Skip to Content
View all events

Transforming APIs into Agents with MCP

Published by O'Reilly Media, Inc.

Intermediate content levelIntermediate

Build intelligent agents from existing APIs

What you’ll learn and how you can apply it

  • Learn best practices for how to AI-enable your personal or public HTTP REST APIs to become AI agents using MCP
  • Get hands-on experience using MCP to have Q&A with custom data sources and data repositories
  • Understand the best practices for creating AI agents with MCP and avoiding the antipatterns
  • Create AI agents with off-the-shelf hardware to convert dumb appliances into smart appliances for your home

Course description

The Model Context Protocol is the open standard for creating AI agent clients and AI agent servers with official SDK support in Python, Java, JavaScript, C#, and other languages. Created by Anthropic in 2024, it has emerged as the de-facto protocol for AI interoperability between any compliant MCP client and MCP server.

Bruce Hopkins shows you how to take your existing APIs, data sources, and services and turn them into AI agents that conform to the MCP standard. By adopting the MCP, your end users can easily interact with your services using a standard MCP client application. You’ll also learn how to create custom MCP servers that work with off-the-shelf hardware units to convert home appliances to AI-enabled.

This live event is for you because...

  • You’re a software developer who needs to create AI-based applications and services that connect to corporate, confidential, or personal data sources.
  • You’re familiar with ChatGPT, but you want to understand the best practices for creating AI agents that conform to the MCP standard.
  • You’re a technical manager who wants to understand the capabilities and benefits of AI agents and the MCP standard.

Prerequisites

  • An API key for OpenAI or Anthropic Claude
  • An IDE configured with your preferred programming language
  • A basic understanding of programming in any major programming language such as Python, Java 21, JavaScript, or C#
  • Basic experience using ChatGPT or Claude

Recommended follow-up:

Schedule

The time frames are only estimates and may vary according to how the class is progressing.

Why is the Model Context Protocol (MCP) great for external APIs? (60 minutes)

  • Presentation: Getting started with MCP clients and servers
  • Demonstration: Creating a basic MCP server from an external API
  • Q&A

Advanced MCP concepts—resources and sampling (60 minutes)

  • Presentation: Using custom URIs and MCP sampling
  • Q&A
  • Break

MCP AI agent orchestration (50 minutes)

  • Presentation: The best practices for converting APIs to MCP server
  • Demonstration: Multi-agent orchestration
  • Q&A
  • Break

Using MCP to create AI agents for your home (60 mins)

  • Presentation: Creating MCP servers for remote control of your home
  • Demonstration: Asking how much power is being used by appliances
  • Q&A

Next steps (10 minutes)

Your Instructor

  • Bruce Hopkins

    Bruce Hopkins is a technical writer, an AI expert, an Intel Software Innovator for AI, and an Oracle Java Champion. He’s also the author of ChatGPT for Java and the coauthor of Beginning ChatGPT for Python.

    Xlinksearch

Skill covered

Web APIs