# Performance Pass

> Use when changes might impact latency, throughput, rendering, memory, or DB load. Do NOT micro-optimize; focus on obvious bottlenecks and regression prevention.

- Skill: `askaret/performance-pass` (Agent Skill)
- Install (CLI): `npx skillmds@latest add askaret/performance-pass`
- Raw SKILL.md: https://api.skillmd.com/api/skills/askaret/performance-pass/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: askaret (https://skillmd.com/u/askaret)
- Updated: 2026-09-10
- Page: https://skillmd.com/skills/askaret/performance-pass

---


Backend checklist:
- API
  - Pagination defaults for list endpoints
  - Avoid N+1 patterns; verify includes/joins
  - Add caching headers or server caching where appropriate
  - Validate payload sizes and streaming for large responses
- DB
  - Verify indexes align with query predicates and sort order
  - Check migration/backfill impacts (locks, write amplification)
  - Watch for chatty queries
- Runtime
  - Avoid unnecessary allocations in hot paths
  - Ensure async usage is correct (no sync-over-async)
- Guardrails
  - Add perf-focused tests/benchmarks only if repo already uses them

Frontend checklist (React):
- Rendering
  - Avoid unnecessary re-renders; use memoization only where it helps
  - Verify list virtualization for large lists if relevant
- Data fetching
  - Avoid duplicate requests; handle loading/error states cleanly
  - Ensure caching strategy is consistent (query libs, cache keys)
- Bundling
  - Avoid adding heavy dependencies without justification
  - Ensure code splitting for large new routes/features if repo supports it

Verification:
- Run build and measure obvious regressions (build output, bundle warnings)
- Smoke test core flows affected
- If repo has profiling tools/scripts, run them and report results

Finish with:
- Potential perf risks identified
- Changes made (if any) to mitigate
- Commands run + results
- Follow-ups (monitoring, dashboards, load testing)