← Back
LangChain
LangChain releases Skills for AI coding agents; boosts Claude Code performance from 29% to 95%
· featuresdkintegrationopen-source · blog.langchain.com ↗

What Are Skills?

Skills are curated instructions, scripts, and resources designed to improve coding agent performance in specialized domains. Rather than overwhelming agents with all available tools at once (which historically degrades performance), skills are dynamically loaded through progressive disclosure—agents only retrieve relevant skills when needed.

LangChain Skills Release

LangChain is releasing its first set of 11 skills organized into three categories:

  • LangChain Skills: Guidance on create_agent(), middleware, and tool patterns for foundational agent work
  • LangGraph Skills: Instructions for working with LangGraph primitives, including Human-in-the-Loop, durable execution, and more
  • Deep Agents Skills: Documentation for the open-source Deep Agents package and its prebuilt middleware and FileSystem utilities

Key Performance Improvements

When tested with Claude Code (Sonnet 4.6), LangChain Skills delivered exceptional results:

  • Without Skills: 25% pass rate on LangChain/LangGraph/Deep Agents tasks
  • With Skills: 95% pass rate on the same tasks

This represents a 280% improvement in coding agent effectiveness on LangChain ecosystem tasks.

Installation and Usage

Skills are portable and shareable markdown files available through the langchain-skills GitHub repository. Install them locally or globally using:

npx skills add langchain-ai/langchain-skills --skill '*' --yes

To bind skills to Claude Code specifically:

npx skills add langchain-ai/langchain-skills --agent claude-code --skill '*' --yes --global

Next Steps

LangChain plans to continue expanding Skills coverage as new capabilities are added to the ecosystem. The team is also releasing complementary LangSmith skills for the evaluation and monitoring platform.