mcpField
D
databricks-config

databricks-config

github.com/databricks-solutionsCloud & DevOps2.0k37 installsUpdated 6d ago
Repository

About databricks-config

databricks-config is a skill for guiding agents through Databricks workspace management via an MCP tool. It covers checking, listing, switching, and logging into workspaces, and notes that switching lasts only for the current session.

Key features

  1. 01Check current workspace status
  2. 02List available workspaces
  3. 03Switch workspace configuration
  4. 04Log in or connect to a workspace
  5. 05Present operation results

Use cases

  1. 01Verify the connected workspace
  2. 02Find and switch to a workspace
  3. 03Authenticate with a workspace

MCP support

The skill requires the Databricks manage_workspace MCP tool for workspace operations and instructs agents to load it through ToolSearch first.

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/databricks-solutions/ai-dev-kit --skill databricks-config

After installing, type/databricks-configin Claude Code to confirm it loaded.

Details

From the project README

Use the manage_workspace MCP tool for all workspace operations. Do NOT edit ~/.databrickscfg, use Bash, or use the Databricks CLI.

Steps

  1. Call ToolSearch with query select:mcp__databricks__manage_workspace to load the tool.

  2. Map user intent to action:

    • status / which workspace / current → action="status"
    • list / available workspaces → action="list"
    • switch to X → call list first to find the profile name, then action="switch", profile="<name>" (or host="<url>" if a URL was given)
    • login / connect / authenticate → action="login", host="<url>"

  3. Call mcp__databricks__manage_workspace with the action and any parameters.

  4. Present the result. For status/switch/login: show host, profile, username. For list: formatted table with the active profile marked.

Note: The switch is session-scoped — it resets on MCP server restart. For permanent profile setup, use databricks auth login -p <profile> and update ~/.databrickscfg with cluster_id or serverless_compute_id = auto.

Comments

Markdown supported