Open Source AI Agent Improvement Platform

TraceRoot is the open-source AI agent improvement platform. Find the patterns in your agent's production issues. Ship verified fixes with your coding agent.

Install TraceRoot CLI
npm install -g traceroot-cli

Trusted by
  • Vector Legal
  • Perfectly
  • DINQ
  • Trident

Your agent passed the tests. Production found the gaps.

Wrong answers can look like successful runs. Tool loops waste time and tokens. Prompt and model changes introduce regressions that manual spot checks miss. Turn those failures into clear fixes. Test the changes, improve your agent, and keep learning from production.

Trace production model and tool calls. Detectors use LLM judges to evaluate runs. Cluster judge outputs into signals; background coding agents investigate the traces and code to recommend changes. Apply changes with your coding agent and verify them by comparing offline evals. New production traffic starts the cycle again.
AgentYour applicationTracesModel + tool callsLLM judgesEvaluate runsSignalsSuggested fixesEvalsCompare results

Turn production failures into tested improvements.

Background coding agents investigate production signals using GitHub code in Daytona sandboxes and recommend fixes. Get alerts in Slack, then apply and verify changes with your coding agent and offline evals.

Cluster judge outputs at scale into improvement signals. A background coding agent investigates traces and connected GitHub code in a sandbox, recommending changes down to the commit, file, and line.

Signal analysis with root cause, recommended changes, frequency over time, and affected traces

Built for your coding agent.

Read traces, signals, and workspace context; manage detectors, datasets, evals, dashboards, and alerts—all through an agent-native CLI. Give your coding agent the context to make changes and verify them.

TraceRoot CLI connects your coding agent to Home, traces, dashboards, detectors, evaluations, datasets, signals, and alerts. Connections are bidirectional except for traces, which point toward the agent.
TracesDashboardsDetectorsEvalsDatasetsAlertsHomeSignals
Your coding agent
via TraceRoot CLI

Enterprise-grade security

Built with security in mind. Your data is protected with industry-leading standards.

Simple, transparent pricing

Start free — no credit card needed. BYOK supported on all plans.

Free

Get started with core features

$0per month

Discord support

  • 2 seats
  • 50k events/month included
  • 15-day retention
  • 30 chat runs/month
  • 30 RCA runs/month
  • 100 detector runs/month
  • Slack alerts for detectors
  • BYOK or hosted LLM

Starter

For small teams

$30per month

Discord support

  • Everything in Free
  • Unlimited seats
  • 150k events/month included
  • $4 per 50k overage events
  • 30-day retention
  • 100 chat runs/month
  • $10 per 100 overage chat runs
  • 100 RCA runs/month
  • $10 per 100 overage RCA runs
  • Unlimited detector runs
  • Hosted LLM: token cost × 1.05 on overage

Pro

Popular

For growing teams

$200per month

Discord + Slack support

  • Everything in Starter
  • 90-day retention
  • Higher rate limits

Enterprise

For scaling organizations

Custom

Discord + Slack + SLA support

  • Everything in Pro
  • Custom retention
  • SLA support

Made better, together ❤️

Thanks to the 73+ contributors helping TraceRoot grow!

Frequently asked questions

What is TraceRoot?

TraceRoot is an open-source AI agent improvement platform. Find the patterns in your agent's production issues, then ship verified fixes with your coding agent. Trace agent runs, detect failures, investigate signals, and compare changes with evals, all from the TraceRoot CLI.

How does the improvement loop work?

TraceRoot automatically detects failures and inefficiencies in production agent traffic, then clusters related judge outputs into improvement signals. Your coding agent uses the TraceRoot CLI to investigate those signals and make changes in a local worktree. Compare offline agent eval runs to verify the fix and catch regressions, then repeat the loop with new production traffic.

How do detectors and evals work together?

Detectors run online evaluations on production traces to identify failures and inefficiencies. Patterns across judge outputs surface as improvement signals. Offline agent evals then test your changes against datasets and compare results with a baseline to verify the fix and catch regressions.

Can I use TraceRoot with my coding agent?

Yes. The TraceRoot CLI gives your coding agent access to production traces, detectors, datasets, evals, dashboards, and alerts. Install it with npm install -g traceroot-cli, then run traceroot login.

CLI setup
Does TraceRoot work with my agent stack?

TraceRoot offers Python and JavaScript/TypeScript SDKs, OpenTelemetry-compatible tracing, and integrations with frameworks such as LangGraph, Vercel AI SDK, and CrewAI. Start by sending traces from your application, then add detectors and evals as you need them.

Explore the docs
Is TraceRoot open source? Can I self-host it?

Yes. You can use TraceRoot Cloud or self-host with Docker. Code outside directories named ee is licensed under Apache 2.0; those enterprise directories use a separate Enterprise License.

Self-hosting and source code
Can I start for free?

Yes. TraceRoot Cloud has a free plan with no credit card required. Bring your own model API keys or use hosted models. See pricing for included usage and paid plans.

View pricing