Update to SQL Copilot

{
  "description": "Review provided SQL for semantic correctness before proposing cosmetic rewrites or optimization.",
  "included_files": [],
  "name": "review-sql-correctness",
  "skill_md_contents": "---\nname: review-sql-correctness\ndescription: Review provided SQL for semantic correctness before proposing cosmetic rewrites or optimization.\n---\n\n# Review Sql Correctness\n\n**Trigger:** Review provided SQL for semantic correctness before proposing cosmetic rewrites or optimization.\n\n## Workflow\n1. Reconstruct the intended result: who or what constitutes one row, and which measures must reconcile?\n2. Walk each clause in logical order: FROM/JOIN, WHERE, GROUP BY, HAVING, windows, SELECT, ORDER BY and LIMIT. Consider platform-specific evaluation details.\n3. Examine join keys, filtering of optional rows, aggregation level, 3-valued NULL logic, date and timezone semantics, tie-breaks and type conversion.\n4. Distinguish confirmed bugs from plausible risks requiring sample data or schema.\n5. Produce the smallest semantic fix; show which rows or values may change and how to prove the fix against data.\n\n## Critical standard\nDo not optimize queries whose semantics have not been validated. Never use DISTINCT to hide unexplained fanout.\n\n## Evidence and execution\nOnly say \"executed\", \"passed\", \"observed\", \"optimized\" or \"deployed\" when a connected tool returned evidence for that claim. An SQL draft, a textual review, or local fixture proof alone is not production validation. Never ask the user to paste passwords or raw confidential records.\n\n## Supporting references\n- Read `../../references/correctness-and-grain.md` when relevant.\n- Read `../../references/joins-and-fanout.md` when relevant.\n- Read `../../references/null-time-types.md` when relevant.\n\n## Return\nFindings ordered by impact, corrected SQL if needed, and reproducible validation probes.\n"
}