Building an Agentic AI Assistant with LangGraph and MCP

Описание и применение

Создан агентический ассистент, использующий LangGraph и MCP для координации инструментов, поддерживающий многократные диалоги и безопасный доступ к файловой системе.

Задача

Create an AI assistant that can perform multi-step tasks by selecting and executing tools, rather than just generating text responses.

Результат

The assistant can perform multi-step tasks, maintain conversation context, and safely execute file operations within a sandbox, demonstrating the benefits of combining LangGraph, MCP, and tool orchestration.

Рабочий процесс

• Interpret user request
• Select appropriate tool via MCP discovery
• Execute tool call (web-search or filesystem)
• Receive and process tool output
• Decide if further actions are needed
• Return final response to user

Доказательства

The assistant supports multi-turn conversations and can use a sandboxed filesystem tool and a web-search tool.

Источник

The Model Context Protocol (MCP) defines a standard way for applications to communicate with tool-providing servers and discover available capabilities.

Источник

My filesystem tool restricts access to a designated sandbox directory and resolves requested paths against that directory.

Источник