Document existing data assets as MD-DDL baseline files.
npx skills add https://github.com/Semprini/md-ddl --skill baseline-capture
Please help me install this Agent Skill. Skill: baseline-capture Source: https://github.com/Semprini/md-ddl/tree/main/agents/agent-ontology/skills/baseline-capture Command: npx skills add https://github.com/Semprini/md-ddl --skill baseline-capture
This skill addresses the challenge of documenting existing data schemas, models, ETL pipelines, and catalog metadata for audit, historical record-keeping, or as a starting point for canonical modeling.
Use the baseline-capture skill to document the provided CREATE TABLE statements as a baseline.
High-intent search queries and answers about installing and using this skill.
To document existing data assets, you can feed raw DDL scripts, dbt models, or catalog exports into a baseline generator to automatically create structured MD-DDL baseline files with metadata headers for audit and historical record-keeping.
Yes, you can use natural language prose descriptions to generate schema documentation. The baseline capture process accepts unformatted text descriptions and converts them into structured baseline files without requiring strict input formatting.
The best way to create a baseline for a migration audit is to process your ETL pipeline definitions and catalog exports through an automated baseline generator, which outputs structured MD-DDL files within a dedicated baselines folder.
Yes, baseline documentation works directly with raw catalog exports and ERDs. The system accepts these various input formats natively, allowing you to generate comprehensive structured baseline files without needing to manually reformat the source materials.
You can identify candidate entities for canonical modeling by processing existing data schemas through a baseline capture tool, which analyzes the structured baseline files generated from your inputs and automatically flags suitable entities for canonical modeling.
You should not use automated baseline files when your goal is to create canonical entities directly, as this documentation process is strictly for audit, historical record-keeping, and generating raw inputs for downstream canonical modeling.
SempriniConvert database schemas and ETL logic into draft MD-DDL domain models.
SempriniGenerate Neo4j Cypher DDL schemas from MD-DDL domain models.
yu-iskwRegenerate TypeScript types from dbt artifact JSON schemas.
scumunnaOptimize TikTok ad creative assets, catalog data, and targeting strategies.
