build-polished-visual-frontends

build-polished-visual-frontends is a skill for Claude Code, Codex from EverMind-AI/Raven. It costs 142 tokens per session (13,148 once invoked), scanned A, original, Apache-2.0.

A guide for building polished websites and other interfaces consumed directly in a web browser, using HTML, CSS, JavaScript, or web frameworks. It focuses on the browser's final visual result while partnering with a domain guide for games, maps, charts, simulations, or other specialist content.

In plain words
What is it for?
Use it to create or improve websites, browser-based products, data stories, maps, diagrams, simulations, games, and web-to-print layouts with a deliberate visual system and working interactions.
Why use it?
It helps turn a domain model into a coherent web experience rather than an untested shell or temporary viewer. It checks real tasks, interaction feedback, responsive behavior, visual consistency, and the final pixels users see.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: $skill-name invocation.

Good fit Use it to create or improve websites, browser-based products, data stories, maps, diagrams, simulations, games, and web-to-print layouts with a deliberate visual system and working interactions.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/evermind-ai/raven/build-polished-visual-frontends
About the project

Raven is an open-source agent harness for running long-term AI work with terminal execution, tracing, memory, skills, evaluation, and reusable workflows. People use the current release to operate and improve persistent AI workflows, while its described future direction is a multi-agent system that combines specialized harnesses.

EverMind-AI/Raven · 3,825 stars · on GitHub · raven.evermind.ai

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.

Any agent
npx skills add EverMind-AI/Raven --skill build-polished-visual-frontends
Clone the repo
git clone --depth 1 https://github.com/EverMind-AI/Raven

Made for: Claude Code, Codex.

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 build-polished-visual-frontends

README.md
[![agentmods](https://agentmods.dev/badge/skills/evermind-ai/raven/build-polished-visual-frontends/github.svg)](https://agentmods.dev/skills/evermind-ai/raven/build-polished-visual-frontends)
Your own site
<a href="https://agentmods.dev/skills/evermind-ai/raven/build-polished-visual-frontends"><img src="https://agentmods.dev/badge/skills/evermind-ai/raven/build-polished-visual-frontends/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 build-polished-visual-frontends

Your own site · 80×15
<a href="https://agentmods.dev/skills/evermind-ai/raven/build-polished-visual-frontends"><img src="https://agentmods.dev/badge/skills/evermind-ai/raven/build-polished-visual-frontends.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 142 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 13,148 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.
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 $0.00142 $0.13148
Opus 5 $0.00071 $0.06574
Sonnet 5 $0.00028 $0.02630
Haiku 4.5 $0.00014 $0.01315

Measured today against content hash c402d34c02d1, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

build-polished-visual-frontends 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-dist/design-engine/raven_design/skills/build-polished-visual-frontends/SKILL.md · 595 lines

How it starts

The opening of the file, as written. The whole thing — 595 lines — stays where its author put it; the contents beside it link to each section on GitHub.

构建高完成度视觉前端

核心目标

让用户直接完成任务,让视觉、交互和专业对象来自同一组清楚的决定。让成熟工具解决已经 解决的问题,把模型的判断留给当前产品独有的任务、构图和体验。

遵循三个原则:

  1. 任务决定形态。 不因交付是 HTML 就默认应用壳、dashboard、landing page、卡片网格或 16:9。
  2. 一套视觉语言。 控件、导航、列表、面板、专业画布和反馈必须像同一个产品。
  3. 标准能力交给工具。 模型主要设计任务模型、工作顺序、整体构图和必要的连接层。

主要领域 Skill 决定内容模型、专业语义和领域门槛;visual-artifact-design 决定共享流程与 证据;本 Skill 只负责浏览器中的视觉方向、技术组合、工具边界和最终像素。用户当前要求与 已接受参考优先,其次是既有品牌、平台和领域约束。题材联想不能为任意颜色或装饰提供理由。

什么算好,八条判据(后文以代号引用):

