本文へ移動
cccskills
無料GitHub で公開

code-review-web

Reviews browser-rendered HTML, CSS, JSX, TSX, JavaScript, and TypeScript for accessibility, XSS, forms, focus, rendering performance, styles, and lifecycle cleanup. Loaded by dh:code-reviewer when source or framework entrypoints establish frontend execution; composes with TypeScript checks.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md5.7 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Web Frontend Code Review Patterns

Load these checks for browser-targeted HTML, CSS, JavaScript, TypeScript, JSX, and TSX. Confirm the rendering/runtime path from source and relevant framework configuration; apply TypeScript checks alongside these checks when applicable.

Read Review principles before applying these checks; it defines authority, applicability, evidence, and blocking criteria.

Accessibility

  • Every interactive element (button, link, input, select) must have an accessible name — either visible text, aria-label, or aria-labelledby
  • Verify focus behavior for the component's semantics: modal interaction needs appropriate containment and restoration; do not impose a modal focus trap on an ordinary nonmodal panel.
  • Check text contrast and large-text eligibility against the target's applicable accessibility standard; record the actual foreground/background and sizing evidence rather than guessing from visual appearance.
  • Icon-only buttons without visible text must have aria-label or a visually-hidden text span
  • Verify form controls have programmatically associated names/labels appropriate to their semantics; visual proximity alone does not establish that relationship.
  • Images conveying information must have descriptive alt text; decorative images use alt=""
  • Keyboard navigation must work — focus order must follow visual order, no focus traps outside intentional modal patterns

XSS Prevention

  • element.innerHTML = userValue is a blocking finding — use textContent for plain text
  • dangerouslySetInnerHTML (React) or equivalent without sanitization is a blocking finding
  • User-controlled values used in eval(), Function(), or setTimeout(string) are a blocking finding
  • Validate user-controlled URL schemes for the actual attribute and content context; prevent executable or unauthorized navigation/content while preserving deliberately supported media sources.

Performance

  • Investigate interleaved layout reads (offsetWidth, getBoundingClientRect) and style writes for repeated layout work; establish the affected rendering path and cost before assigning severity.
  • Check that image layout reserves the required space through dimensions or an equivalent supported layout rule; report observed or source-established layout-shift risk.
  • Review off-screen image loading against the critical rendering path and intended preload behavior; recommend lazy loading where it addresses a demonstrated cost.
  • Check the actual script-loading semantics and critical rendering effect of large bundles; absence of a particular attribute alone does not establish a performance defect.
  • Trace expensive render work to its input size, frequency, and user-visible cost before recommending memoization or another optimization.

CSS

  • Inspect !important for an actual cascade conflict or project-policy violation; a justified third-party override is not defective solely because of the keyword.
  • Apply the project's design-token conventions where they exist; do not impose a new token system on unrelated styles.
  • Trace undocumented z-index values to stacking behavior and project policy before requiring named tokens or a new scale.
  • position: fixed or position: sticky without overflow and scroll container awareness is flagged
  • Verify whether broad component selectors affect unintended descendants; identify the affected component and style consequence.

Forms

  • Verify user-operable fields have accessible names and any visible/programmatic labels required by their semantics and the target standard; preserve supported wrapping, association, or equivalent naming mechanisms.
  • Where the target accessibility standard requires identifying input purpose, verify the field's purpose and relevant autocomplete metadata.
  • Validation error messages must be associated with the field via aria-describedby — color alone is not sufficient to communicate errors
  • Form submission must not clear field values without user confirmation when validation fails

Event Listener Cleanup

  • Verify listeners stop affecting the component when its lifetime ends, through explicit removal, supported abort/disposal behavior, or an equivalent lifetime guarantee.
  • For document.addEventListener, trace the long-lived emitter and actual cleanup mechanism; report retained listeners or stale effects rather than requiring one cleanup spelling.
  • AbortController is preferred for fetch-and-cleanup patterns — pass the signal and abort on cleanup

Anti-Patterns

<!-- WRONG: no label, no accessible name -->
<input type="text" placeholder="Search..." />

<!-- RIGHT: associated label -->
<label for="search">Search</label>
<input type="text" id="search" autocomplete="off" />

<!-- WRONG: XSS vector -->
<div id="output"></div>
<script>
  document.getElementById("output").innerHTML = userInput;
</script>

<!-- RIGHT: safe text insertion -->
<script>
  document.getElementById("output").textContent = userInput;
</script>
/* Investigate this value against the actual stacking contract. */
.modal { z-index: 9999; }

/* RIGHT: design token */
.modal { z-index: var(--z-modal); }

/* Investigate the override's scope and reason. */
.button { color: red !important; }

/* RIGHT: explained override */
/* Overrides third-party widget styles that cannot be targeted more specifically */
.widget-container .button { color: red !important; }

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Add automated documentation updater to any Claude skill. Creates a Python sync script that downloads upstream docs, processes markdown for AI consumption, and maintains local cache with configurable refresh. Collects template variables, then delegates implementation through 5-phase workflow. Use when adding auto-updating reference documentation to plugins or skills.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

SAM-style feature initiation workflow — discovery through codebase analysis, architecture spec, task decomposition, validation, and context manifest. Use when a user asks to add a feature, plan a feature, or convert an idea into an executable SAM plan.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

Browser automation for AI agents using the agent-browser CLI and Playwright. Use when navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, logging into sites, or automating any browser task. Triggers on "open a website", "fill out a form", "click a button", "scrape data", "test this web app", "automate browser actions", or any programmatic web interaction request.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

Runs the description-drift experiment — spawns all Claude Code agents simultaneously to collect self-reported capabilities, then compares them against static frontmatter descriptions to reveal how reliable orchestrator routing based on descriptions actually is. Use when measuring description drift across the agent fleet, re-running the capability collection experiment, analyzing a specific agent's self-reported capabilities, or auditing whether frontmatter descriptions accurately reflect agent behavior.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

Create or adapt Claude Code agent definitions. Use when creating an agent, changing subagent configuration, selecting agent scope, or designing a specialized delegation role.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

Jamie-BitFlight のスキルをすべて見る

このスキルの問題を報告する