​
Model Context Protocol (MCP)

​
What is MCP?

The Model Context Protocol (MCP) is an open protocol that standardizes how applications provide context to large language models (LLMs). Think of MCP like a USB-C port for AI applications - just as USB-C provides a standardized way to connect your devices to various peripherals, MCP provides a standardized way to connect AI models to different data sources and tools.

MCP helps you build agents and complex workflows on top of LLMs by providing:

  • A standardized way to connect AI models to data sources and tools
  • The flexibility to switch between LLM providers and vendors
  • Best practices for securing your data within your infrastructure

​
Key Features

  • Structured Communication: Defines how applications should communicate with language models
  • Context Management: Efficiently manages the limited context windows of LLMs
  • Tool Integration: Standardizes how LLMs interact with external tools and functions
  • Security: Helps prevent prompt injection attacks and unauthorized access
  • State Management: Maintains conversation history and relevant context

​
Infactory MCP Integration

Infactory provides a TypeScript package for integrating with the Model Context Protocol:

MCP SDK

@infactory/infactory-mcp

Our MCP implementation allows you to:

  • Connect your applications to various LLMs through a standardized interface
  • Provide structured context to enhance AI responses
  • Integrate with Infactory’s data sources and tools
  • Build agentic workflows with improved context management

​
Getting Started

Install the MCP package:

npm install @infactory/infactory-mcp
# or
yarn add @infactory/infactory-mcp
# or
pnpm add @infactory/infactory-mcp

Basic usage:

import { InfactoryMCP } from '@infactory/infactory-mcp';

// Initialize the MCP client
const mcp = new InfactoryMCP({
  apiKey: 'your-api-key'
});

// Use MCP to provide context to an LLM
const response = await mcp.sendMessage({
  content: "What insights can you provide from my sales data?",
  context: {
    dataSources: ["sales_q1_2025"]
  }
});

​
Learn More

For more information about the Model Context Protocol: