From the course: Build with AI: Production-Ready AI Agent Systems on Databricks
Join today to access over 25,600 courses taught by industry experts.
Adding external MCP (web search) - Databricks Tutorial
From the course: Build with AI: Production-Ready AI Agent Systems on Databricks
Adding external MCP (web search)
So far, our agents work with your internal data, assembly manuals, inventory reviews. But sometimes, you need to reach beyond your own systems. Maybe your agent needs to search the web for tutorials. Maybe it should open a support ticket in your customer service platform. Maybe it needs to coordinate with other AI agents across your organization. This is where ecosystem matters. Model Context Protocol, or MCP, is an open standard that lets AI agents connect to external tools and services. Instead of building custom integrations, you plug into a growing ecosystem of MCP servers that provide ready-made capabilities. Databricks has an MCP marketplace with curated servers you can install directly. This enables complex workflows where your agents interact with external systems. The architecture is straightforward. Databricks creates a managed proxy through Unity Catalog connections. You install from Marketplace, authenticate once, and the server is available to all your agents with…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.