log-correction

Track data analysis errors by recording mistakes and their corresponding fixes in a structured YAML storage system. This builds a shared knowledge base that logs SQL before and after comparisons, categorizes error types, and prevents analysts from repeating the same query mistakes in future projects.

The best way to categorize metric and schema errors is by classifying them into distinct types like SQL, metric, schema, or logic. This systematic categorization identifies common failure patterns and builds a centralized correction log for future reference.

Record the severity of a data validation error by assigning critical, high, medium, or low severity classifications. This severity tracking prioritizes fixes for high-impact mistakes and structures the error correction log to highlight the most disruptive data analysis issues.

Yes, you can log manual corrections when automatic feedback capture is insufficient. The system supports manual correction entry to track systemic gaps in data validation processes, ensuring that analyst mistakes and their corresponding fixes are accurately recorded.

No, the log-correction skill requires no external dependencies or components. It operates independently using structured YAML storage for correction logs and index tracking to manage error categorization and SQL comparison data.