postgresql-database-engineering

postgresql-database-engineering is a comprehensive reference skill for professional PostgreSQL work, spanning schema design, query optimization, indexing strategies, performance tuning, partitioning, replication, backup and recovery, high availability, and production database management. It exists to help engineers design, optimize, and maintain high-performance PostgreSQL databases at scale, providing a structured body of knowledge rather than a single narrow tool.

The skill explains PostgreSQL's process-based architecture (postmaster, backend processes, shared memory, background workers, WAL, and the storage layer including TOAST, FSM, and VM) and its foundational concurrency model, MVCC — snapshots, tuple versions, xmin/xmax transaction IDs, visibility rules, and why regular VACUUM and FREEZE matter for reclaiming space and preventing transaction wraparound. It covers the four transaction isolation levels and, per its description, extends into indexing for complex query patterns, partitioning large tables (100M+ rows), streaming and logical replication, connection pooling with pgBouncer or PgPool, monitoring, and scalability patterns. Its 'When to Use' list enumerates concrete scenarios from tuning slow queries to planning migrations and setting up high availability.

It targets backend engineers, database administrators, and platform teams running PostgreSQL in production who need authoritative guidance on performance, reliability, and scaling. It is a knowledge and best-practices reference with no destructive automation, remote code execution, or credential exfiltration, making it a benign engineering resource.