senior-frontend
Skill48 starsChanged 40 days agoApache-2.0View on GitHub
Build modern, performant frontend UIs: component design, state management, rendering performance, and accessibility. Covers React and Next.js (web) plus React Native, with framework-specific mechanics in reference files; the judgment applies cross-framework. Use whenever the user builds or optimizes UI components, manages client or server state, or fixes frontend performance including list virtualization and scroll or render jank (e.g. FlatList). Do NOT use for visual or aesthetic design direction (use frontend-design), or for animation and gesture craft (use motion-craft).
29 other projects use a version of this.
Adopted by
Most recently taken up by
- darkweb19/skillstheir own version, first seen Nov 8, 2026
- kazemcodes/skill-treetheir own version, first seen Nov 8, 2026
- codeApe-7/ai-agent-workflowGrouptheir own version, first seen Nov 1, 2026
- nguyenthanhtat/screen1-claudetheir own version, first seen Nov 1, 2026
- OpenxIntelligence/Coding-Agent-Skillstheir own version, first seen Nov 1, 2026
- weininghui/skillstheir own version, first seen Oct 30, 2026
Used with
Projects with it also use senior-qa, ui-ux-pro-max and 3 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
nek1987Skillsenior-qa
Comprehensive QA and testing skill for quality assurance, test automation, and testing strategies for ReactJS, NextJS, NodeJS applications. Includes test suite generation, coverage analysis, E2E testing setup, and quality metrics. Use when designing test strategies, writing test cases, implementing test automation, performing manual testing, or analyzing test coverage.
Also in 3 projects that use this
StrayDragonSkillui-ux-pro-max
UI/UX design intelligence. 50 styles, 21 palettes, 50 font pairings, 20 charts, 8 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app, .html, .tsx, .vue, .svelte. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient.
Also in 5 projects that use this
JetBrainsSkilltheme-factory
Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors/fonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.
Also in 3 projects that use this
VercelSkillvercel-react-best-practices
React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.
Also in 3 projects that use this
OpenAISkillbrainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
Also in 3 projects that use this
Other versions29 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 29
- Ohh-889/skyroc-admin★ 795same name
- roedyrustam/vibes-plug★ 75same name
- tranhieutt/software_development_department★ 71same name
- nilecui/SkillsBase★ 27same name
- leadgenjay/claude-skills★ 22same name
- codeApe-7/ai-agent-workflowGroup★ 18same name
- Hassaan146/claude-skills★ 18same name
- EmersonBraun/skills★ 4same name
- I-Onlabs/claude-code-skills★ 4same name
- nek1987/auto-agent-harness★ 4same name
19 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository31 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- code-reviewerSkill
- commit-workSkill
- establish-charterSkill
- frontend-designSkill
- grill-with-docsSkill
- motion-craftSkill
- plan-with-docsSkill
- review-codebaseSkill
- review-workSkill
- senior-architectSkill
- senior-backendSkill
- senior-frontendSkill
- tddSkill
- update-docsSkill
- work-stepSkill
- code-reviewerSkill
- commit-workSkill
- establish-charterSkill
- frontend-designSkill
- grill-with-docsSkill
- motion-craftSkill
- plan-with-docsSkill
- review-codebaseSkill
- review-workSkill
- senior-architectSkill
- senior-backendSkill
- tddSkill
- update-docsSkill
- work-stepSkill
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 4 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 28 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 28
Sizing threshold: keep a component focused. When it pushes past ~200 lines or starts doing two unrelated jobs (fetching *and* laying out …
- Distinctive: line 47
Lift when shared: When two siblings need the same value, lift it to their nearest common parent — and no higher.
- Distinctive: line 82
Semantic elements first: Use the real interactive primitive (a button, a link, a labeled input) rather than re-implementing one on a …
- Very specific: line 116
Next.js: when you want server rendering, routing, and data-fetching conventions out of the box — content sites, app shells, anything that …
- Distinctive: line 137
Read the file that matches your target before writing framework-specific code:
- Very specific: line 150
After making any code changes, run codument context --file <path> --owner to find the owning doc — one line, not a read of the whole …
“Common” is measured: kept by at least 6 in 10 of the 28 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Frontend Development
This skill is the platform-neutral judgment layer for building UIs. The principles below hold whether you target the web or native. Framework-specific mechanics (the actual API names, files, and config) live in the reference files at the end — read the one that matches your target before writing code.
Component design
The unit of a UI is a component. Get its size and boundaries right and most other problems shrink.
- One component per file, named to match the export. Co-locate its styles, tests, and types.
- Sizing threshold: keep a component focused. When it pushes past ~200 lines or starts doing two unrelated jobs (fetching and laying out and branching on three states), split it. Length is a smell, not a law — a long but linear render is fine; a medium one juggling four concerns is not.
