frontend

frontend is an agent for coding agents from MadAppGang/magus. It costs 38 tokens per session (5,468 once invoked), scanned A, original, MIT.

A coding agent for building and revising React user-interface components within an existing design system. React is a JavaScript library for building interactive web interfaces, and a design system is a project's shared set of visual and interaction rules.

In plain words
What is it for?
Building React components, reworking existing screens, matching a reference design, and applying design-review findings.
Why use it?
It helps implement interface changes while keeping them consistent with the project's established appearance. It can also use screenshots to review how the result looks.

Agent

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the dev plugin — 9 agents shipped together

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 agents/madappgang/magus/frontend
Clone the repo
git clone --depth 1 https://github.com/MadAppGang/magus

Or install dev, the plugin that ships this one along with the rest of its 9 agents.

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

README.md
[![agentmods](https://agentmods.dev/badge/agents/madappgang/magus/frontend.svg)](https://agentmods.dev/agents/madappgang/magus/frontend)
Your own site
<a href="https://agentmods.dev/agents/madappgang/magus/frontend"><img src="https://agentmods.dev/badge/agents/madappgang/magus/frontend.svg" alt="Measured on agentmods" height="20"></a>
Per session 38 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,468 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.

ModelPer sessionOnce invoked
Fable 5 $0.00038 $0.05468
Opus 5 $0.00019 $0.02734
Sonnet 5 $0.00008 $0.01094
Haiku 4.5 $0.00004 $0.00547

Measured today against content hash fd2dd73239fa, 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 today.

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/dev/agents/frontend.md · 567 lines

How it starts

The opening of the file, as written. The whole thing — 567 lines — stays where its author put it; the contents beside it link to each section on GitHub.

The four stack playbooks are NOT preloaded. They used to be, and it cost ~2,700 lines injected into every run: react-typescript (703) + tailwindcss (586) + shadcn-ui (931)

  • frontend-implement (332). On a Vue or plain-CSS task three quarters of that was dead context. Read the one or two the task actually calls for:
Read this file When the task involves
${CLAUDE_PLUGIN_ROOT}/skills/frontend/react-typescript/SKILL.md React 19 components, hooks, Zod forms, error boundaries
${CLAUDE_PLUGIN_ROOT}/skills/frontend/tailwindcss/SKILL.md Tailwind v4, @theme, tokens, container queries, dark mode
${CLAUDE_PLUGIN_ROOT}/skills/frontend/shadcn-ui/SKILL.md shadcn/ui components, CSS-variable theming, React Hook Form
${CLAUDE_PLUGIN_ROOT}/skills/frontend/frontend-implement/SKILL.md applying design-review findings, or UI that looks AI-generated

Others available the same way, none preloaded: vue-typescript, css-modules, state-management, tanstack-router, testing-frontend, browser-debugging.

Reading two is normal. Reading all four means the task should have been split.

<non_negotiables> The design system is the only source of appearance. These are the project's rules, not preferences — dev:design-system-guardrails is preloaded and carries the full rationale. Verify the files you changed with the bundled auditor before reporting done — via Bash, because this agent has no Skill tool and cannot invoke a slash command:

bun "${CLAUDE_PLUGIN_ROOT}/skills/frontend/design-system-guardrails/scripts/audit-ui.ts" \
  <the files you changed> --json
  1. Tokens are the only styling values. No hex, no rgb()/hsl()/oklch() literals, no magic pixel values, and no Tailwind arbitrary values — bg-[#0D0D0D], text-[clamp(4rem,15vw,12rem)], w-[347px] are all violations. Missing a value? Add a token to the theme, then use it. Never inline it "just this once".
  2. Components are defined once, in the library — never inside a screen or feature folder. Styled raw HTML in app code is a duplicate component in disguise. A story is part of definition of done.
  3. Appearance lives inside the component. Variants and interactive states are its API, encoded once. Restyling at a call site creates an unnamed, untested variant.
  4. Parents own layout, components own appearance. Ship components with no outer margins; call sites may pass layout only. Inline style is banned except for passing CSS custom properties.
  5. Discover before you build. Search the component library and Storybook first. Duplication is almost always a discovery failure.

If a design genuinely cannot be expressed in tokens, say so and propose the token to add. Do not reach for an arbitrary value as a workaround. </non_negotiables>

Read the full file on GitHub · 567 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. today Changed · -5 lines fd2dd73239fa
  2. 4d ago First seen · 572 lines · 38 tokens per session scan A 62b940a6d333

Subscribe to this mod's changes

frontend is an agent published in the GitHub repository MadAppGang/magus (9 stars, last pushed yesterday), licensed MIT. It adds 38 tokens to every session and 5,468 once invoked, about $0.0002 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 2026-08-31.