Everything you need to know about MCP for building connected AI agents
AI agents are no longer limited to answering questions. They’re beginning to take action, execute tasks, and collaborate across systems. However, to do that effectively, they need more than traditional APIs. They need structure, discoverability, and a shared language. That’s where the Model Context Protocol (MCP) comes in.
Previously aired on June 4th, 2025
Register to watch the replay right away.
What we covered
Aired on June 4th, 2025, this webinar with Daniel Petro (Developer Advocate) and Cameron Booth (Developer Education) explained the Model Context Protocol (MCP) for building connected AI agents. Agents are starting to take action, execute tasks and collaborate across systems, and to do that well they need more than traditional APIs: they need structure, discoverability and a shared language. MCP gives them that.
The session covered the principles behind MCP for agentic workflows, use cases for building MCP, the Xano tools you can use to build your own MCP server and bring LLMs into your workflows, and a walkthrough of setting up a real-world agent on Xano.
Key takeaways
- MCP gives agents structure, discoverability and a shared language that traditional APIs lack.
- You can build your own MCP server on Xano with the Xano MCP Builder.
- Part 2, MCP tools for AI agents, breaks down the tools, APIs, functions and database behind the XanoBnB Agent demo.
In this webinar, we cover:
- The principles behind MCP for AI agentic workflows
- Use cases for building MCP
- Xano tools you can use to build your own MCP server and integrate LLMs into your workflows
- A walkthrough of setting up a real-world agent using Xano