This repository provides a collection of solutions for the Atlan Metadata Lakehouse (MDLH), enabling Atlan customers to deploy production-ready tools on their own compute environments. Solutions include analytical models (Gold Layer), maintenance utilities, and more.

• Snowflake ❄️
• Databricks 🔷
• BigQuery ☁️
• DuckDB 🦆 (coming soon)
• Trino 🚀 (coming soon)

The Gold Layer provides curated, analytics-ready metadata views that serve as the single entry point for both human and AI consumption of lakehouse metadata.

What it delivers:

• Unified Asset Registry: Centralized view of all assets across SQL, BI, pipelines, data quality, and object domains
• Relational Asset Details: Consolidated metadata for databases, schemas, tables, views, columns, queries, and procedures
• Governance & Classification: Tag and custom metadata views for data governance
• Lineage: Complete multi-hop upstream and downstream lineage relationships
• Data Quality: Views for Atlan-native and third-party data quality rules
• Pipeline Details: Orchestration and pipeline asset metadata
• Glossary: Business glossary terms, categories, and hierarchies

Snowflake supports native Iceberg REST Catalog federation, allowing you to query the Atlan Lakehouse directly without any external scripts or scheduled refresh jobs. The catalog integration guide walks through creating an External Volume, Catalog Integration, and Linked Database.

Databricks and BigQuery do not currently support querying federated Iceberg REST catalogs natively. These scripts provide a workaround by creating local table references (foreign Iceberg tables in Databricks Unity Catalog, external Iceberg tables in BigQuery) that point directly to the Atlan Lakehouse metadata files. The scripts handle both initial table creation and ongoing metadata refresh to keep tables in sync. Databricks supports both AWS S3 and Azure ADLS storage.

An Agent Skills skill that teaches AI coding agents (Claude Code, Cortex Code, Genie Code) how to connect to and query the lakehouse. Includes ~45 SQL templates covering metadata completeness, lineage analysis, glossary export, and usage analytics (active users, feature adoption, engagement, retention, health scoring). See skills/atlan-lakehouse/ for installation instructions.

A native Snowflake Streamlit app that identifies stale Iceberg tables and provides an option to repair them by refreshing metadata and enabling auto-refresh.

• Access to one of the supported compute environments
• Appropriate permissions to create databases, schemas, views, and tables
• Connection to your Atlan metadata catalog

Navigate to the platform-specific folder for available solutions and setup instructions:

• Snowflake
• Databricks
• BigQuery

• Analytics & Reporting: Query-ready metadata for business intelligence and analytics
• AI/ML Consumption: Structured metadata for AI agents and machine learning models
• Data Governance: Standardized views for compliance and governance reporting
• Lineage Analysis: Complete lineage visualization and impact analysis
• Asset Discovery: Unified search and discovery across all metadata types

lakehouse-solutions/
├── README.md                              # This file
├── snowflake/
│   ├── README.md                          # Snowflake solutions overview
│   ├── catalog-integration/
│   │   └── README.md                      # Catalog integration setup guide
│   ├── gold-layer/
│   │   ├── README.md                      # Gold Layer setup guide
│   │   └── MDLH_Gold_layer.sql            # Gold Layer deployment script
│   └── mdlh-table-maintenance/
│       ├── README.md                      # Table maintenance setup guide
│       └── MDLH_table_refresh_repair.py   # Streamlit app
├── databricks/
│   ├── README.md                          # Databricks solutions overview
│   ├── gold-layer/
│   │   ├── README.md                      # Gold Layer setup guide
│   │   ├── MDLH_Gold_layer.sql            # Gold Layer deployment script
│   │   └── refresh_materialized_views.sql # Scheduled refresh script
│   └── foreign-iceberg-tables/
│       ├── README.md                      # Foreign Iceberg Tables setup guide
│       ├── dbx_foreign_iceberg_tables_create.py   # Table creation script
│       └── dbx_foreign_iceberg_tables_refresh.py  # Table refresh script
├── bigquery/
│   ├── README.md                          # BigQuery solutions overview
│   ├── gold-layer/
│   │   ├── README.md                      # Gold Layer setup guide
│   │   └── MDLH_Gold_layer.sql            # Gold Layer deployment script
│   └── external-iceberg-tables/
│       ├── README.md                      # External Iceberg Tables setup guide
│       └── bq_external_iceberg_tables_create_refresh.py  # Create/refresh script
├── skills/
│   └── atlan-lakehouse/
│       ├── README.md                      # Installation & usage guide
│       └── SKILL.md                       # Agent skill definition
├── duckdb/                                # Coming soon
└── trino/                                 # Coming soon

This repository is maintained by the Atlan team. For issues, questions, or contributions, please contact the Atlan engineering team.

[Specify license here]

For support and questions:

• Documentation: See platform-specific README files
• Atlan Support: [Contact information]

Note: This repository contains deployment scripts for customer-managed infrastructure. All scripts are designed to be idempotent and production-ready.