From role-database
Cross-engine query optimization guide. EXPLAIN/EXPLAIN ANALYZE across PostgreSQL, MySQL, MongoDB, Cassandra, ClickHouse. Index strategies (B-tree, hash, GIN, GiST, BRIN, partial, covering, composite). N+1 detection, cursor-based pagination, materialized views, query plan analysis, slow query diagnosis. Use when optimizing slow queries, designing indexes, or analyzing query performance.
npx claudepluginhub rnavarych/alpha-engineer --plugin role-databaseThis skill is limited to using the following tools:
Load from `references/` based on what's needed:
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Designs, implements, and audits WCAG 2.2 AA accessible UIs for Web (ARIA/HTML5), iOS (SwiftUI traits), and Android (Compose semantics). Audits code for compliance gaps.
Load from references/ based on what's needed:
EXPLAIN analysis for PostgreSQL (red flags, buffer hit ratio), MySQL (type column, FORMAT=TREE), MongoDB (executionStats, COLLSCAN vs IXSCAN), ClickHouse (system.query_log). Index type comparison table (B-tree, Hash, GIN, GiST, BRIN, partial, covering). Composite index design rules (equality first, range last, selectivity order). Partial indexes and covering indexes with INCLUDE. Statistics and cardinality management. Load when: analyzing query plans, designing indexes, or fixing stale statistics.
N+1 detection via pg_stat_statements, elimination patterns (eager load, DataLoader, window functions). Offset vs cursor-based pagination with multi-column cursor example. Materialized views: creation, blocking vs concurrent refresh, automated refresh strategies. Query anti-patterns: SELECT *, functions on indexed columns, correlated subqueries, implicit type conversions, missing LIMIT. Load when: fixing N+1 problems, implementing pagination, or eliminating slow query patterns.