CheckOps home

Tools directory

L

CheckOps

Laminar

Laminar is an Apache-2.0 open-source observability platform for AI agents with OpenTelemetry-native tracing, evaluations and natural-language signals that flag agent failures.

Visit website

Latest public signals

Not refreshed yet. Refresh reads the public site and, when the entry has a GitHub repository, the latest release.

Visit Website

Introduction

Laminar is an open-source observability platform purpose-built for AI agents, backed by Y Combinator. Its README highlights an OpenTelemetry-native tracing SDK that needs one line of code to trace the Vercel AI SDK, Browser Use, Stagehand, LangChain, OpenAI, Anthropic, Gemini and more.

A distinctive feature is describing agent behaviour you want to track in plain English, such as ‘agent is stuck in a loop’; Laminar reads agent runs and flags matching cases. It also supports evaluations over datasets, SQL queries over trace data, and browser session recordings alongside traces for browser agents.

Laminar can be self-hosted with Docker Compose or used as a hosted service. In the quality loop it helps teams find and reproduce agent failures in production and turn them into evaluations.

The source code is at lmnr-ai/lmnr on GitHub under the Apache-2.0 licence, with about 3,500 stars; the most recent commit was on 10 November 2026.

Features

  1. OpenTelemetry tracing: Auto-instrumentation for many SDKs.
  2. Natural-language signals: Describe behaviours to detect.
  3. Evaluations: Datasets and scored runs.
  4. Browser agent replays: Session recordings with traces.
  5. SQL over traces: Query trace data.
  6. Self-hosting: Docker Compose deployment.

Pros

  1. Agent-focused: Designed around multi-step agents.
  2. Free to use: Open source under the Apache-2.0 licence, so the core tool costs nothing to use.
  3. Actively maintained: The repository had commits as recent as 10 November 2026.
  4. Community: About 3,500 GitHub stars.

Cons

  1. Young project: Features change quickly.
  2. Self-hosting effort: Several services to run.
  3. Overlap: Similar to Langfuse and Phoenix in parts.
from lmnr import Laminar
Laminar.initialize(project_api_key="...")
# subsequent OpenAI / Anthropic / LangChain calls are traced automatically

Resources:

Pricing

Laminar is open source under Apache-2.0 and free to self-host. A hosted version is available; see laminar.sh for current plans.

  1. Teams building browser or coding agents.
  2. Developers wanting open-source agent tracing.

Conclusion

Laminar targets the specific debugging needs of agents.

Its plain-English signals are worth testing on real failure modes.

On CheckOps, Laminar is filed under the Production monitoring & observability and Release & CI gates stages. Related tools listed here include Langfuse, Helicone, OpenLLMetry, AgentOps, Arize Phoenix and Opik. When comparing them, weigh the licence, the languages and platforms each supports, whether it is self-hosted or managed, and how cleanly it slots into your existing CI pipeline. Details on this page were checked against the official website, documentation and repository on 11 November 2026.

Alternatives

Other tools on CheckOps that cover similar ground. Open a page, or compare one with Laminar side by side.