skills.rest
AnanyaDBJ
AnanyaDBJCOMMUNITY
Explore all skills by this author →
GitHub

supervisor-api

Execute agent loops and tool calls server-side via the Databricks Supervisor API.

6|10|Updated Mar 23, 2026
Search Tags:#AI integration#agent management#Databricks#server-side processing#Supervisor API
One-click install
npx skills add https://github.com/AnanyaDBJ/databricks-ai-workshops --skill supervisor-api-ananyadbj
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: supervisor-api
Source: https://github.com/AnanyaDBJ/databricks-ai-workshops/tree/main/medium/.claude/skills/supervisor-api
Command: npx skills add https://github.com/AnanyaDBJ/databricks-ai-workshops --skill supervisor-api-ananyadbj

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires databricks-sdk, openai-python, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill facilitates the replacement of the client-side agent loop with the Databricks Supervisor API, enabling server-side management of tool calls and loops, and simplifying the integration of Genie spaces, UC functions, and other tools.

Core Features & Use Cases

  • Server-Side Agent Loop: Handles tool calls and loops on the Databricks platform.
  • Hosted Tools Integration: Supports integration with Genie spaces, UC functions, Knowledge Assistants, and MCP servers as hosted tools.
  • Runtime Model Selection: Allows selection of models at runtime for dynamic tool usage.
  • Scalability: Efficiently handles complex multi-tool workflows, ideal for production environments.
  • Use Case: Use this Skill to automate and streamline the management of complex workflows that involve multiple tools and require server-side execution.

Quick Start

Declare hosted tools and use the Databricks Supervisor API to manage your agent loop server-side.

Frequently Asked Questions about supervisor-api

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

FAQPage Schema
How do I run an agent loop server-side in Databricks instead of on the client?▼

To run server-side agent loops in Databricks, you replace the client-side loop with the Supervisor API to manage tool calls and loops directly on the platform for scalable execution.

Can I integrate Genie spaces and UC functions as hosted tools for AI agents?▼

You can integrate Genie spaces, UC functions, Knowledge Assistants, and MCP servers as hosted tools using the Databricks Supervisor API for server-side agent management.

Do I need the Databricks SDK and openai-python to execute server-side tool calls?▼

You need the Databricks SDK and openai-python dependencies to enable API interactions and execute server-side agent loops and tool calls.

How do I select different AI models at runtime for dynamic tool usage in Databricks?▼

Runtime model selection for dynamic tool usage is supported by the Supervisor API, allowing you to choose models dynamically during server-side agent loop execution in Databricks.

What is the best way to manage complex multi-tool workflows in a production environment?▼

Managing complex multi-tool workflows in production is best handled by the Databricks Supervisor API, which efficiently executes server-side agent operations and scalable tool loops.

Why should I move my agent loop from the client to the Databricks Supervisor API?▼

Moving your agent loop to the Databricks Supervisor API simplifies hosted tool integration, enables runtime model selection, and provides scalable server-side execution for production workflows.

Related Skills

Explore More Software Engineering Skills→
ThomazRossitoThomazRossito
31

databricks-config

Manage Databricks workspace profiles and authentication via the Python SDK.

Community
Intermediate
plurigridplurigrid
60

hoot

Compile Scheme programs into standalone WebAssembly modules with JavaScript interop.

Official
Advanced
itsadijmbtitsadijmbt
1

databricks-agent-bricks

Create and manage Databricks Agent Bricks for knowledge retrieval, SQL exploration, and multi-agent orchestration.

Community
Intermediate
victorlouvictorlou
4

supervisor-api

Delegate agent tool looping to the Databricks Supervisor API via responses.create().

Community
Advanced