Update to SQL Copilot

{
  "description": "Interpret user-provided SQL execution plans and warehouse query profiles.",
  "included_files": [],
  "name": "explain-query-plans",
  "skill_md_contents": "---\nname: explain-query-plans\ndescription: Interpret user-provided SQL execution plans and warehouse query profiles.\n---\n\n# Explain Query Plans\n\n**Trigger:** Interpret user-provided SQL execution plans and warehouse query profiles.\n\n## Workflow\n1. Identify engine, plan format, version and whether it is estimated or actual; do not treat estimated rows as observed.\n2. Locate high-cost scans, expensive joins, sorting, spilling, shuffle/data movement, skew, cardinality-estimation gaps and queueing.\n3. Correlate plan nodes to SQL operators and filter/projection semantics; distinguish root bottlenecks from downstream symptoms.\n4. Suggest a targeted experiment with expected changes in profile metrics and query output.\n5. Avoid claiming the plan proves a specific index or distribution strategy without workload evidence.\n\n## Critical standard\nPostgreSQL EXPLAIN ANALYZE executes the statement: do not casually recommend it for mutating SQL against production.\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/query-plans.md` when relevant.\n- Read `../../references/performance-diagnosis.md` when relevant.\n- Read `../../references/official-docs.md` when relevant.\n\n## Return\nPlan interpretation, bottleneck hypotheses, safe experiments and validation.\n"
}