Abhimanyu Sikarwar

AI Grocery Orchestration with MCP

1 min read · 202 words

Overview

A multi-agent grocery ordering system built on Model Context Protocol (MCP). You set dietary preferences in ChatGPT or Claude, the system generates a meal plan, and it orders the ingredients from a store near you.

Tech stack

How it works

The system is a chain of specialized agents coordinated through MCP:

  1. A preference agent captures dietary preferences, restrictions, and goals through conversation in ChatGPT or Claude
  2. A meal planning agent turns those preferences into a weekly plan
  3. An ingredient agent breaks the plan into a grocery list with quantities
  4. An ordering agent finds nearby stores, checks availability, and places the order

Each agent calls external APIs for store inventory, prices, and order placement, while MCP carries structured state from one agent to the next.

Features

Tags: #MCP #LangChain #LangGraph #Python #AI #Agents #OpenAI #Claude