data-context-extractor

To extract tribal knowledge and bootstrap data context, this skill formalizes company-specific analyst knowledge through schema discovery and structured questions. It generates tailored reference files and frontmatter to establish a foundational data warehouse knowledge base.

The best way to create a new data analysis skill is using Bootstrap Mode, which discovers warehouse schemas, asks key domain questions, and generates initial reference files. This method rapidly builds a tailored skill from scratch without requiring prior metadata.

Yes, you can enhance an existing data skill using Iteration Mode. This mode loads your current skill and appends domain-specific reference files, allowing you to continuously update and tailor the data context with new company knowledge without starting over.

Yes, the data context extractor supports Snowflake warehouse schema discovery. It coordinates schema discovery to automatically map your warehouse structure, using that information to generate accurate metadata and bootstrap tailored data skills for your specific Snowflake environment.

Generated reference files use a structured template format with enforced frontmatter. This ensures consistent metadata organization across your data skills, allowing the extracted domain knowledge to be reliably navigated and utilized by downstream data analysis processes.

Formalizing company-specific data knowledge ensures your data skills understand your unique warehouse schemas and domain context. Generic skills lack this tailored reference data, leading to inaccurate analysis, whereas formalized tribal knowledge enables precise, company-specific data navigation.