Official WordPress skill for Gutenberg block work: block.json, attributes and serialization, dynamic rendering, and the deprecation path that keeps existing content valid.

React Native Best PracticesSkill
Summary
Callstack's official React Native performance skill: a priority-ordered framework for FPS, time-to-interactive, bundle size, memory leaks and re-render problems.
Features
- Priority-ordered guidelines separating JS, native and bundling causes
- Nine JavaScript/React guides: profiling, lists, animations, memory
- Nine native guides: Turbo Modules, threading, Hermes, native profiling
- Nine bundling guides: tree shaking, code splitting, size analysis
- FlashList and list virtualisation guidance for scroll performance
- Time-to-interactive and startup optimisation
- Supply-chain and remote-chunk-loading security rules
Install This Skill
Add this skill to your favorite AI agent in a few steps.
Skill Content
Usage Instructions
Learn how to use this skill with different AI agents.
Description
What it is
Callstack's official agent skill for React Native performance work. Callstack are long-standing React Native core contributors and consultants, and this skill distils their "Ultimate Guide to React Native Optimization" into guidance a coding assistant can apply while it works.
What it is for
Performance problems in React Native are hard to delegate to an assistant because the symptom rarely indicates the layer at fault. Dropped frames might be a JS thread block, an over-eager re-render, a bridge round-trip, or a native measurement pass — and the fixes are entirely different. This skill gives the agent a priority-ordered framework for telling them apart.
Reach for it when:
- debugging janky UI, frame drops or slow animations
- investigating memory leaks, in JavaScript or native
- optimising time-to-interactive and app startup
- reducing bundle or installed app size
- writing Turbo Modules or other native modules
- reviewing React Native code specifically for performance
What it covers
Guidance spans three layers, with roughly nine reference guides each:
- JavaScript and React — profiling, list virtualisation (including FlashList), animation strategy, re-render elimination, memory
- Native (iOS and Android) — Turbo Modules, threading, Hermes tuning, native profiling
- Bundling — tree shaking, code splitting, and bundle size analysis
The references sit in separate files and are pulled in on demand, so the agent loads the bundle-analysis material only when it is actually analysing a bundle.
A note on its security guidance
Unusually for a performance skill, it carries explicit supply-chain instructions: treat the shell commands in its references as local developer operations to review rather than paste, pin tooling versions instead of piping remote scripts to a shell, and apply normal provenance controls to third-party libraries. Its rule for remote chunk loading is the strictest — hosted chunks must come from HTTPS origins you control and be pinned to the current app release, with app-bundled chunks or signed CI manifests preferred. That is sound advice for anyone shipping over-the-air updates, skill or no skill.
Installing
MIT licensed. Install the single skill with npx skills@latest add callstackincubator/agent-skills --skill react-native-best-practices.
Related Skills
Persistent cross-session memory for coding agents: hooks capture each session, a local SQLite + vector store compresses it, and a mem-search skill reads it back.
The master skill in PlanetScale's official pack: runs a full read-only best-practices assessment of a database or org, then produces one evidence-backed report.
Build crash-safe AI agents on AgentKit and step.ai — durable tool calls, human-in-the-loop approval, realtime progress, and when not to reach for an agent loop at all.
