add-tools

Add tools to a Databricks agent and grant permissions in databricks.yml.

Updated Mar 15, 2026
One-click install
npx skills add https://github.com/sumitsaxena-git/databricks-app --skill add-tools-sumitsaxena-git
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: add-tools
Source: https://github.com/sumitsaxena-git/databricks-app/tree/main/agent-openai-agents-sdk/.claude/skills/add-tools
Command: npx skills add https://github.com/sumitsaxena-git/databricks-app --skill add-tools-sumitsaxena-git

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Add tools to your Databricks agent and grant required permissions in databricks.yml to enable new MCP servers, Genie spaces, vector search, and Unity Catalog functions without runtime errors.

Core Features & Use Cases

  • Add MCP servers and grant app access: connect new back-end services to your agent and expose them to the Databricks runtime.
  • Enable resources on demand: configure Genie spaces, vector-search indexes, and UC functions as part of the agent's toolset.
  • Operational safety & deployment readiness: ensure permissions are granted in databricks.yml before deployment to prevent permission errors.
  • Real-world use case: a team adds a new MCP-backed memory for a memory-enabled agent and needs immediate tool availability after deployment.

Quick Start

Tell the agent to add an MCP server and grant the necessary permissions in databricks.yml.

Frequently Asked Questions about add-tools

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

FAQPage Schema
How do I add MCP servers to a Databricks agent and grant permissions?

To add MCP servers to a Databricks agent and grant permissions, you configure the required resources and access levels directly in databricks.yml. This exposes the back-end services to the Databricks runtime and prevents deployment errors.

Why does my Databricks agent encounter runtime permission errors after adding new tools?

Runtime permission errors occur when required access levels are missing from databricks.yml. You must explicitly grant permissions for new tools like Genie spaces or vector search before deployment to satisfy Databricks App template sequencing requirements.

Can I configure Unity Catalog functions and vector search as tools for my Databricks agent?

Yes, you can configure Unity Catalog functions and vector search indexes as agent tools. Adding these resources to databricks.yml ensures they are enabled on demand and accessible by the Databricks runtime without encountering access errors.

What is the best way to enable Genie spaces in a Databricks App template?

The best way to enable Genie spaces in a Databricks App template is to add the resource to your agent's toolset and update databricks.yml. This ensures operational safety by granting necessary permissions before deployment.

When do I need to update databricks.yml for my agent's tools?

You need to update databricks.yml when integrating new MCP servers, Genie spaces, vector search indexes, or Unity Catalog functions. Updating this file ensures the Databricks runtime grants required permissions and avoids deployment sequencing failures.