ワンクリックで
review
Delegate UI code reviews to a subagent. Reviews components, pages, git changes, or specification files using web design guidelines.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Delegate UI code reviews to a subagent. Reviews components, pages, git changes, or specification files using web design guidelines.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Fetches and normalizes http(s) web pages into clean Markdown for LLM ingestion. Use when a task includes a URL, needs to fetch docs or asks to convert web docs/articles/pages into Markdown for summarizing, quoting, diffing, or saving.
Creates a commit with repo-matching style and intentional staging.
Fetches and normalizes http(s) web pages into clean Markdown for LLM ingestion. Use when a task includes a URL, needs to fetch docs or asks to convert web docs/articles/pages into Markdown for summarizing, quoting, diffing, or saving.
Structured code and document review workflow for finding actionable issues in diffs, PRs, commits, files, directories, and architecture docs.
This skill should be used when the user asks to "write React tests", "test a component", "set up test utils", "mock an API", "use React Testing Library", "configure MSW", "test async behavior", "choose query type", "test Mantine components", "test a modal", or when writing tests for React components using Vitest, RTL, MSW, or Mantine. Provides best practices for component and integration testing.
Use when building or modifying frontend UI and implementation choices should follow shared web interface guidelines for interactions, accessibility, forms, motion, performance, responsive behavior, or copy.
| name | review |
| description | Delegate UI code reviews to a subagent. Reviews components, pages, git changes, or specification files using web design guidelines. |
| allowed-tools | ["Agent","Glob","AskUserQuestion"] |
With argument: Use it directly.
Without argument: Determine from session context:
AskUserQuestionAsk when unclear. Use AskUserQuestion with options:
Always include:
Add context-specific options if identifiable:
Quick glob (don't read):
glob: "docs/**/design-system.md"
glob: "**/DESIGN_SYSTEM.md"
Do NOT gather content yourself. Just pass identifiers — the agent gets the content.
tool: Agent
parameters:
description: "UI review: [target]"
subagent_type: "web-design-guidelines:ui-reviewer"
prompt: |
Review: [TARGET IDENTIFIER - e.g. "commit abc123", "src/Button.tsx", "uncommitted changes"]
Design system: [PATH or "none"]
Context: [BRIEF SESSION CONTEXT - what feature, what spec, focus areas]
Don't:
git diff or git showDo: