AI Agents with MCP

Chapter 4. MCP Clients: Advanced Use and Best Practices

A Note for Early Release Readers

With Early Release ebooks, you get books in their earliest form—the author’s raw and unedited content as they write—so you can take advantage of these technologies long before the official release of these titles.

This will be the 4th chapter of the final book. Supplemental material (code examples, exercises, etc.) is available for download at https://github.com/kylestratis/ai_agents_mcp_examples

If you’d like to be actively involved in reviewing and commenting on this draft, please reach out to the editor at [e-mail].

In Chapter 3 you learned how to build a client that supported the server primitives: tools, resources, and prompts. You learned a standard architecture for building clients and how to manage stdio and streamable HTTP connections. To round out your client-building knowledge, this chapter will show you how to provide the server with client-based capabilities, how to use OAuth to allow servers to authorize your client users, how to build for different models and model providers, how to discover servers with the MCP Registry, techniques for connecting to multiple servers, managing your context window with context engineering, and some best practices for building your client.

Providing MCP Client Capabilities

This section will introduce the use of callback functions to provide client capabilities to the connected MCP server. There are three major capabilities that a ...