What problem does it solve?
Getting SaaS data from tools like HubSpot, Stripe, or Salesforce into a SQL-queryable semantic layer normally requires manual ETL scripting, schema mapping, and model authoring. This Skill automates the entire path from API credentials to a verified Wren project that answers real SQL queries.
Core Features & Use Cases
- Guided dlt Pipeline Setup: Walks through installing dlt, picking a verified SaaS source, configuring credentials, and loading data into a local DuckDB file.
- Automatic Project Generation: The introspect_dlt.py script introspects the DuckDB schema, filters dlt internal tables and columns, detects parent-child relationships, normalizes types via sqlglot, and emits a complete Wren v2 YAML project.
- End-to-End Verification: Builds the MDL and runs real SQL queries against every model before declaring success, catching catalog mismatches and type errors.
- Use Case: A data analyst with a Stripe secret key runs the pipeline, generates a Wren project from stripe_data.duckdb, and immediately queries customers, charges, and subscriptions with SQL.
Quick Start
Ask the assistant to connect your HubSpot account to Wren Engine using your private app token so you can analyze contacts and deals with SQL.