Audits an existing rendered interface or frontend diff for hierarchy, content clarity, interaction states, accessibility, responsive behavior, visual consistency, performance-sensitive UI patterns, and anti-generic design defects. Use for a UI/UX review, redesign audit, screenshot review, pre-release visual check, or design-system consistency pass. Not for building a new interface from scratch or for backend-only review.
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。
Review the interface users actually receive, not only the component source.
Identify routes, primary flows, target users, supported viewports, themes, content states, and the design system or brand constraints.
Inspect the diff and direct consumers before forming findings.
Run the deterministic audit helper against relevant HTML/CSS/JS when a static surface is available:
python3 skills/ui-ux-review/scripts/audit_ui.py ./path/to/ui
Zero supported source files is inconclusive and exits with status 2. Use
--allow-empty only when an empty UI scope is intentional; it is not proof
that a surface is clean.
Extract the current token and component system with frontend-design when
consistency claims depend on it; compare against existing tokens before
proposing a redesign language.
Render the surface in a browser. Capture normal, empty, error, loading, permission, narrow, intermediate, and desktop states that matter to the request. If runtime access is blocked, mark visual findings unverified.
Check whether the first viewport explains the page, the primary action is obvious, copy is scannable, labels are specific, and repeated containers add meaning rather than noise.
Check semantics, names, keyboard order, focus visibility, contrast, target size, error recovery, reduced motion, screen-reader announcements, and whether color is the only signal. Run axe or the repository's equivalent when available, then manually traverse the primary flow by keyboard; automated accessibility scans do not prove task usability.
Check narrow/intermediate/desktop reflow, long content, tables, dialogs, overflow, theme contrast, tokens, type rhythm, spacing, icon consistency, and layout stability.
Check image dimensions and loading, layout shift, blocking work, feedback during async actions, destructive confirmation, and misleading placeholder or proof content.
Use review-checklist.md and report only
actionable findings with priority, exact location/route, failure mode, impact,
and the smallest correction. Separate static findings from unverified visual
claims. Use the runtime state/viewport matrix supplied by frontend-design
rather than treating a single desktop screenshot as evidence.
The report covers the requested surface and states, every finding is evidence-backed or marked unverified, and the next design or engineering action is clear.
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。
概要と使いどころ
Designs and runs reproducible evaluations for AI agents, prompts, tools, skills, and model-backed workflows using realistic datasets, isolated baselines, objective assertions, rubric grading, trajectory analysis, cost/latency tracking, and regression comparison. Use when measuring agent quality, optimizing skill triggering, comparing prompts or models, or gating an AI feature release. Not for ordinary deterministic unit tests.
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。
Designs or reviews HTTP, REST, GraphQL, RPC, CLI, webhook, event, and service interfaces with explicit inputs, outputs, errors, compatibility, idempotency, pagination, authentication, versioning, and observability. Use when introducing or changing an API or cross-component contract. Not for internal implementation details with no boundary or for debugging one API failure; use root-cause-debugging there.
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。
Reviews an existing codebase for structural friction, unclear ownership, leaky or shallow interfaces, excessive coupling, misplaced state, poor testability, and risky dependency direction, then prioritizes evidence-backed improvement candidates. Use for architecture audits, modularization, modernization, or recurring cross-cutting change pain. Not for designing one new interface, simplifying a local function, or fixing a reproduced bug.
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。
Validates a running application, CLI, API, service, or generated artifact as a user or operator against a prewritten observable behavior contract while remaining source-blind. Use for acceptance checks, runtime proof, anti-fake probes, release smoke tests, or an independent companion to code review. Not for source-quality findings, root-cause diagnosis, or visual design judgment outside the contract.
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。
Closes a completed development branch by checking the final diff and proof, presenting merge, pull-request, keep, or discard options, and cleaning up only after the user or repository workflow chooses a path. Use when feature work is complete and the branch must be integrated or retired. Not for claiming a feature is complete before verification.
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。
Tests a real web user flow with a browser by asserting semantic behavior, network and loading states, keyboard access, responsive layouts, and stable visual evidence. Use for browser bugs, end-to-end UI behavior, responsive or accessibility checks, and screenshot baselines. Not for static source review without a browser or for backend-only tests.
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。