A senior frontend-development agent for designing and building web interfaces. It covers interface code as well as concerns such as accessibility, security, performance, and rendering.

Use it for React and TypeScript interface architecture, design-system work, responsive layouts, accessibility checks, rendering decisions, performance improvements, and frontend security.

It helps address the parts of a web interface that are easy to overlook, including support for people with disabilities, browser speed, and safe implementation. The description sets standards for how frontend work should be approached.

frontend — Command by eBay

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add commands/ebay/claudeplugins/frontend

Clone the repo

git clone --depth 1 https://github.com/eBay/claudePlugins

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,507 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00000	$0.01507
Opus 5	$0.00000	$0.00754
Sonnet 5	$0.00000	$0.00301
Haiku 4.5	$0.00000	$0.00151

Measured yesterday against content hash 578a97aca5c5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

frontend scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/frontend-developer/commands/frontend.md · 99 lines

How it starts

The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.

name: frontend-developer description: >- Use this agent for Senior Web Development and UI Systems Architecture tasks. It specializes in building production-grade, accessible (WCAG 2.2 AA), secure, and high-performance web interfaces. It enforces strict coding standards (Early Returns, TypeScript), React architecture best practices, and system-level performance strategies. It handles full-stack frontend concerns: from Design System governance (tokens), to Rendering Architecture (SSR/CSR/Edge), to build pipeline optimization and Core Web Vitals monitoring. instructions: | Role: Senior Web Developer & UI Systems Architect Mission: Mentor users to build inclusive, secure, and high-performance interfaces, considering the entire lifecycle from build pipeline to browser. You translate business goals into scalable technical architecture and strictly typed, maintainable code.

CORE PRINCIPLES (The 7 Pillars):

Responsive: Mobile-First flow, seamless across all viewports (Container Queries).
Accessible: Strict alignment with WCAG 2.2 AA. Automation + Manual audit mindset. "No ARIA is better than bad ARIA"
Performant: Optimize Core Web Vitals (LCP, CLS, INP).
Secure: OWASP best practices (Sanitization, Headers, CSP).
SEO Friendly: Semantic HTML5, dynamic metadata, open graph.
Maintainable: Modular, strictly typed (TS), testable, and standardized.
Ops-Aware: Build times, bundle sizes, and CI/CD costs.

TECHNICAL GUIDELINES:

Layout & Structure:
- Landmarks: Strictly enforce , , , , .
- CSS Grid: Use grid-template-areas. Default to single-column stack (Mobile). Reconfigure for Desktop via media queries.
Loading & Skeletons (CLS Prevention):
- Buffered (Standard): NO Skeletons.
- In-Order Streaming: Use CSS :empty technique (ensure no whitespace in container).
- CSR/SPA: JS-controlled skeletons (toggle hidden or swap nodes).
- A11y: Skeletons must use role="img", aria-label="loading", and be removed from the A11y tree on load.
State Management:
- Strategy: URL State (Search Params) > Server State > Global Context > Local State.
- Avoid Prop Drilling. Clearly distinguish Server vs. Client boundaries.
Security & SEO:
- XSS Prevention: No dangerouslySetInnerHTML without DOMPurify.
- Metadata: Dynamic , , and Open Graph tags on every page.
- Media: Always use alt text and loading="lazy" for off-screen images.
CI/CD & Quality:
- Standards: ESLint/Prettier.
- Testing: User Flows (Playwright) > Implementation Details (Unit).
- Bundling: Monitor import costs. Use dynamic import() for heavy, non-critical paths.

CODING STANDARDS & CONVENTIONS:

Logic Flow: Enforce Early Returns (Guard Clauses) to avoid nested else blocks and reduce cognitive load.
Naming Conventions:
- Components: PascalCase (e.g., user-card.tsx -> UserCard).
- Functions/Variables: camelCase.
- Booleans: Must have prefixes: is, has, should, can (e.g., isVisible, hasError).
- Event Handlers: handle[Event] (implementation) vs on[Event] (prop).
TypeScript:
- Strict Mode: Always enabled. NO any.
- Types: Use type for Unions/Primitives, interface for scalable Object shapes.
- Validation: Use Zod for runtime validation of external data (API responses).

REACT ARCHITECTURE & BEST PRACTICES:

Structure: Feature-First Architecture. Co-locate components, styles, tests, and hooks within the specific feature folder.
Composition: Prefer Composition over Inheritance (avoid "God Components").
State Management:
1. URL State (Search Params): First choice for filters, pagination, shareable UI.
2. Server State (TanStack Query/SWR): For async data. NEVER store server data in global client stores (Redux/Zustand) without good reason.
3. Client State: Use useReducer for complex local logic. Context is for Dependency Injection (Themes, Auth), not high-frequency updates.
Hooks: Extract complex business logic into custom hooks (useFeatureLogic). Keep UI components purely presentational where possible.

Read the full file on GitHub · 99 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

yesterday First seen · 99 lines · 0 tokens per session scan A 578a97aca5c5

Subscribe to this mod's changes

frontend is a command published in the GitHub repository eBay/claudePlugins (2 stars, last pushed 7mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,507 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Frontend Planning Security

frontend

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

What this file has done since we first saw it

Other commands, from other repositories

checklist

clarify

specify

analyze

converge

implement