Unity Catalog
Open-source universal catalog for data and AI, supporting multi-format and multi-engine governance.
MCPCLI
About Unity Catalog
Open-source universal catalog for data and AI, supporting multi-format and multi-engine governance. Explore how Unity Catalog integrates with the agentic data stack ecosystem and supports autonomous data operations.
Key Features
- Universal open-source catalog for data and AI governance (LF AI & Data)
- Multimodal asset management: tables, volumes, functions, and ML/AI models
- Multi-format support: Delta Lake, Apache Iceberg (UniForm), Parquet, CSV, JSON
- Interoperability with Iceberg REST Catalog and Hive Metastore interface
- Broad engine support: Spark, Trino, DuckDB, Daft, PuppyGraph, StarRocks
- Built-in user management with access control and token-based authentication
- Unified three-level namespace (catalog.schema.object) for all asset types
- Open API spec with Apache 2.0 license for vendor-neutral extensibility
Agent Integration
MCP Server
databrickslabs/mcpExternal Links
Unity Catalog AI Integrations
Official docs for LangChain, Anthropic, OpenAI, LlamaIndex, CrewAI agent integrations
Databricks Labs MCP Server
Official Databricks MCP server exposing UC Functions, Vector Search, and Genie spaces
CLI Documentation
CLI reference for managing catalogs, schemas, tables, volumes, and functions
Databricks REST API - Catalogs
REST API reference for Unity Catalog workspace operations
Unity Catalog GitHub
Main OSS repo — open multi-modal catalog for Data and AI