What problem does it solve? Turning a data model design into actual dbt or SQL implementation code is repetitive and error-prone, especially when tracking which tables are done, which are context-only, and how lineage and join keys should map to SQL. This Skill automates that implementation loop one task at a time. ## Core Features & Use Cases - Task-driven code generation: Reads pending tasks from .modscape/changes/<name>/tasks.md and generates dbt/SQL code for staging, core, and mart tables in lineage dependency order. - Spec-aware SQL generation: Builds SELECT expressions, JOIN clauses from relationships, incremental WHERE filters, and SCD Type 2 merge patterns directly from spec-model.yaml. - Governance and safety: Skips Context Only tables, records open questions in questions.md with PII-safe investigation queries, and never edits generated files directly. - Use Case: After designing a new mart layer with @modscape-spec-design, run this Skill repeatedly to implement each table task, with checkboxes updated automatically after every generation. ## Quick Start Ask the AI to run @modscape-spec-implement with your spec name to implement the next pending task from the change's tasks.md file.
