What problem does it solve?
When writing SQL against a dbt-powered data warehouse, it is hard to know which BigQuery tables exist, what each one contains, and which model fits a vague data question. This Skill gives the agent a lookup index of dbt models so queries target the right tables with the right grain, filters, and join keys.
Core Features & Use Cases
- Model Lookup Index: Organizes dbt models into domain sections (Users, Activity, Revenue, Events) with descriptions of each table's grain and content.
- Query Guidance: Each entry lists "Useful for" bullets covering common query patterns, join keys, standard filters, and partition fields.
- Cost and Correctness Guardrails: Documents mandatory filters (e.g., excluding internal users), partition pruning rules, and fully-qualified production table paths.
- Use Case: A stakeholder asks "How many weekly active subscribers churned last quarter?" The agent consults the index to pick the activity and subscription models, applies the standard filters, and writes a partition-safe BigQuery query.
Quick Start
Ask the agent to find which dbt models answer your data question and write a BigQuery query using the indexed tables.