create-tools

Create Databricks Genie spaces, Vector Search indexes, UC functions, and HTTP connections.

4|Updated May 9, 2026
One-click install
npx skills add https://github.com/victorlou/housing-assistant --skill create-tools-victorlou
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-tools
Source: https://github.com/victorlou/housing-assistant/tree/main/app/app-templates/.claude/skills/create-tools
Command: npx skills add https://github.com/victorlou/housing-assistant --skill create-tools-victorlou

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the friction of manually setting up Databricks resources that an agent must call, so you can go from “I need a tool” to a working, permissioned integration faster.

Core Features & Use Cases

  • Creates Databricks tool-backed resources: Genie spaces for structured Q&A, Vector Search indexes for RAG, UC functions for custom logic, and UC HTTP connections for external MCP servers.
  • Guides end-to-end wiring: Ensures you create the right resource first, then hand off to the add-tools and deploy skills to connect it to the agent.
  • Covers the typical “what do I need first?” setup path: Helps when resources don’t exist yet or when users ask what prerequisites are required before adding tools.

Quick Start

Ask your agent to create the Genie space, vector search index, UC function, or UC connection needed for your tool-backed workflow before you add-tools and deploy.

Frequently Asked Questions about create-tools

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I provision Databricks resources for an AI agent to call?

To provision Databricks resources for an agent, you create managed resources like Genie spaces for structured Q&A, Vector Search indexes for RAG, or UC functions for custom logic. This establishes the permissioned endpoints your agent needs.

What Databricks prerequisites do I need before adding tools to my agent?

Before adding tools to your agent, you must create the underlying Databricks resources first, such as Genie rooms, Vector Search endpoints, or Unity Catalog HTTP connections. Once these managed resources exist, you can wire MCP and tool permissions.

How do I set up a Unity Catalog HTTP connection for external MCP servers?

Setting up a Unity Catalog HTTP connection for external MCP servers involves creating a UC connection resource to bridge the external tool. After creating the UC connection, you align it with subsequent add-tools and deploy steps to enable federated tool access.

When should I create a Vector Search index versus a Genie space for my agent?

Create a Vector Search index for semantic retrieval and RAG workloads, while a Genie space is for structured querying and Q&A. Both are Databricks tool resources that require creating the managed resource first before wiring MCP permissions.

Does this handle wiring MCP permissions after creating the Databricks tool resources?

This creates the Databricks tool resources and aligns them with subsequent MCP permission wiring. It guides the end-to-end setup path by creating the managed resource first, then handing off to the add-tools and deploy steps to connect it to the agent.