Langsmith Sdk For Llm Tracing And Evaluation

---
name: "LangSmith SDK for LLM Tracing and Evaluation"
slug: "langsmith-sdk-for-llm-tracing-and-evaluation"
description: "LangSmith provides tracing, evaluation, and debugging workflows for LLM applications. This skill is useful when an agent team needs structured observability around prompts, chains, tool calls, datasets, and eval runs across multiple frameworks."
github_stars: 843
verification: "security_reviewed"
source: "https://github.com/langchain-ai/langsmith-sdk"
author: "langchain-ai"
category: "Monitoring & Alerts"
framework: "Multi-Framework"
tool_ecosystem:
  github_repo: "langchain-ai/langsmith-sdk"
  github_stars: 843
---

# LangSmith SDK for LLM Tracing and Evaluation

LangSmith provides tracing, evaluation, and debugging workflows for LLM applications. This skill is useful when an agent team needs structured observability around prompts, chains, tool calls, datasets, and eval runs across multiple frameworks.

## Installation

Use the upstream install or setup path that matches your environment:
- pip install -U langsmith
- pnpm add langsmith

Requirements and caveats from upstream:
- [![Release Notes](https://img.shields.io/github/release/langchain-ai/langsmith-sdk?logo=python)](https://github.com/langchain-ai/langsmith-sdk/releases)
- [![Python Downloads](https://img.shields.io/pypi/dm/langsmith)](https://pypi.org/project/langsmith/)
- This repository contains the Python and Javascript SDK's for interacting with the [LangSmith platform](https://smith.langchain.com/). Please see [LangSmith Documentation](https://docs.smith.langchain.com/)

Basic usage or getting-started notes:
- bash
- export LANGSMITH_TRACING=true
- export LANGSMITH_API_KEY=ls_...

- Source: https://github.com/langchain-ai/langsmith-sdk
- Extracted from upstream docs: https://raw.githubusercontent.com/langchain-ai/langsmith-sdk/HEAD/README.md

## Source

- [Agent Skill Exchange](https://agentskillexchange.com/skills/langsmith-sdk-for-llm-tracing-and-evaluation/)