Does it help with data cleaning?

Absolutely. It identifies 'orphaned records' (child records pointing to non-existent parents), which is a critical first step in database cleaning and referential integrity repair.

Can it identify many-to-many relationships?

Yes, the skill specifically looks for junction table characteristics, such as tables containing multiple foreign keys with few other attributes, to map many-to-many connections.

What databases are supported?

The skill is designed for any relational database that supports standard SQL queries, including SQLite, PostgreSQL, MySQL, and SQL Server.

How does this skill detect foreign keys that aren't defined in the schema?

It uses a combination of naming heuristics (searching for columns like 'user_id' or 'customerID') and statistical analysis to verify if the data values in a column actually match the primary keys of another table.

Database Relationship Detector

Name: Database Relationship Detector
Author: tilmon-engineering

bytilmon-engineering

•

Database Management

Identifies and validates foreign key relationships in relational databases using naming heuristics and deep value overlap analysis.

About

The Database Relationship Detector skill provides a systematic framework for reverse-engineering undocumented database schemas. It guides Claude through a 5-phase process to discover foreign key candidates based on naming conventions, verify those relationships through statistical value overlap checks, and determine relationship cardinality. By identifying orphaned records and validating referential integrity, this skill is indispensable for developers auditing legacy data, preparing migrations, or performing complex data analysis on unfamiliar schemas.

Key Features

Statistical value overlap analysis to calculate the confidence of potential relationships
Referential integrity auditing to pinpoint orphaned records and unused primary keys
2 GitHub stars
Automated cardinality assessment to distinguish between 1:1, 1:many, and many:many relationships
Heuristic-based candidate discovery using naming patterns like _id, fk_, and table name matching
Detection of complex structures including composite keys, junction tables, and self-referencing hierarchies

Use Cases

Validating data quality and referential integrity before critical data migrations
Reverse-engineering undocumented legacy databases to create accurate ER diagrams
Optimizing join performance by understanding data distribution and relationship cardinality

About

Key Features

Statistical value overlap analysis to calculate the confidence of potential relationships
Referential integrity auditing to pinpoint orphaned records and unused primary keys
2 GitHub stars
Automated cardinality assessment to distinguish between 1:1, 1:many, and many:many relationships
Heuristic-based candidate discovery using naming patterns like _id, fk_, and table name matching
Detection of complex structures including composite keys, junction tables, and self-referencing hierarchies

Use Cases

Validating data quality and referential integrity before critical data migrations
Reverse-engineering undocumented legacy databases to create accurate ER diagrams
Optimizing join performance by understanding data distribution and relationship cardinality