Skip to main content
Veritrix is the observability platform built for AI agents. Whether you’re running a single assistant or a complex multi-agent pipeline, Veritrix gives you a complete, navigable timeline of every LLM call, tool execution, and agent handoff — so when something breaks, you know exactly why in minutes, not hours.

Introduction

Learn what Veritrix is, the problems it solves, and who it’s built for — from engineers to compliance teams.

Quickstart

Install the SDK, add two lines of code, and get your first trace inside ten seconds.

OpenAI Agents SDK

Native instrumentation for OpenAI’s Agents SDK — zero configuration, full trace coverage from day one.

Trace Waterfall

See every span across every agent on one timeline — nested handoffs, parallel calls, and retries all in view.

​
Get up and running in three steps

You can go from zero to your first trace in under a minute. Here’s all it takes.
1

Install the SDK

Add Veritrix to your Python environment with a single command.
2

Instrument your agent

Import trace and call trace.init at the top of your agent entry point. Two lines — that’s it.
3

View traces in the dashboard

Run your agent as normal, then open the Veritrix dashboard to see a live waterfall of every span, token count, and payload.Dashboard: app.veritrix.xyz