Trace and evaluate LLM application behavior with Langtrace
Instrument Python or TypeScript agent applications with OpenTelemetry traces, metrics, and evaluation views for debugging production LLM workflows.
npx skills add agentskillexchange/skills --skill trace-and-evaluate-llm-application-behavior-with-langtrace
langtrace-python-sdk with pip or @langtrase/typescript-sdk with npm, initialize Langtrace before LLM/framework imports, then run agent traffic and inspect traces in Langtrace.Use Langtrace when an operator needs an agent application to emit reviewable traces before debugging latency, cost, retrieval, tool-call, or evaluation failures. The workflow is to install the Python or TypeScript SDK, initialize it before model and framework imports, run realistic agent traffic, then inspect traces and metrics in Langtrace. The scope is observability instrumentation and review for LLM workflows, not a generic monitoring platform card or a listing for every Langtrace SDK.
What this skill actually does
Inputs and prerequisites: Python or Node.js runtime, Langtrace API key or self-hosted Langtrace instance.
Setup notes: Install langtrace-python-sdk with pip or @langtrase/typescript-sdk with npm, initialize Langtrace before LLM/framework imports, then run agent traffic and inspect traces in Langtrace.
Source and verification boundary: use https://github.com/Scale3-Labs/langtrace as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.