# langchain-ai/langsmith-sdk

LangSmith Client SDK Implementations

Repository: https://github.com/langchain-ai/langsmith-sdk
Canonical: https://ross.abutalabs.com/products/langsmith-sdk
Homepage: https://docs.smith.langchain.com/
Language: Python
License: MIT
License Family: permissive
Topics: evaluation, language-model, observability
Last push: 2026-09-03T00:07:05+00:00

## Health v2 (maintenance only)
Score: 93/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 100, release rhythm 88, longevity 85
- inputs: {"age_days": 1191, "days_push": 0, "days_rel": 1, "gap_med": 1.0, "n_releases_24m": 301}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1044, forks 288 (observed 2026-09-03T02:15:14.285131+00:00)

## What it is
Official Python and JavaScript/TypeScript client SDKs for the LangSmith platform, enabling tracing, debugging, evaluation, and monitoring of LLM applications and agents. It provides decorators and wrappers (e.g., traceable, wrap_openai) that capture end-to-end traces of LLM calls with minimal code changes.

## Use cases
- trace llm calls in production
- debug ai agent failures
- evaluate llm app quality with datasets
- monitor chatbot performance
- instrument openai calls with tracing
- collect user feedback on model outputs
- build evaluation datasets from production traces

## When to choose
- you need observability for LLM or agent applications
- you use LangChain or LangGraph and want one-variable tracing setup
- you want to evaluate and benchmark LLM outputs against datasets
- you need production monitoring and alerting for AI apps

## When to avoid
- you want fully self-hosted open-source observability without the LangSmith platform
- your project has nothing to do with LLMs or agents
- you need a general-purpose APM tool for non-LLM services

## Facets
- artifact type: library
- maturity: active
- function: monitoring, tracing, llm-inference, agent-framework, sdk, testing, analytics
- domain: large-language-models, developer-tools, monitoring, machine-learning
- platform: python, cross-platform
- tags: langsmith, llm-observability, tracing, llm-evaluation, langchain, openai-wrappers, agent-debugging, ai-agents, nodejs

## Member repositories
- langchain-ai/langsmith-sdk (main) score 93

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:14.285131+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:05:09.308418+00:00, confidence not recorded.
  - readme: https://github.com/langchain-ai/langsmith-sdk (fetched 2026-09-03T02:15:14.285131+00:00, sha a28ad2f34709)
  - homepage: https://docs.smith.langchain.com/ (fetched 2026-08-29T13:06:06.761494+00:00, sha 69d22a6c5e5e)
  - site_page: https://docs.langchain.com/ (fetched 2026-08-29T13:06:06.763969+00:00, sha e6cbe67a291c)
  - site_page: https://docs.langchain.com/langsmith/observability-quickstart (fetched 2026-08-29T13:06:06.765807+00:00, sha e2173e54f36b)
  - site_page: https://docs.langchain.com/langsmith/integrations (fetched 2026-08-29T13:06:06.767742+00:00, sha 211ddfcc915d)
  - site_page: https://docs.langchain.com/use-these-docs (fetched 2026-08-29T13:06:06.769312+00:00, sha 5ab7220bb555)
- Data as of 2026-08-30T08:39:29.467469+00:00.