Skip to content

FDL

DuckLake catalog management CLI.

FDL manages the lifecycle of DuckLake catalogs — initialization, building, and distribution. With a Git-like command interface, you can consistently manage your data catalogs from development to deployment.

Features

  • Initialize and manage DuckLake catalogs (fdl init)
  • Push and pull to S3 or local storage via Named Remotes
  • 3-layer configuration management (project → workspace → user)
  • Automatic metadata generation from dbt artifacts
  • Environment variable injection for pipeline execution (fdl run)
  • HTTP server with CORS and Range request support (fdl serve)

Installation

uv add frozen-ducklake

pip install frozen-ducklake

Requires Python 3.13 or later.

Quick Start

# Initialize a project
fdl init my-dataset

# Configure a remote
fdl config remotes.origin s3://my-bucket

# Run the pipeline
fdl run -- dbt run

# Generate metadata
fdl metadata

# Upload to remote
fdl push origin

See Quick Start for a complete walkthrough.

Learn More