Agent-Skills.md

Agent Skills: Performance Auditor

Expert assistant for monitoring and optimizing performance in the KR92 Bible Voice project. Use when analyzing query performance, optimizing database indexes, reviewing React Query caching, monitoring AI call costs, or identifying N+1 queries. Helps diagnose slow operations across database, frontend, and AI systems.

UncategorizedID: Spectaculous-Code/raamattu-nyt/performance-auditor

Repository

Spectaculous-Code/raamattu-nyt
Spectaculous-Code

Install this agent skill to your local

pnpm dlx add-skill https://github.com/Spectaculous-Code/raamattu-nyt/tree/HEAD/.claude/skills/performance-auditor

Skill Files

Browse the full folder contents for performance-auditor.

Download Skill

Loading file tree…

.claude/skills/performance-auditor/SKILL.md

Skill Metadata

Name
performance-auditor
Description
Expert assistant for monitoring and optimizing performance in the KR92 Bible Voice project. Use when analyzing query performance, optimizing database indexes, reviewing React Query caching, monitoring AI call costs, or identifying N+1 queries. Helps diagnose slow operations across database, frontend, and AI systems.

Performance Auditor

Expert performance optimization for the Raamattu Nyt application across database, frontend, and AI systems.

Quick Start

Choose your performance concern:

  • Database performance: Slow queries, missing indexes, sequential scans → See references/sql-queries.md
  • React Query optimization: Caching strategy, stale times, unnecessary refetches → See references/react-query-patterns.md
  • AI performance: Latency, costs, cache effectiveness → See references/ai-monitoring.md
  • N+1 query detection: Finding and fixing N+1 patterns → See references/n+1-detection.md
  • Optimization checklist: Complete performance audit → See references/checklist.md
  • Common gotchas: Known issues and pitfalls → See references/learnings.md

Performance Targets

Reference targets for healthy performance:

| Operation | Target | |-----------|--------| | Single verse lookup | <20ms | | Chapter load | <50ms | | Text search | <100ms | | AI translation | <500ms | | Page load (FCP) | <1.5s | | API response | <200ms |

How to Use This Skill

  1. Identify your performance issue (database, frontend, AI, or N+1)
  2. Read the appropriate reference file above
  3. Run SQL scripts or code patterns from scripts/ if needed
  4. Check references/learnings.md if results seem unexpected

Key Tools

  • Database analysis: Read references/sql-queries.md then run scripts/analyze-queries.sql
  • N+1 detection: See references/n+1-detection.md and scripts/detect-n+1.js
  • Cache optimization: See references/react-query-patterns.md for staleTime/gcTime strategy
  • AI monitoring: See references/ai-monitoring.md for latency and cost tracking

Context Files

For codebase structure reference:

  • Docs/context/db-schema-short.md - Database tables and indexes
  • Docs/context/supabase-map.md - Edge functions to monitor