#3535th of 48 in Gateways and observability
traceroot
LLM tracing, detectors and evals that feed fixes back to your coding agent
- Stars
- 1.4k
- License
- custom license
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- TypeScript
504 on 21 Apr 20261,387 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Traceroot records model calls, tool calls and responses from agent applications, with inputs, outputs, latency and cost. Detectors flag production runs that miss defined behavior, and flagged runs are clustered into signals. A background agent investigates traces and connected GitHub code in a sandbox, and offline evals check the changes. Python and TypeScript SDKs cover frameworks such as LangChain, CrewAI and the OpenAI Agents SDK.
Who it is for: Teams running LLM agents in production who want tracing and evals
Strengths
- Closes the loop: tracing, detectors, signals and dataset evals in one tool
- Python and TypeScript SDKs with integrations for 14 agent frameworks and 5 providers
- CLI setup command has a coding agent add instrumentation and waits for first trace
- Self-hostable with Docker Compose via make prod-lite; hosted cloud also available
Weaknesses
- Mixed licensing: Apache 2.0 outside ee directories, Enterprise License inside them
- Several provider integrations (Gemini, Mistral) are Python only
- Signals feature relies on a sandboxed coding agent and GitHub connection
- Self-hosting needs Git, Docker Compose and Make; resource requirements not documented
What it needs
- no GPU
- Docker + Compose
- Needs Docker Compose, Make, Git
- Models: OpenAI, Anthropic, Google Gemini, Mistral, OpenRouter
- port 3000
Quick answers
Is traceroot open source?
Its code is public on GitHub under its own license terms; read them before you use it at work.
Can you run traceroot with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does traceroot need a GPU?
No, it runs without one.
Also in Gateways and observability
See all 48| Rank | Project | Score |
|---|---|---|
| 1 | OmniRouteLLM gateway routing one OpenAI-compatible endpoint across 372 providers | 96 out of 100 |
| 2 | headroomCompresses tool outputs, logs and history before they reach the LLM | 92 out of 100 |
| 3 | new-apiGateway that unifies LLM providers behind OpenAI, Anthropic and Gemini APIs | 91 out of 100 |
| #4 | MLflowTracing, evaluation and prompt registry for LLM apps, plus ML tracking | 90 out of 100 |
| #5 | LiteLLMOpenAI-format gateway and Python SDK for calling 100+ LLM providers | 87 out of 100 |