Foglamp
Log inStart free
SDK

Two lines. Every AI SDK call, instrumented.

Register Foglamp once with AI SDK v7 and every generateText and streamText call is traced and costed. Using v4-v6? The version-aware wrapper gives you the same traces.

Start freeSee pricing
model.ts
import { registerTelemetry } from "ai";import { foglamp } from "foglamp";registerTelemetry(foglamp());
Drop-in

Built for the code you already wrote.

Foglamp uses AI SDK v7's native telemetry integration and a version-aware wrapper for v4-v6. Your prompts, models, tools, and control flow stay exactly as they are.

  • Works with Vercel AI SDK v4 through v7
  • No prompt or business-logic changes
  • TypeScript-first, fully typed
Start free
Install
npm i foglamp
Zero overhead

Telemetry that stays out of the hot path.

Spans are batched and flushed asynchronously, so instrumentation never blocks a response. Set one environment variable and you're streaming traces to your dashboard.

  • Async, batched span export, no added latency
  • Single FOGLAMP_API_KEY to authenticate
  • Open standards under the hood
See the dashboard
Captured per call
100%
calls traced
0ms
added latency
2
lines to add

Your agents are running in the fog

What they cost, when they break, what they say. One prompt turns the light on.

Start free