Azure SQL Database Best Practices Skill

A comprehensive skill for AI coding agents that provides best practices for Azure SQL Database development. Based on Microsoft's SQL Assessment API, Azure SQL Database performance guidance, and official documentation.

🚀 Quick Install

npx skills add <your-github-username>/azure-sql-best-practices

📋 What This Skill Does

This skill helps AI agents:

• Analyze T-SQL code for performance, security, and maintainability issues
• Fix SQL injection vulnerabilities in application code (Python, Node.js, C#)
• Optimize queries by identifying non-SARGable predicates, missing indexes, and inefficient patterns
• Improve connection management with pooling and retry logic
• Configure databases with Query Store, automatic tuning, and proper settings
• Review security including permissions, encryption, and row-level security

📁 Repository Structure

azure-sql-best-practices/
├── skills/
│   └── azure-sql-best-practices/
│       ├── SKILL.md           # Main skill definition
│       ├── AGENTS.md          # Compiled guide for agents
│       ├── rules/             # Individual rule files
│       │   ├── query-avoid-select-star.md
│       │   ├── query-parameterize.md
│       │   ├── index-cover-queries.md
│       │   ├── connection-pooling.md
│       │   ├── security-least-privilege.md
│       │   ├── tsql-set-nocount.md
│       │   └── config-query-store.md
│       ├── scripts/           # Helper scripts
│       │   ├── analyze-tsql.py
│       │   ├── check-indexes.sql
│       │   └── security-audit.sql
│       └── references/        # Reference documentation
├── test-app/                  # Test application with violations
│   ├── sql/                   # T-SQL with issues
│   ├── python/                # Python app with issues
│   ├── node/                  # Node.js app with issues
│   ├── csharp/                # C# app with issues
│   └── tests/                 # Test runner
└── README.md

🎯 Rule Categories

Priority	Category	Impact	Example Rules
1	Query Performance	CRITICAL	query-avoid-select-star, query-parameterize, query-sargable
2	Indexing Strategy	CRITICAL	index-cover-queries, index-missing-index-dmv
3	Security	HIGH	security-least-privilege, security-encrypt-connections
4	Connection Management	HIGH	connection-pooling, connection-retry-logic
5	T-SQL Patterns	MEDIUM-HIGH	tsql-set-nocount, tsql-error-handling
6	Database Configuration	MEDIUM	config-query-store, config-auto-tuning
7	Data Modeling	MEDIUM	model-normalization, model-appropriate-types
8	Monitoring	LOW-MEDIUM	monitor-dmvs, monitor-query-performance-insight

💡 Example Usage

Once installed, your AI agent will automatically apply these best practices when working with Azure SQL Database code.

Ask Your Agent:

"Review this stored procedure for best practices"
"Fix SQL injection vulnerabilities in my Python database code"
"Optimize this query that's running slowly"
"Check my database for security issues"
"Add retry logic and connection pooling to this application"

Example: Before & After

Before (Vulnerable):

def get_user(user_id):
    query = f"SELECT * FROM Users WHERE UserID = {user_id}"
    cursor.execute(query)

After (Fixed):

def get_user(user_id: int) -> Optional[User]:
    query = "SELECT UserID, Name, Email FROM Users WHERE UserID = ?"
    cursor.execute(query, (user_id,))

🧪 Testing the Skill

The test-app/ directory contains applications with intentional violations for testing:

# Run the T-SQL analyzer
cd skills/azure-sql-best-practices/scripts
python analyze-tsql.py ../../test-app/sql/

# Run all tests
cd test-app/tests
python test_runner.py

📚 Sources

This skill is based on:

• SQL Assessment API
• Azure SQL Database Performance Guidance
• Azure SQL Database Security Best Practices
• Query Store Best Practices

🤝 Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Add new rules following the existing format
4. Include tests for new rules
5. Submit a pull request

📄 License

MIT License - see LICENSE for details.

🔗 Related Skills

• vercel-labs/agent-skills - React/Next.js best practices