database-sql-expert

name	database-sql-expert
description	Provides expert guidance on writing clean, performant, and maintainable SQL queries. Covers performance optimization, readability improvements, and database design best practices. Use when you need to optimize SQL queries, improve query readability, or follow database design principles.
license	MIT
compatibility	Requires access to a SQL database (PostgreSQL, MySQL, SQL Server, etc.)

SQL Expert Skill

This skill provides expert guidance on writing clean, performant, and maintainable SQL queries. It covers the most important best‑practice recommendations for performance, readability, and database design.

When invoked, the skill will:

• Analyze your SQL query and provide specific improvement recommendations
• Explain performance optimization techniques with examples
• Suggest readability improvements for complex queries
• Recommend database design best practices when applicable

Key Best Practices

Performance Optimization

See references/sql-performance.md for detailed guidance on:

• Selecting only necessary columns instead of using SELECT *
• Filtering data early with WHERE clauses
• Effective index usage and verification
• Avoiding functions on indexed columns
• Using EXISTS for existence checks
• Optimizing JOIN operations

Readability and Maintainability

See references/sql-readability.md for guidance on:

• Using Common Table Expressions (CTEs) to break down complex queries
• Consistent formatting (uppercase keywords, lowercase snake_case identifiers)
• Wise commenting practices
• Using parentheses for clarity in WHERE clauses

Database Design

See references/sql-design.md for recommendations on:

• Appropriate data normalization
• Using stored procedures for frequently used logic
• Performance monitoring techniques
• Preventing SQL injection with parameterized queries

Gotchas

• Implicit conversions: Comparing different data types can cause performance issues as indexes may not be used
• NULL handling: Remember that NULL = NULL is unknown, not true - use IS NULL instead
• Query plan caching: Complex queries with many parameters may not benefit from plan caching
• Locking behavior: Different isolation levels can cause unexpected locking and blocking
• Aggregate functions: COUNT(*) counts rows, COUNT(column) counts non-null values

How to Use

When you have a SQL query that needs improvement, share it with specific context about:

• What database system you're using (PostgreSQL, MySQL, etc.)
• What performance issues you're observing (slow execution, high resource usage)
• What specific aspects you want improved (performance, readability, safety)

Example invocation:

How can I optimize this query for better performance?
SELECT * FROM orders WHERE YEAR(order_date) = 2023

The skill will respond with targeted recommendations including:

• Specific indexes to add
• Query rewrites for better performance
• Readability improvements
• Safety considerations

Evaluation Workflow

This skill includes test prompts to verify its effectiveness:

# Run evaluation (if scripts are implemented)
python scripts/run_evaluation.py

Test prompts cover:

• Performance optimization scenarios
• Readability improvements
• Database design best practices

See evals/evals.json for test cases.

Output Format

When providing SQL recommendations, the skill will typically:

1. Identify issues in the current query
2. Provide an improved version
3. Explain why the changes improve performance, readability, or safety
4. Offer additional tips for similar situations

---

Following Agent Skills specification for proper progressive disclosure and best practices