What problem does it solve? Manually creating and configuring GCP data pipeline resources across dev, staging, and prod environments is error-prone and inconsistent. This Skill automates declarative resource provisioning through a single deployment.yaml file, keeping environment-specific configurations synchronized and reproducible. ## Core Features & Use Cases - Declarative Resource Provisioning: Define BigQuery datasets and tables, Dataform repositories and workflow configs, Dataproc clusters, and BigQuery Data Transfer Service configs in one deployment.yaml file. - Environment Management: Map dev, staging, and prod environments to their own projects, regions, variables, and secrets with automatic placeholder resolution. - Secure Secret Handling: Reference existing Secret Manager secrets via variable substitution without ever reading or exposing secret values. - Use Case: A data engineer needs to stand up a new BigQuery dataset, a DTS transfer config, and a scheduled Dataform workflow for the staging environment. The agent discovers the project and region, generates the deployment.yaml, validates it, and deploys everything with gcloud orchestration-pipelines commands. ## Quick Start Ask the agent to create a deployment.yaml that provisions a BigQuery dataset and a Dataform workflow for the dev environment, then validate and deploy it.
