Skip to main content

Operations

8 skills for managing, monitoring, and maintaining your Starlake deployment.

Skills​

validate​

Validate project configuration. Checks your entire project for YAML schema compliance, broken references, and configuration errors.

You: /validate Check my project for errors and suggest fixes

Validates:

  • YAML schema compliance against starlake.json
  • Connection references exist
  • Domain and table configurations are complete
  • Type definitions are valid
  • Environment variables resolve correctly

metrics​

Metrics collection and management. Configure and query data pipeline metrics including row counts, processing times, and data quality scores.

You: /metrics Set up metrics collection for all load operations

freshness​

Data freshness monitoring. Track how current your data is and alert when tables go stale.

You: /freshness Configure freshness checks for the analytics domain with a 4-hour SLA

gizmosql​

GizmoSQL process management. Manage DuckLake SQL endpoints for serving data via SQL interfaces.

You: /gizmosql Set up a GizmoSQL endpoint for the analytics database

Capabilities:

  • Start/stop SQL endpoints
  • Configure connection pooling
  • Manage DuckDB-based SQL serving
  • Monitor active queries

migrate​

Schema migration. Manage database schema changes and migrations across environments.

You: /migrate Apply schema changes to update the customers table in production

console​

Console access and management. Access the Starlake interactive console for ad-hoc operations.

You: /console How do I access the Starlake console for my BigQuery project?

serve​

Data serving configuration. Set up APIs and endpoints for serving data from your pipeline.

You: /serve Configure a REST endpoint for the product catalog

settings​

Application settings management. Configure global application settings including logging, parallelism, and engine options.

You: /settings Configure parallel loading with 4 concurrent jobs