What problem does it solve? Translating a written requirements spec into a concrete data model is error-prone and hard to iterate on, especially when changes must not touch the main model file. This Skill turns a spec.md into a validated, work-scoped YAML data model that can be refined table by table. ## Core Features & Use Cases - Spec-driven model design: Reads spec.md and business context, then creates or updates tables, lineage, and relationships in changes/<name>/spec-model.yaml via the modscape CLI. - Incremental one-table-per-invocation workflow: Tracks design progress in design.md, generates stub design docs for all affected tables, and details one table per run. - Impact and traceability management: Classifies downstream impact, records open questions and assumptions in questions.md, detects tacit-knowledge risks, and maintains a project glossary. - Use Case: A data engineer runs the skill against a new feature spec to extract relevant tables from the main YAML, design a new fact table with lineage and FK relationships, validate the model, and produce review-ready design documentation. ## Quick Start Ask the AI to run @modscape-spec-design with your spec name to design the next pending table in your data model.
