data-context-extractor

You can document institutional knowledge for a data warehouse by running interactive schema discovery and targeted analyst interviews to capture metric formulas, entity disambiguation, and common gotchas into structured reference files.

The best way to bootstrap company-specific data analysis skills is to extract existing analyst knowledge through guided discovery questions and package it into a structured skill directory with SKILL.md and SQL examples.

Yes, this data context extraction approach works with Snowflake, BigQuery, PostgreSQL, Redshift, and Databricks. It guides connection to these warehouses to list datasets, surface key tables, and inspect schema details for documentation.

You capture common SQL query mistakes and standard filters by running targeted interview questions focused on metric definitions, primary identifiers, and common gotchas, then converting the answers into reusable domain reference files.

Yes, you can generate a packaged distributable skill. The extraction process produces a structured skill directory containing SKILL.md, domain reference files, SQL dialect notes, and example queries ready for delivery or iteration.

You need to create a company-specific data context skill during onboarding, documentation updates, or analytics troubleshooting to reduce onboarding time, prevent inconsistent metrics, and avoid common query mistakes across the team.