Starlake Skills
Open-source Claude Code plugin providing 48 specialized skills for building, configuring, and operating Starlake data pipelines.
git clone https://github.com/starlake-ai/starlake-skills.git ~/.claude/skills/starlake-skillsWhy Starlake Skills?
Your AI-powered co-pilot for declarative data pipeline development.
Open source & auditable
Every skill, every prompt, every configuration pattern is inspectable and extensible. Apache-2.0 licensed for full transparency.
Cross-platform, not single-vendor
Covers BigQuery, Snowflake, DuckDB, PostgreSQL, Redshift, and Databricks. Write once, deploy anywhere.
AI-native workflow
Purpose-built for Claude Code. Ask questions in natural language and get expert Starlake guidance with ready-to-use configurations.
Complete coverage
48 skills covering every CLI command, configuration pattern, write strategy, data quality expectation, and production best practice.
48 Specialized Skills
Covering the full Starlake lifecycle — from ingestion to orchestration.
Ingestion & Loading
9 skills- autoload
- load
- cnxload
- esload
- kafkaload
- ingest
- preload
- stage
Transformation
2 skills- transform
- job
Extraction
5 skills- extract
- extract-schema
- extract-data
- extract-bq-schema
- extract-script
Schema Management
5 skills- bootstrap
- infer-schema
- xls2yml
- yml2ddl
- yml2xls
Lineage & Dependencies
4 skills- lineage
- col-lineage
- table-dependencies
- acl-dependencies
Operations & Orchestration
10 skills- dag-generate
- dag-deploy
- validate
- metrics
- freshness
- gizmosql
See It in Action
Natural language commands that generate production-ready configurations.
# Set up a new Starlake project with BigQuery
> /bootstrap a new project targeting BigQuery with Airflow orchestration
# Configure data ingestion for CSV files
> /load CSV files from GCS into the customers domain with OVERWRITE strategy
# Generate column-level lineage
> /col-lineage for the revenue_summary transform
# Create Airflow DAGs from your pipeline config
> /dag-generate for all domains using Airflow with daily schedule
# Validate your entire project configuration
> /validate the full project and fix any schema errors
# Extract schemas from an existing Snowflake database
> /extract-schema from Snowflake connection "prod" for the analytics schemaMulti-Platform Support
One plugin, every warehouse and orchestrator.