What problem does it solve?
This Skill simplifies the creation and management of pre-built AI components, known as Agent Bricks, enabling the development of sophisticated conversational AI applications on the Databricks platform without deep AI expertise.
Core Features & Use Cases
- Knowledge Assistants (KA): Build document-based Q&A systems using RAG on files stored in Unity Catalog Volumes. Ideal for creating chatbots that answer questions from company policies, manuals, or reports.
- Genie Spaces: Enable natural language to SQL queries for data exploration, allowing users to interact with Unity Catalog tables conversationally. Perfect for business analysts and data scientists.
- Supervisor Agents (MAS): Orchestrate multiple specialized agents (KAs, Genie Spaces, model serving endpoints, UC functions, external MCP servers) into a single, unified interface. Use this to create complex AI assistants that can handle diverse user requests.
- Use Case: A company wants to provide a single point of contact for employees to ask HR policy questions (using a KA), get sales performance reports (using a Genie Space), and submit IT support tickets (using an external system via MAS). This Skill allows them to build and manage all these components.
Quick Start
Use the databricks-agent-bricks skill to create a Knowledge Assistant named 'HR Policy Assistant' that answers questions from documents in '/Volumes/my_catalog/my_schema/raw_data/hr_docs'.