OpenLIT
An open-source observability and evaluation platform that traces LLM calls, tool use, prompts, cost, and coding-agent sessions using OpenTelemetry.
OpenLIT is an open-source observability and evaluation platform for AI applications and coding agents, built on OpenTelemetry. After starting its Docker Compose stack and adding one line to initialize its SDK, it automatically instruments supported LLM providers, frameworks, and vector databases, tracing every LLM call, tool invocation, prompt, agent step, token, cost, error, and evaluation to a local dashboard.
A separate CLI instruments coding agents such as Claude Code, Cursor, and Codex, giving an OpenTelemetry-native view of coding-agent sessions in a dedicated dashboard so a developer can see what an agent actually did during a task. Telemetry is exported over the OTLP protocol to a configurable endpoint, and a doctor command checks that the CLI's instrumentation installed correctly.
It is written in TypeScript and released under the Apache License 2.0.