KiloClawPowered by OpenClaw
Back to Cookbook
OpenClaw recipe

dbt Copilot

Generate models, tests, and YAML configs from descriptions

Describe a dbt model in plain English and get the SQL, schema YAML, and tests auto-generated. Also debugs compilation errors, suggests materializations, and analyzes your DAG for orphaned models and circular dependencies.

CommunitySubmitted by CommunityWork5 min setup
Try in KiloClawFree 7-day trial

INGREDIENTS

🐙GitHub

PROMPT

Create a skill called "dbt Copilot". When I describe a data model in plain English (e.g., "monthly revenue by customer segment, only completed orders, excluding refunds"), generate: (1) The dbt SQL model with appropriate CTEs, source references, and ref() calls. (2) A schema.yml entry with column descriptions and appropriate tests (unique, not_null, accepted_values, relationships). (3) A recommended materialization strategy with reasoning. When I paste a dbt compilation or runtime error, diagnose the issue and suggest a fix. When I ask for a project health check, analyze the DAG for: orphaned models (no downstream consumers), unused sources, circular dependencies, models that should be ephemeral, and missing tests. Support dbt Core and dbt Cloud syntax.

How It Works

dbt is powerful but verbose. Every new model needs SQL, a schema.yml entry,

test definitions, and documentation. As projects grow to 500+ models, the

YAML alone becomes a maintenance burden. This skill generates the boilerplate

and keeps the project healthy.

What You Get

  • SQL model generation from plain-English descriptions
  • Auto-generated schema.yml with column descriptions and tests
  • Test suggestions based on data patterns (uniqueness, not_null, accepted_values, relationships)
  • Compilation error debugging with plain-English explanations
  • DAG analysis (orphaned models, unused sources, circular dependencies)
  • Materialization recommendations (table vs. view vs. incremental vs. ephemeral)

Setup Steps

  1. Ask your Claw to create a "dbt Copilot" skill with the prompt below
  2. Point it at your dbt project directory
  3. Ask for new models, debug errors, or run a project health check

Tips

  • The DAG analysis is worth running monthly — projects accumulate dead models fast
  • Test suggestions are based on actual data, not just column names
  • Use it to generate the YAML config that nobody wants to write by hand
  • The materialization recommendations can significantly improve run times
Tags:#dbt#data-engineering#sql#automation

Related Recipes

DAG Builder

Describe a pipeline in English, get a working Airflow DAG

Skip the boilerplate. Describe your data pipeline in plain English and get a complete Airflow DAG with proper dependencies, error handling, retries, and alerting. Also debugs failing DAGs by analyzing task logs.

Work5 min setup

Schema Sentinel

Get alerted when upstream schemas change before your pipelines break

Monitors your source databases, APIs, and third-party integrations for schema changes — new columns, removed columns, type changes, enum value additions — and alerts you before the downstream breakage happens.

Work5 min setup

CLAWBITE AI

Local-first AI assistant that automates small daily tasks safely on your device

A personal, local-first AI assistant that automates small daily tasks—organizing files, setting reminders, and monitoring system events—without touching sensitive data or taking risky actions without your approval.

Personal5 min

Bulk Change Airbnb Rates

Update pricing across all your Airbnb listings without clicking through each one

Automate bulk rate changes across multiple Airbnb listings using your Claw. Useful for seasonal pricing updates, last-minute discounts, or syncing rates after a change in your hosting strategy.

Home5 min