The next generation of AI observability

Understand exactly what your AI agents are doing.

Rackle is an open-source telemetry platform built for multi-agent systems. Capture real-time token costs, debug LLM responses, and trace every single step in your workflow.

View on GitHub

Everything you need to ship reliable AI.

Stop flying blind. Rackle provides the visibility you need to build trust in your autonomous systems.

Cost & Analytics

Automatically calculate exactly how much each LLM call costs based on token usage. View rich charts of your daily spend and latency distributions.

Waterfall Traces

Break down complex agent runs into a visual waterfall. Inspect exactly what data went into the prompt and what the model outputted at every step.

Real-time WebSockets

Watch your agents think in real-time. Rackle uses WebSockets to instantly stream agent activity to your dashboard without a single page refresh.

Integrate in two lines of code.

Our lightweight TypeScript SDK drops perfectly into any Node.js, Bun, or Edge environment. Generate a secure API key and start tracing instantly.

import { Tracer } from "@rackle-labs/sdk"; // 1. Initialize the tracerconst tracer = new Tracer({  apiKey: process.env.RACKLE_API_KEY}); // 2. Wrap your agent executionconst run = await tracer.startRun({ agentName: "Customer-Bot" }); await run.log({  type: "llm_call",  model: "gpt-4o",  tokens: 350,  input: "Help me reset my password."}); await run.end({ status: "completed" });