Custom MCP Server Template

This solutions accelerator demonstrates how to deploy existing MCP servers, or your own custom MCP servers, on Databricks Apps. It provides a template you can adapt to your own custom MCP server implementation, or merge with an existing MCP server implementation, to support deployment on Databricks apps.

After installing the solutions accelerator, view the code in the examples/custom-server directory to access the custom server Databricks Apps template.

If you’re interested in using MCP to access data on Databricks, use Databricks-managed MCP servers ([AWS](https://docs.databricks.com/aws/en/generative-ai/agent-framework/mcp) | [Azure](https://learn.microsoft.com/en-us/azure/databricks/generative-ai/agent-framework/mcp) | [GCP](https://docs.databricks.com/gcp/en/generative-ai/agent-framework/mcp)). These servers are available out of the box and maintained by Databricks, with no need to host your own.