.agents/skills/col-lineage/SKILL.md
Generate column-level lineage for a specific task
npx skillsauth add starlake-ai/starlake-skills col-lineageInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Generates column-level lineage for a specific transform task, showing how each output column traces back to its source columns through SQL transformations. This provides fine-grained data provenance information.
starlake col-lineage [options]
--task <value>: Task name in the form domain.task (required)--output <value>: Output JSON file path (default: console output)--accessToken <value>: Access token for authentication (e.g. GCP)--reportFormat <value>: Report output format: console, json, or htmlAnalyzes the SQL query for a task and traces each output column back to its source columns. For example, given:
-- metadata/transform/kpi/revenue_summary.sql
SELECT
o.order_id,
o.timestamp AS order_date,
SUM(ol.quantity * ol.sale_price) AS total_revenue
FROM
starbake.orders o
JOIN starbake.order_lines ol ON o.order_id = ol.order_id
GROUP BY
o.order_id, o.timestamp
The column lineage output would show:
order_id ← starbake.orders.order_idorder_date ← starbake.orders.timestamptotal_revenue ← starbake.order_lines.quantity, starbake.order_lines.sale_pricestarlake col-lineage --task kpi.revenue_summary
starlake col-lineage --task kpi.order_summary --output col_lineage.json
starlake col-lineage --task kpi.revenue_summary --accessToken $GCP_TOKEN
development
Design SQL transformations for data pipelines with quality checks and dependency management. Use when the user says "design transforms" or "create SQL transformations".
devops
Plan and track sprint progress for data pipeline implementation. Use when the user says "sprint planning" or "plan data sprint".
testing
Analyze data sources in depth: schema, quality, volume, and extraction strategy. Use when the user says "analyze data source" or "profile this data source".
data-ai
Design Starlake-compatible table schemas with types, constraints, privacy, and expectations. Use when the user says "design schema" or "create table definition".