🔍 TraceRoot helps engineers debug production issues 10× faster using AI-powered analysis of traces, logs, and code context.
- Visit the TraceRoot website to start debugging your production issues.
- Explore the TraceRoot documentation to get started with the TraceRoot library.
- Join our Discord community to learn more and discuss on AI Agent for observability, debugging, tracing and root cause analysis.
TraceRoot accelerates the debugging process with AI-powered insights. It integrates seamlessly into your development workflow, providing real-time trace and log analysis, code context understanding, and intelligent assistance.
| Feature | Description |
|---|---|
| 🚀 Ease of Use | Get started with TraceRoot in minutes with our simple setup process |
| 🤖 LLM Flexibility | Bring your own model (OpenAI, Anthropic, local LLMs) for AI-powered debugging |
| 🌐 Distributed Services | Cross-platform support with distributed setup for enterprise-scale debugging |
| 💻 AI Debugging Interface | Cursor-like interface specialized for debugging with AI assistance |
| 🔌 Integration Support | Native integration with GitHub, Notion, Slack, and other tools |
The fastest and most reliable way to start with TraceRoot is by signing up for free to TraceRoot Cloud for a 7-day trial. You’ll get:
- 100k traces + logs storage with 30-day retention
- 1M LLM tokens
- AI agent with chat mode
Usually new features will be available in TraceRoot Cloud first, and then they will be released to the self-hosted version.
If you want to self-host TraceRoot, you can deploy a starter instance in one line on Linux with Docker:
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/traceroot-ai/traceroot/HEAD/bin/deploy-starter)"Open source deployments should scale to a certain point and may not cover all the features, thus we recommend migrating to TraceRoot Cloud.
In general the open source version will start the UI at http://localhost:3000 and the API at http://localhost:8000.
If you don't want to use Docker, please refer to the DEVELOPMENT.md for more details to setup the environment manually.
Whether you're using TraceRoot Cloud or our open source version, it's required to use our SDK:
| Language | Repository |
|---|---|
| Python | traceroot-sdk |
| JavaScript/TypeScript | traceroot-sdk-ts |
For more details on SDK usage and examples, please check out this Quickstart.
Here is an overview for our AI Agent Framework:
Please checkout the README.md in the rest/agent directory for more details.
If you find our exploratory TraceRoot useful in your research, please consider citing:
@article{traceroot_2025,
title={TraceRoot Is All You Need for Debugging and Tracing},
author={Zecheng Zhang and Xinwei He},
year = {2025},
publisher = {GitHub},
url = {https://github.com/traceroot-ai/traceroot}
}Thanks to all our contributors for helping make TraceRoot better!