Inflearn LangGraph Agent

What it does

This project provides a comprehensive set of tools and example servers designed to teach the development of AI Agents using LangGraph. It demonstrates how to implement complex agentic workflows including Self-RAG, Corrective RAG, and Multi-Agent systems with RouteLLM, integrated via the Model Context Protocol (MCP).

Tools

• custom_mcp_server: Implements a custom MCP server as part of the educational framework to showcase tool registration.
• rag_retrieval: Tools for preprocessing PDFs and performing retrieval for Agentic RAG.
• human_in_the_loop: Mechanisms for managing agent history and human interaction.
• multi_agent_orchestrator: Tools for routing queries between specialized agents using create_react_agent.

Installation

{
  "mcpServers": {
    "langgraph-agent": {
      "command": "python",
      "args": ["mcp_sse_server.py"]
    }
  }
}

Supported hosts

Confirmed support for Claude Desktop and Cursor via standard MCP configuration.