Agent Skills for LangChain, LangGraph, LangSmith & Deep Agents

langchain-agent-skills is a collection of agent-optimized skills for AI coding assistants working in the LangChain ecosystem. Each skill is a self-contained package with a SKILL.md entry point plus optional scripts, references and templates, covering the development lifecycle from project setup to production deployment, monitoring and debugging.

The nine skills include initializing and configuring LangGraph projects, multi-agent coordination patterns such as supervisor, router and orchestrator-worker, state schema and checkpoint management, retry and human-in-the-loop error handling, testing and evaluation patterns including trajectory evaluation and dataset evaluation, a trace analyzer for debugging and performance work, deploying and monitoring LangGraph applications in the cloud, hybrid or standalone, and setting up and configuring Deep Agents projects in Python or JavaScript, including their planning and to-do tooling.

A further skill documents how to create and maintain additional skills in the same repository.