Extracts tribal knowledge and domain expertise to generate structured, reusable data documentation for accurate AI-driven analysis.
The Data Context Extractor is a specialized Claude Code skill designed to bridge the gap between raw database schemas and actual business logic. By enforcing a strict interview-first protocol, it prevents Claude from making incorrect assumptions about your data. It guides users through documenting data sources, entity relationships, metric definitions, and common 'gotchas' or data quality issues. The result is a persistent, project-specific data context that allows Claude to perform accurate analysis, write correct SQL, and understand the nuances of your specific dataset without repeated explanations.
Key Features
015 GitHub stars
02Strict verification gates to ensure generated documentation matches user expertise
03Native integration with financial data skills like WRDS and LSEG
04Guided 5-round interview process to capture undocumented domain knowledge
05Structured generation of entity maps, metric definitions, and data hygiene logs
06Bootstrap and Iteration modes for creating or updating project contexts
Use Cases
01Documenting project-specific data filters and 'gotchas' for collaborative research
02Onboarding Claude to a complex legacy database with ambiguous column names
03Standardizing business metric definitions and SQL formulas across a data team