+ P Evaluate for Vercel AI SDK agents
Capture every Vercel AI SDK run through its own telemetry integration interface — the same lifecycle hooks the SDK uses for its own instrumentation, not a bolted-on tracer.
Prefactor evaluates your Vercel AI SDK agents — score outcome quality against the captured spans, track drift by comparing custom spans across versions and environments, and classify every run's risk on two axes: data sensitivity and action consequence.
How to add evaluate to Vercel AI SDK
Install the package
Add @prefactor/ai to your project.
Create the middleware
Build it from your config — API URL, token, and agent.
Spans flow to Prefactor
LLM calls, tool invocations, and agent spans arrive as trace data.
// npm install @prefactor/ai
import { init } from "@prefactor/ai";
const prefactor = init({
transportType: "http",
httpConfig: {
apiUrl: process.env.PREFACTOR_API_URL!,
apiToken: process.env.PREFACTOR_API_TOKEN!,
agentIdentifier: "1.0.0",
},
});How the Vercel AI SDK integration actually works
- Registers as a Telemetry integration — via registerTelemetry() globally, or passed per-call through the telemetry option (the current name for what generateText/streamText still accept as the deprecated experimental_telemetry alias).
- The SDK's own executeTool and executeLanguageModelCall wrappers are the real enforcement point: each wrapper receives the actual tool or model call as a function before it runs, and decides whether to call it, throttle it, or block it — the same mechanism the SDK uses internally to compose multiple integrations, not a side channel.
- Beyond the auto-captured lifecycle events, use withSpan to record any custom step you define — an API call, a quality check, a business action.
What Prefactor captures for Vercel AI SDK agents
Outcome quality
Score agent outputs against your own criteria, task by task, using the continuous run history as the record to measure against.
Drift detection
Compare custom spans for the same operation across versions and environments — a model update or prompt change shows up as a difference between spans, before it shows up as a quality drop.
Risk classification
Every run scored on data sensitivity and action consequence, weighted into a Low / Medium / High / Critical classification.
Data tagging
Tag PII and other sensitive fields once in the schema; every run carrying that tag becomes searchable — the record enforcement acts on.
Trace anything — not just LLM calls
The SDK captures LLM, tool, and agent spans automatically. With withSpan you wrap any operation in your own span type — an API call, a database query, a quality check, a business action — each with its own payload and schema. It all flows into the same Observe, Evaluate, and cost views.
import { withSpan } from "@prefactor/core";
// Wrap ANY operation in a span you define — an API call, a quality
// check, a business action — with its own spanType, inputs and schema
await withSpan(
{
name: "research competitor",
spanType: "research_competitor",
inputs: { competitor },
},
async () => {
const results = await search(competitor); // your tool / API calls
return summarize(results); // captured as one span
},
);Nest spans to capture business-level actions, and start with permissive schemas you tighten over time. Instrumentation strategy →
An example run, span by span
Illustrative — a single Vercel AI SDK run as nested spans.
Illustrative example.
Manual logging vs DIY vs Prefactor
| Capability | Manual | DIY OpenTelemetry | Prefactor |
|---|---|---|---|
| LLM / tool / agent spans | Hand-rolled | ✓ build it | ✓ via the SDK |
| Token usage captured per call | — | Build it | ✓ |
| Configurable capture & sampling | — | Partial | ✓ |
| Hosted Admin UI (agents, instances) | — | — | ✓ |
| Risk profiles & audit trail | — | — | ✓ |
Vercel AI SDK evaluate — FAQ
Does evaluation reuse the same Vercel AI SDK integration?+
Yes — evaluation runs on the spans the SDK already records for Vercel AI SDK, so there is one integration, not a separate one for evaluation.
What is quality drift and how is it detected for Vercel AI SDK agents?+
Drift is a change in behaviour over time — after a model update, a prompt change, or a data shift. Because a custom span records the same Vercel AI SDK operation every time it runs, this week's spans can be compared directly against last week's, or against a different version or environment.
How is risk scored for Vercel AI SDK runs?+
On two axes — data sensitivity (what kind of data the run touched) and action consequence (what the agent did with it) — combined via configurable weights into a Low/Medium/High/Critical classification per run.
Can I tag and find PII across Vercel AI SDK runs?+
Yes — tag the fields that matter once in your agent's data schema, and every run carrying that tag becomes searchable across your whole Vercel AI SDK deployment, not just the run it was tagged in.
Do I need a dedicated package for Vercel AI SDK?+
Vercel AI SDK has a native package (@prefactor/ai) that registers as a Telemetry integration; you can also instrument it today with the framework-agnostic prefactor-core SDK.
What does Prefactor capture from Vercel AI SDK?+
Prefactor hooks the SDK's own telemetry lifecycle — onStart/onEnd, onStepStart/onStepEnd, onLanguageModelCallStart/End, and onToolExecutionStart/End — as structured, timestamped spans, so every Vercel AI SDK run is captured as trace data you can reconstruct, search and export end to end.
Does Prefactor add latency or change how Vercel AI SDK runs?+
No. Observability capture is designed to stay off your agent's critical path, so it doesn't alter your Vercel AI SDK logic or your users' responses. The only part that acts inline is the optional runtime guardrails you enable per agent — by design, so a high-risk or low-confidence action can be held for human approval before it executes.
Can I evaluate agents built with Vercel AI SDK and catch regressions?+
Yes. Once runs are captured, eval suites score quality and groundedness on real traffic, drift detection flags behaviour changes after deployment, and versioned eval history catches regressions before they ship — the observe → evaluate → improve loop applied to your Vercel AI SDK agents.
Keep going
Evaluate for other frameworks
LangChain →CrewAI →LangGraph →OpenAI Agents SDK →Claude Agent SDK →Microsoft AutoGen →See it on your Vercel AI SDK agents
Book a 15-minute setup and our team gets you evaluate in production.