database-query

Database Query (Natural Language)

⚡ UNIQUE FEATURE: Query any database using natural language - automatically generates optimized SQL/NoSQL queries, explains query plans, suggests indexes, and visualizes results. Supports PostgreSQL, MySQL, MongoDB, SQLite, and more.

What This Skill Does

Transform natural language into optimized database queries:

• Natural language to SQL: "Show me users who signed up last month" → SELECT * FROM users WHERE created_at >= NOW() - INTERVAL '1 month'
• Multi-database support: PostgreSQL, MySQL, MongoDB, SQLite, Redis
• Query optimization: Analyzes queries and suggests improvements
• Index suggestions: Recommends indexes for slow queries
• Visual results: Formats query results as tables, charts, JSON
• Query explanation: EXPLAIN ANALYZE with human-readable insights
• Safe mode: Read-only by default with confirmation for writes
• Schema discovery: Auto-learns database structure

Why This Is Unique

First Claude Code skill that:

• Understands intent: Translates vague requests to precise queries
• Cross-database compatible: Same natural language works across SQL/NoSQL
• Performance-aware: Automatically optimizes and suggests indexes
• Safety-first: Prevents destructive operations without confirmation
• Learning mode: Improves by understanding your schema

Instructions

Phase 1: Database Connection & Discovery

1. Identify Database:

   Ask user:
   - Database type (PostgreSQL, MySQL, MongoDB, SQLite, etc.)
   - Connection method (local, remote, Docker, MCP server)
   - Connection string or credentials
   

2. Test Connection:

   # PostgreSQL
   psql -h localhost -U user -d database -c "SELECT version();"
   
   # MySQL
   mysql -h localhost -u user -p database -e "SELECT VERSION();"
   
   # MongoDB
   mongosh "mongodb://localhost:27017/database" --eval "db.version()"
   
   # SQLite
   sqlite3 database.db "SELECT sqlite_version();"
   

3. Discover Schema:

   # PostgreSQL: Get all tables and columns
   psql -d database -c "\dt"
   psql -d database -c "\d+ table_name"
   
   # MySQL: Show database structure
   mysql database -e "SHOW TABLES;"
   mysql database -e "DESCRIBE table_name;"
   
   # MongoDB: List collections and sample documents
   mongosh database --eval "db.getCollectionNames()"
   mongosh database --eval "db.collection.findOne()"