Dbt Ingest

If you're pushing dbt schema.yml files into a semantic layer, this is the bridge. It reads dbt model and source definitions, maps column descriptions and data tests (not_null, unique, accepted_values, relationships) into ktx overlays, and grounds everything against the actual warehouse schema before writing. The skill won't invent columns or metrics from thin air. It verifies identifiers with entity_details and sql_execution probes, which matters because dbt YAML is documentation, not a schema of record. Useful when you want dbt's human context (descriptions, test intent) feeding your semantic layer without waiting for a full manifest sync, but only if the tables already exist in the warehouse.

GitHub stars: 1.1K
last updated: Jun 11, 2026

View on GitHub

Install

$ npx -y skills add kaelio/ktx --skill dbt_ingest --agent claude-code

Installs into .claude/skills of the current project.

Files

dbt → ktx (bundle ingest)

Use this skill for uploaded dbt projects (dbt_project.yml at stage root, models/**, sources/**, schema.yml). There is no fetch() in v1 - scheduled dbt parse / manifest.json pulls are out of scope; host-provided dbt sync may still backfill structured test metadata into _schema on the next sync.