- Extraction threshold: extract a shared component only when it is genuinely reused (the rule of three — pull it out at the third copy, not the first). Premature shared components calcify into the wrong abstraction. Duplication is cheaper to fix than the wrong shape.
- Props are a contract: type them precisely. Prefer a closed set
(
"sm" | "md" | "lg") over an open one (string). The narrower the prop type, the more the compiler does your reviewing for you.
State management
Reach for the least powerful tool that holds the state, and escalate only when you feel the pain. The decision order:
- Local first. State that only one component cares about lives in that component. Most state is local.
- Lift when shared. When two siblings need the same value, lift it to their nearest common parent — and no higher. Lifting too far makes everything below re-render.
- Context for truly global, low-churn state. Theme, auth, locale, current user. Context is for things that are genuinely app-wide and change rarely. It is not a fix for prop drilling through two levels, and it is not a state manager — every consumer re-renders when the value changes.
- External store for high-churn or server-cache state. Cross-cutting state that updates often, or that mirrors the server, belongs in a dedicated store or data-fetching library, not in Context.
The rule that overrides all four: derive, don't sync. Never copy one piece of state into another and keep them in lockstep with effects. If a value can be computed from existing state or props, compute it during render. State you sync is state that drifts.
Rendering performance
- Measure before you memoize. Memoization is not free — it adds allocation and complexity, and most components are already fast enough. Profile first, find the component that actually re-renders too often or computes something genuinely expensive, then memoize that one. Sprinkling memoization preventively makes code slower to read and rarely faster to run.
- Stable identity matters. Lists need stable, unique keys tied to the data, not array indices. Callbacks and objects passed to memoized children must keep a stable reference or the memoization is defeated.
- Defer what isn't needed yet. Lazy-load routes and heavy, below-the-fold, or rarely-hit components. Split the bundle along the lines users actually traverse.
- Big lists need virtualization. Rendering thousands of rows at once janks. Use the platform's windowing/virtualization primitive so only visible rows mount.
Accessibility by default
Accessibility is not a pass at the end — it is the default way you build.
- Semantic elements first. Use the real interactive primitive (a button, a link, a labeled input) rather than re-implementing one on a generic container. You get focus, keyboard handling, and assistive-tech semantics for free.
- Every control is reachable and labeled. Keyboard (or platform equivalent) must reach every action; every interactive element exposes an accessible name.
- Visible focus. Interactive elements show a clear focus state. Never strip it without replacing it.
- Adequate contrast. Text meets a 4.5:1 contrast ratio.
- Accessibility APIs are the escape hatch, not the default. Reach for explicit accessibility attributes only when the semantic primitive can't express it.
Loading and errors
- Every loading state needs an error state. Async work has three outcomes — pending, success, failure. A spinner without an error path is a bug waiting for a flaky network. Design all three before you ship the happy path.
- Don't manually sync server state. Server data is a cache, not local state. Use the platform's data-fetching layer so caching, revalidation, and error states come built in.
Architecture hygiene
- Keep data-fetching out of the render body. Extract API calls into hooks, loaders, or server-side code — components should consume data, not orchestrate its retrieval.
- Test behavior, not implementation. Assert what the user sees and does, not internal state names. Tests coupled to implementation break on every refactor.
Framework selection
Choose the target before reaching for mechanics:
- React (web) for in-browser SPAs and embedded UI where you control the host page.
- Next.js when you want server rendering, routing, and data-fetching conventions out of the box — content sites, app shells, anything that benefits from server components and SSR/SSG.
- React Native for iOS/Android apps from a shared codebase. The judgment above transfers; the primitives (views instead of DOM, native navigation, list virtualization, platform APIs) differ — see its reference.
The judgment on this page is the same across all three. Only the mechanics differ, and those live in the references below.
Motion and animation
For anything beyond a basic, static transition — deciding whether to animate,
choosing easing/duration/springs, building drawers/toasts/sheets, gestures and
drag-to-dismiss, or reviewing and fixing janky motion — invoke the motion-craft
skill instead of hand-rolling it. It covers both web (CSS/WAAPI/Motion) and React
Native (Reanimated + Gesture Handler). Do not duplicate motion guidance here or in
the references.
References
Read the file that matches your target before writing framework-specific code:
references/react.md— read this when targeting React or Next.js on the web: Server Components and"use client",loading.tsx/error.tsx,next/dynamic,React.memo/useMemo/useCallback, Tailwind (@apply,tailwind.config), web performance, and HTML/ARIA accessibility.references/react-native.md— read this when targeting React Native (Expo or bare): StyleSheet/Flexbox and NativeWind, FlatList/SectionList/FlashList virtualization, React Navigation / Expo Router, Hermes and bridge performance, Platform APIs, and RN accessibility props.
Definition of Done
After making any code changes, run codument context --file <path> --owner to find the owning doc — one line, not a read of the whole registry — and update it.
This is part of the project's Definition of Done.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

