IRLFirst physical meetup — Bengaluru, Sat May 23, 4PM · RSVP on Luma
HomeToolsMCPHow It WorksStoriesPhilosophyCommunityArchitectureStar on GitHub
All Tools
L
MonitoringFreeOpen Source

LAMINAR

Open-source observability for AI agents

Apache-2.0

ABOUT

AI agents are complex, long-running, and non-deterministic — traditional monitoring tools designed for deterministic API calls and CRUD apps fall short. Developers can't trace agentic decision loops, evaluate tool-call quality, or detect regressions across versions. Laminar solves this with OpenTelemetry-native tracing purpose-built for agentic workflows, natural-language monitoring signals, and integrated evals that let you ship reliable agents 10x faster.

INSTALL
pip install lmnr npm install @lmnr/lmnr

INTEGRATION GUIDE

1. Trace complex multi-step agentic workflows with OpenTelemetry-native instrumentation in one line of code 2. Run evaluations on agent outputs and compare results across different model versions or configurations 3. Define custom monitoring signals with natural language to track issues, logical errors, and agent behavior 4. Query all traces, metrics, and events with a built-in SQL editor for deep debugging and analysis 5. Build custom dashboards to visualize agent performance and reliability over time at scale

TAGS

observabilitymonitoringai-agentstracingevaluationsopen-source
Laminar — AI Tool | Agentic AI For Good