From the course: Build with AI: Agentic Applications with LlamaIndex and MCP

Unlock this course with a free trial

Join today to access over 25,200 courses taught by industry experts.

Interacting with your own agents served as an MCP via Claude Code

Interacting with your own agents served as an MCP via Claude Code - LlamaIndex Tutorial

From the course: Build with AI: Agentic Applications with LlamaIndex and MCP

Start my 1-month free trial Buy for my team

Interacting with your own agents served as an MCP via Claude Code

“

So far, you've built your own agent workflows, you've learned about MCP, you've used MCP server tools in tool calling agents, and now it's time for the final piece of the puzzle. Instead of just existing MCP servers and tools, this time we'll expose our very own agent workflow as an MCP server itself. This is great because it means that once you have an agent that you are happy with, you can expose it to other applications and MCP clients, essentially giving those applications additional capabilities. An example of an MCP client that you can really easily use in your own code spaces alongside me is actually called Cloud Code. In this lesson, we'll see how we can serve our own e-commerce agent as an MCP server that Cloud Code can access, and essentially we'll be using Cloud Code as a chat assistant that can access our e-commerce agent. Again, we're not starting with a clean slate here. You can see that this e-commerce agent is the exact same as the e-commerce agent that was able to do…

Contents