langsmith-evaluator

Official

Build scalable, code-driven LangSmith evaluators.

Authorlangchain-ai
Version1.0.0
Installs0

System Documentation

What problem does it solve?

LangSmith-based evaluation pipelines for agent outputs with reproducible, code-driven evaluators that ensure consistent scoring across tasks and datasets.

Core Features & Use Cases

  • Define deterministic run evaluators in Python or TypeScript to validate outputs, trajectories, and schema conformity.
  • Upload offline evaluators attached to datasets or online evaluators attached to projects for real-time quality checks.
  • Align evaluation with LangSmith workflows, enabling structured scoring and debugging insights.

Quick Start

Capture agent outputs, write a small code evaluator in Python or TypeScript, and upload it to LangSmith to validate results against your dataset or project.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: langsmith-evaluator
Download link: https://github.com/langchain-ai/skills-benchmarks/archive/main.zip#langsmith-evaluator

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.