AI Agents with MCP

Chapter 8. Connecting Clients and Servers: MCP Transports

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 8th 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].

The Model Context Protocol’s transport layer is what links clients and servers together, enabling them to send and receives messages between each other. Understanding the transport layer from the JSON-RPC level where it interacts directly with streaming data up through the SDK abstractions and managing real-life connections and custom transports is crucial to effectively debugging and securing your own MCP projects. This chapter explores the MCP transport layer in-depth, covering the JSON-RPC foundations of the layer, the connection lifecycle from the transport’s perspective, and the transport types available in MCP by default as well as custom transports. Moving from the low-level design to the high-level implementation, we will dig into practical issues, like implementing your own custom transport, security threat modeling and mitigation strategies, and debugging ...