nexus-gateway: Skill for Claude Code

.claude/skills/frontend-arch-review/SKILL.md

frontend-arch-review is a skill for Claude Code from AlphaBitCore/nexus-gateway. It costs 265 tokens per session (4,099 once invoked), scanned A, original, Apache-2.0.

A code-review checklist for checking that frontend colors, spacing, and other visual settings use the project's shared design tokens and support light and dark themes.

In plain words
What is it for?
Use it to review changes in the Control Plane UI, Agent Dashboard, and shared UI code, especially CSS modules, inline styles, Tailwind v4, and shadcn components.
Why use it?
It catches hard-coded styles that can break theme switching or make the interface inconsistent. This reduces the need to find visual problems manually across several frontend packages.

Skill for Claude Code

Written for Claude Code: user-invocable in frontmatter. Also seen: mentions CLAUDE.md.

This is AlphaBitCore/nexus-gateway's own configuration. It tells Claude Code how to work on nexus-gateway itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything nexus-gateway configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is node scripts/check-design-tokens.mjs --json > /tmp/violations.json.

Reuse

Borrowing it

Nothing to install: this file belongs to AlphaBitCore/nexus-gateway. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/AlphaBitCore/nexus-gateway/main/.claude/skills/frontend-arch-review/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/AlphaBitCore/nexus-gateway

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for frontend-arch-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/alphabitcore/nexus-gateway/frontend-arch-review/github.svg)](https://agentmods.dev/skills/alphabitcore/nexus-gateway/frontend-arch-review)
Your own site
<a href="https://agentmods.dev/skills/alphabitcore/nexus-gateway/frontend-arch-review"><img src="https://agentmods.dev/badge/skills/alphabitcore/nexus-gateway/frontend-arch-review/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for frontend-arch-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/alphabitcore/nexus-gateway/frontend-arch-review"><img src="https://agentmods.dev/badge/skills/alphabitcore/nexus-gateway/frontend-arch-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 265 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,099 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00265 $0.04099
Opus 5 $0.00133 $0.02049
Sonnet 5 $0.00053 $0.00820
Haiku 4.5 $0.00026 $0.00410

Measured 12d ago against content hash 4d66c66261bb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

frontend-arch-review 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 12d ago.

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.

.claude/skills/frontend-arch-review/SKILL.md · 297 lines

How it starts

The opening of the file, as written. The whole thing — 297 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Frontend Architecture Review

Verifies that every visual value in packages/control-plane-ui/, packages/agent/ui/frontend/, and packages/ui-shared/ resolves through the documented design-token layers — and that the recently-introduced Tailwind v4 + shadcn surface stays inside the token system instead of escaping back to raw palette utilities.

Current layer map:

  • Layer 1 — raw tokens. packages/ui-shared/src/styles/global.css (--g-* legacy palette + spacing/radius/shadow/transition constants).
  • Layer 1.5 — Tailwind v4 @theme {} raw palette. Defined in packages/ui-shared/src/styles/prime-shadcn-tokens.css (--color-navy-900, --color-brand, --color-text-secondary, --font-sans, …). Tailwind resolves utilities like text-foreground and arbitrary values like bg-[var(--color-primary)] against this block.
  • Mode flip. prime-shadcn-tokens.css :root defines light values for the shadcn vars (--background, --foreground, --card, --primary, --sidebar, …); .dark, [data-theme="dark"] re-binds them. Theme switching is driven by <html data-theme="…"> AND <html class="dark"> (set together by ThemeProvider).
  • Layer 2 — semantic bridge. light.css / dark.css map the legacy --color-* / --sidebar-* / --shadow-* names onto the shadcn vars so the 230+ existing CSS Modules keep working.
  • Component surface. Mostly *.module.css (~234 files) plus a small but growing set of shadcn-style components in packages/ui-shared/src/shadcn/ that use Tailwind utility classes, all of which must resolve through the token layers (never raw bg-white / text-gray-500).

This skill is the user's single entry point for "is the frontend architecturally clean for theme + mode switching, including the new Tailwind v4 + shadcn surface".

What it detects

Category Where Why it matters
Hex / rgba / hsla literals in *.module.css *.module.css outside theme-definition files Doesn't flip with data-theme="dark" / .dark → wrong contrast in dark mode
Hex / rgba inside style={{}} in .tsx All *.tsx Same — inline styles bypass token system
Numeric padding/margin/gap/fontSize/fontWeight/borderRadius/transition/zIndex in inline styles All *.tsx Doesn't break mode but breaks density/skin variation; structural inconsistency
Stale var(--xxx, #fff) references .tsx and .module.css The fallback hex is what actually renders if the token name was never defined — a silent breakage. Cleanup pass renames to the real semantic token and drops the fallback.
Recharts hex props (fill="#…" / stroke="#…") .tsx only Should come from packages/ui-shared/src/theme/chartColors.ts so the chart flips with mode
Raw Tailwind palette utilities (bg-white, text-gray-500, border-slate-200, dark:bg-zinc-900, etc.) .tsx, cva()/cn() strings Tailwind utilities that resolve to raw palette hex bypass the semantic shadcn tokens — must use bg-background / text-foreground / bg-card / bg-[var(--color-…)] instead
Hardcoded English in JSX .tsx Delegated to i18n-gap-check skill

Read the full file on GitHub · 297 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.

  1. 12d ago First seen · 297 lines · 265 tokens per session scan A 4d66c66261bb

Subscribe to this mod's changes

frontend-arch-review is a skill published in the GitHub repository AlphaBitCore/nexus-gateway (23 stars, last pushed 3d ago), licensed Apache-2.0. It adds 265 tokens to every session and 4,099 once invoked, about $0.0013 per session on Opus 5. 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 2026-08-30.

Related

Other skills, from other repositories

generic-static-code-reviewer

Review static site code for bugs, security issues, performance problems, accessibility gaps, and CLAUDE.md compliance. Enforces pure HTML/CSS/JS standards, minimal page weight, mobile-first design. Use when completing features, before commits, or reviewing changes.

travisjneuman/.claude · 59 tokens

audit-expert

Expert-level security auditing, compliance, code review, and vulnerability assessment. Use when the user mentions compliance, security review, code review, vulnerability assessment, SOC 2, or GDPR, or when the task involves Audit Types, Audit Frameworks, Audit Process, or Authentication Review.

personamanagmentlayer/pcl · 60 tokens

accessibility-audit

Fast, high-signal accessibility triage for pages, components, or PRs targeting WCAG 2.2 AA compliance.

NickCrew/Claude-Cortex · 31 tokens

design-systems

Principal-level design system practice — design tokens, multi-platform theming, component API design, accessibility built-in, versioning + governance, contribution model, documentation, and the discipline that turns "one team's component library" into a load-bearing capability for every product surface — including…

Nmor/the-claude-council · 102 tokens

interaction-design

Principal-level interaction design — affordances, signifiers, feedback, mappings, constraints, error prevention, recovery, gesture + input model design, microcopy, motion as functional language, and the discipline that turns flows into experiences users complete without thinking and don't have to recover from.

Nmor/the-claude-council · 58 tokens

wcag-accessibility

WCAG 2.2 AA + AAA accessibility patterns for every user-facing surface — semantic HTML, ARIA 1.2, keyboard navigation, screen reader support, color contrast, motion preferences, and the 9 new SCs introduced in WCAG 2.2. Also lazy-loads a11y.md content migrated from rules/common/ on 2026-06-02.

Nmor/the-claude-council · 84 tokens