Explore these example applications to see how Model Context Protocol can be used in different scenarios.
A personalized shopping assistant that remembers user preferences, compares products, and makes tailored recommendations. This example demonstrates how to use MCP for:
User preference tracking and personalization
Product catalog integration via tools
Long-term memory for improved recommendations over time
// E-commerce Assistant MCP Context
import { MCPContext } from '@modl/mcp';
const shoppingAssistant = new MCPContext({
systemInstruction: "You are a helpful shopping assistant for a fashion retailer.",
userGoal: "Find stylish, waterproof sneakers under €150",
userProfile: {
name: "Alex",
style: ["minimalist", "neutral", "casual-elegant"],
shoeSize: "EU 42",
pastPurchases: ["Nike Air Max", "Common Projects", "Vans Old Skool"],
preferredPriceRange: "100-150 EUR"
},
tools: [
{
name: "searchProducts",
description: "Search the product catalog",
parameters: {
query: "string",
filters: "object",
limit: "number"
}
},
{
name: "getProductDetails",
description: "Get detailed information about a product",
parameters: {
productId: "string"
}
},
{
name: "checkInventory",
description: "Check if a product is in stock",
parameters: {
productId: "string",
size: "string"
}
}
],
memory: {
shortTerm: [],
longTerm: {
stylePreferences: "modern minimalist, neutral colors",
purchaseBehavior: "researches thoroughly before buying"
}
}
});
export default shoppingAssistant;A customer support agent that can access knowledge bases, user history, and support tools to resolve issues efficiently. This example demonstrates:
Knowledge base integration with RAG
Customer history and context awareness
Support ticket management tools
A coding assistant that understands your codebase, development patterns, and can access APIs and documentation. This example demonstrates:
Code repository integration
Documentation search and retrieval
Code generation with project-specific context