How often should I run the /data:init command?

You should run it during initial project setup and whenever your database schema undergoes significant changes to ensure the local metadata remains synchronized.

Does it support large-scale databases?

Absolutely. It includes logic to identify tables with over 100 million rows and flags them so that Claude knows to apply mandatory filters, preventing expensive or slow queries.

How does this skill improve Claude's data analysis capabilities?

It provides Claude with a detailed map of your data warehouse, including table descriptions and column types, which allows the AI to write more accurate SQL queries without guessing schema names.

Where does the skill store the discovered schema information?

It generates a .astro/warehouse.md file in your project root, which is designed to be committed to version control so your entire team can share the context.

Can it read my dbt project documentation?

Yes, it automatically scans for dbt YAML models and SQL files with frontmatter to extract business descriptions and primary key information to enrich the warehouse reference.

Data Warehouse Discovery & Schema Mapper

Name: Data Warehouse Discovery & Schema Mapper
Author: diegosouzapw

bydiegosouzapw

•

데이터 과학 및 ML

Automates data warehouse schema discovery and enriches table metadata with codebase context for accurate AI-driven data analysis.

The Data-AI skill bridges the gap between your data warehouse and Claude by generating a comprehensive, version-controllable schema reference. It intelligently crawls your databases to extract table metadata and row counts while simultaneously scanning your codebase—including dbt models, SQL files, and documentation—to inject business context. By centralizing this information in a structured markdown file and integrating with CLAUDE.md, it transforms Claude's ability to perform concept-to-table lookups, significantly reducing query errors and eliminating the need for repetitive manual schema exploration.

주요 기능

01Automated multi-database schema discovery and metadata extraction

02Intelligent codebase scanning for dbt models and SQL documentation enrichment

03Categorical value family discovery to identify common filter variants

0410 GitHub stars

05Automatic identification of large tables (>100M rows) for optimized querying

06Generation of a shareable, version-controllable .astro/warehouse.md reference

사용 사례

01Synchronizing codebase documentation with live production database structures

02Improving the accuracy of natural language to SQL generation tasks

03Onboarding AI agents to complex data warehouses with undocumented schemas

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add diegosouzapw/awesome-omni-skill data-ai

For use in Claude.ai and ChatGPT

주요 기능

01Automated multi-database schema discovery and metadata extraction

02Intelligent codebase scanning for dbt models and SQL documentation enrichment

03Categorical value family discovery to identify common filter variants

0410 GitHub stars

05Automatic identification of large tables (>100M rows) for optimized querying

06Generation of a shareable, version-controllable .astro/warehouse.md reference

사용 사례

01Synchronizing codebase documentation with live production database structures

02Improving the accuracy of natural language to SQL generation tasks

03Onboarding AI agents to complex data warehouses with undocumented schemas