What levels of normalization does this skill support?

The skill supports normalization from 1st Normal Form (1NF) through Boyce-Codd Normal Form (BCNF), allowing users to target specific levels based on their database requirements.

Can I use this for recurring data imports?

Absolutely. Once a normalization structure is defined, the skill generates a transform_config.json that can be used to instantly process future CSV exports with the same schema.

Does it automatically identify primary keys?

Yes, the analysis script detects candidate keys and suggests primary key selections, which the user can then confirm or modify in the configuration file.

Does this skill provide visual documentation?

Yes, it generates erd.md files containing Mermaid syntax, which allows you to visualize the resulting entity-relationship diagram in compatible markdown viewers like VS Code or GitHub.

CSV Database Normalizer

Name: CSV Database Normalizer
Author: elertan

byelertan

データベース管理

Transforms flat CSV files into normalized relational database structures by detecting functional dependencies and generating SQL schemas.

概要

The csv-normalizer skill empowers Claude to convert denormalized data exports into efficient, relational database schemas. By analyzing data structures and identifying functional dependencies, it automates the decomposition of single CSV files into multiple tables adhering to 1NF, 2NF, 3NF, or BCNF standards. This tool is essential for developers migrating spreadsheet data to SQL databases, as it generates not only the normalized data files but also the corresponding SQL DDL statements and Mermaid entity-relationship diagrams for immediate implementation.

主な機能

0 GitHub stars
Multi-stage decomposition into 1NF, 2NF, 3NF, and BCNF structures
Mermaid ERD generation for visual database schema documentation
Automated functional dependency detection with confidence scoring
SQL DDL generation including primary and foreign key constraints
Reusable transformation configs for recurring data processing

ユースケース

Converting legacy flat-file CSV exports into structured relational database schemas
Designing database architectures directly from existing spreadsheet data
Reducing data redundancy and improving integrity in large datasets

概要

主な機能

0 GitHub stars
Multi-stage decomposition into 1NF, 2NF, 3NF, and BCNF structures
Mermaid ERD generation for visual database schema documentation
Automated functional dependency detection with confidence scoring
SQL DDL generation including primary and foreign key constraints
Reusable transformation configs for recurring data processing

ユースケース

Converting legacy flat-file CSV exports into structured relational database schemas
Designing database architectures directly from existing spreadsheet data
Reducing data redundancy and improving integrity in large datasets