I need to check for duplicate and near-duplicate rows across the dataset's tables. Let me start by examining the main tables for exact duplicates based on their business keys. ```json { "action": "tool_call", "reasoning": "Check for exact duplicates in beer_ratings using the natural key (user_id, beer_id, created_at). This is the largest table and most likely to have duplicates.", "tool": "query_sql", "args": { "sql": "SELECT user_id, beer_id, created_at, COUNT(*) as duplicate_count FROM raw.beer_ratings GROUP BY user_id, beer_id, created_at HAVING COUNT(*) > 1 ORDER BY duplicate_count DESC LIMIT 50", "purpose": "Find exact duplicate ratings by the same user for the same beer at the same timestamp" } } ```