LangChain Documentation

Comprehensive documentation for LangChain and LangGraph frameworks. Covers building LLM applications, agents, and stateful workflows in Python and JavaScript/TypeScript.

Quick Reference

When to use

Use this skill when the user asks about:

• •Building LLM applications with LangChain
• •Creating agents and multi-agent systems
• •Building stateful workflows with LangGraph
• •Integrating with chat models, embeddings, vector stores, and tools
• •Python or JavaScript/TypeScript implementations
• •Prompt engineering, context management, and structured outputs
• •RAG (Retrieval-Augmented Generation) patterns
• •Memory management and state persistence
• •Middleware, guardrails, and observability
• •Deploying LangChain applications

How to find information

1. •First, read TOC.md for complete file listing organized by directory
2. •Choose the appropriate language section (python/ or javascript/)
3. •Pick the framework (langchain/ for core, langgraph/ for stateful workflows)
4. •Navigate to specific integrations, guides, or API references

TOC.md contains all 1700+ files organized by directory - always check it first.

Documentation Organization

• •langchain/: Core framework concepts, agents, tools, models, memory
• •langgraph/: Stateful workflows, graph API, persistence, interrupts
• •python/: 1290 Python files including extensive integrations
• •javascript/: 306 JavaScript/TypeScript files with provider integrations
• •integrations/: Provider-specific documentation (chat models, embeddings, vector stores, etc.)
• •migrate/: Version migration guides
• •concepts/: Architectural concepts
• •reference/: API reference documentation

Use the Quick Reference table above to quickly access common entry points, then explore related files for detailed information.