Databricks AI Dev Kit is a toolkit for building coding agents on the Databricks platform provided by Field Engineering.
The toolkit addresses the challenge of setting up and managing AI coding environments for developers working with Databricks. It provides skills—reusable components that extend agent capabilities—and an MCP server that integrates with coding agents like Claude and Cursor. The skills have been moved into official Databricks AI Tools, which are now installed and maintained through the Databricks CLI, though the repository continues to host the MCP server and a Builder App for experimental tools and environment setup guidance.
Developers should adopt this toolkit if they are building coding agents on Databricks and want pre-built integrations for common tasks. The project suits teams using agents like Claude or Cursor who need to work with Databricks resources. The official skills are now delivered through Databricks AI Tools and work with any agent, including first-class integration with Genie One. The repository itself remains useful as a guide for setting up AI coding environments and as a source for experimental tools from Field Engineering, though the bundled skill files are deprecated in favor of the official Databricks installation.
Almost all open issues are raised by outside users rather than the core team, indicating a substantial base of adopters reporting real-world use. Maintainers typically respond to new issues and pull requests within a few days. Work in the issue tracker is dominated by enhancement requests, skills-related discussions, and MCP server topics.
