tool (markdown page content, no full browser session). An OpenAI loop answers user prompts (CLI or default question) by calling that tool when needed.

Observability is wired via OpenTelemetry + OpenInference: instrumentation.ts exports OTLP traces to Arize (https://otlp.arize.com/v1/traces) with space/API headers, auto-instruments OpenAI for LLM spans, and agent.ts adds manual CHAIN / TOOL spans around the agent run and each fetch. Spans are flushed on shutdown via provider.shutdown().

Also included: .env.example, integration README, package.json / lockfile, and tsconfig.json for npm start / npm run build.

^{Reviewed by Cursor Bugbot for commit d2999b1. Bugbot is set up for automated code reviews on this repo. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit d85b139. Configure here.}