Language:JapaneseEnglish

databricks Claude Code Plugin

Author
Databricks
Category
Database
Topics
Data Engineering & Analytics · Cloud, Deployment & CI/CD · Databases & Storage
First cataloged
2026-07-09 (UTC)
Explanation last updated
2026-08-19 (UTC)
Source (GitHub) last updated
2026-08-19 (UTC) (8 days ago)

The explanation below is AI-generated. Please verify it against the sources.

According to the README, skills can be installed either through the Databricks CLI's aitools install command, which auto-detects the coding agent in use, or through plugin marketplaces for Claude Code, Cursor, GitHub Copilot, and Codex. The plugin also adds slash commands for workspace setup and a read-only health check, plus hooks that route Databricks-related prompts to the matching skill, prime session context at startup, and surface hints when a command fails due to authentication. Beyond the stable skill set, an experimental collection offers further skills that can be installed individually with an extra flag. The homepage describes Databricks as a platform combining data engineering, app hosting, managed Postgres, model serving, and conversational analytics inside one workspace.

About the service

Databricks is a cloud-based data and AI platform (described on its homepage as a lakehouse) offering workspace services such as Databricks Apps for hosting applications, Lakebase as managed Postgres, Model Serving endpoints for hosted models, Lakeflow Jobs and Pipelines for orchestration, and Genie for conversational data analytics. The skills installed by this plugin guide a coding assistant through using the Databricks CLI, SDK, and these workspace services rather than connecting to an unrelated third-party SaaS.

What you can do with databricks

  • Install stable Databricks skills into Claude Code, Cursor, Codex, GitHub Copilot, and other supported agents via the Databricks CLI or plugin marketplaces
  • Access further skills covering areas such as vector search, built-in AI functions, Unity Catalog, Agent Bricks, ML training, and MLflow evaluation
  • Use slash commands for workspace setup/authentication and a read-only health check (Claude Code, Cursor)
  • Benefit from hooks that route Databricks-related prompts to the right skill, prime session context, and hint at fixes when an authentication error occurs
  • Opt into experimental skills, imported from a separate ai-dev-kit repository, by passing an --experimental flag

Sources

Back to list