  • J1 主角与胆量:每个交付面能指认主角时刻,且有一个超出安全默认、能一句话说出的决定。
  • J2 删测:删掉后页面不变差的元素,必须删。
  • J3 一句话可述:整个产物能被一句话描述(“这是一张……的页”);说不出即没有方向。
  • J4 具体胜过修辞:每个宣称可验证;用事实、真数字和真对象替代形容词。
  • J5 主视觉先行:内容型/营销型 Web 先用真实成图锁定世界、空间与取景,再围绕它排版; 不允许先完成模板图槽,再用一张可拆换的图补空。
  • J6 细节成系统:字体、标点、边界、图标、状态、运动与目标环境适配共同实现同一视觉命题; “没有命中反模式”不等于已精致。
  • J7 逐区成品:每个可见区块和页面族都有明确职责、任务特有的内容关系与完成态;不能精修 首屏后,把其余部分留成默认组件、文字倾倒或模板余料。
  • J8 示能诚实:任何看起来可点击、可选择、可展开或可切换的东西都兑现相应行为与反馈; 静态内容不借用按钮、选中态、悬停态或可点击卡片的视觉暗示。

1. 先读懂任务,再决定画什么

新建方向或大改前,先写一张短决策卡:

任务: 谁在什么场景完成哪一件主要事情
主要对象: 页面中真正需要看、读、编辑或操纵的对象
自然形态: 页面、工作台、长文、画布、地图、图解、游戏或其他形态
领域惯例: 该领域真实从业者交付此物时的成品形态与表达习惯;偏离惯例必须写出理由
参考与偏好: 实际看见的参考,以及用户明确喜欢和拒绝的特征
表达强度: quiet | editorial | expressive
视觉系统: 哪一套成熟设计语言负责字体、控件、表面、图标和反馈
专业引擎: 哪个领域工具负责图表、地图、排版、媒体、画布或其他核心对象
衔接合同: 两者如何共享 token、状态、字体、颜色、焦点、图例和导出
目标画面: 最能决定方向的尺寸、内容密度和关键状态

真实参考依次来自用户提供的画面、已接受的同项目产物、实际生产产品和成熟公共系统。参考先 回答领域惯例:该领域真实从业者交付这类产物时,成品长什么样、用什么表达;产物先像该领域的 真实交付物,再谈风格取向。必须看到真实画面再提取 3–6 个可观察特征;品牌印象、文章描述和 模型记忆不算视觉证据。需要自行 研究时,只查看覆盖主要风险的少量真实工作态、密集态和合同所列目标环境,不照搬独特版式、资产或 文案。按需读取审美路由

表达强度由任务和参考决定:

强度 常见任务 默认表现
quiet 产品、数据、地图、技术工作 让主要对象与状态主导;层级来自比例、字重、间距和关系
editorial 内容网站、字体、演示、出版 用排版、图像角色、节奏和叙事顺序形成方向
expressive 品牌、营销、插画、游戏 可使用强色彩和动效,但每个强信号都承担传播、叙事或玩法职责

内容型/营销型 Web 的前置参考分流

内容型或营销型 Web 在形成视觉方向之前,先分别锁定两条路线并写入 REFERENCE-CONTRACTidentity_route 决定 标识、色、字与既有视觉语言;layout_route 决定宏观结构与阅读顺序。identity_routeaccepted_systemnonelayout_routepool_templateobserved_benchmark,不因 identity_route 已锁定而省略。合同中至少 一条宏观结构/阅读顺序关系必须来自 layout_route 的来源画面。

Read the full file on GitHub · 595 lines

Files

What ships with it

35 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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 First seen · 595 lines · 142 tokens per session scan A c402d34c02d1

Subscribe to this mod's changes

build-polished-visual-frontends is a skill published in the GitHub repository EverMind-AI/Raven (3,825 stars, last pushed today), licensed Apache-2.0. It adds 142 tokens to every session and 13,148 once invoked, about $0.0007 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 2026-09-12.