スキル interface-review

📦

interface-review

Name: interface-review
Author: jakubkrehel

コンテンツリビジョン r1 安全 ⚙️ 外部コマンド📁 ファイルシステムへのアクセス

Review Interface Changes with Clear Evidence

Interface changes can introduce visual, accessibility, and content regressions across connected surfaces. This skill resolves the change scope and produces evidence-based findings with clear status labels.

対応: Claude Codex Code(CC)

🥉 77 ブロンズ

自分のエージェントでインストール

このリクエストをエージェントにコピーしてください。正規の Skill ページとマニフェストが含まれています。

エージェントリクエスト

Review the Skillstore skill "interface-review" from https://skillstore.io/skills/jakubkrehel-interface-review.md and its manifest at https://skillstore.io/api/skills/jakubkrehel-interface-review/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

エージェントは引き続き計画を提示し、セキュリティポリシーで必要な確認を求める必要があります。

エージェントが読めるリソース

AI エージェント、クローラー、スクリプトがページ全体ではなく整理されたコンテキストを必要とする場合は、これらのリンクを使ってください。

Markdown詳細 GET /skills/jakubkrehel-interface-review.md 署名済み manifest GET /api/skills/jakubkrehel-interface-review/manifest 署名済み lockfile GET /api/skills/jakubkrehel-interface-review/lockfile

テストする

「interface-review」を使用しています。 Review the current branch and include uncommitted interface changes.

期待される結果:

Scope: seven commits and two uncommitted files against the default branch.
Coverage: twelve files after exclusions, with two affected surfaces reviewed.
Finding: a dialog lost its accessible name in this change. Status: Regression. Severity: High.
Verdict: Block until the accessible name is restored.

「interface-review」を使用しています。 Review pull request 482 for interface issues.

期待される結果:

Scope: the fetched pull request head compared with its merge base.
Coverage: no interface evidence was found in excluded generated files.
No actionable interface findings were introduced or regressed.
Verdict: Approve.

セキュリティ監査

安全

v1 • 2026/9/4 バージョン別レポートを開く

All 44 static findings are false positives. The external-command alerts mistake Markdown formatting and examples for Ruby execution; the other alerts describe bounded review scope or isolated worktree use. No evidence of prompt injection, exfiltration intent, destructive behavior, or other semantic security issues was found.

スキャンされたファイル

289

解析済み行数

レビュー項目

誤検知を無視

リスク要因

最新の完了済み静的・セマンティック監査では、確認済みのセキュリティ検出事項は見つかりませんでした。これは、スキルに副作用がないことを証明するものではありません。

監査者: codex

このレポートを共有・引用

バージョン付き評価レポート、中立的なバッジ、埋め込みカード、引用を共有できます。Skillstore は証拠を報告しますが、この Skill が安全かどうかは判断しません。

バージョン別レポートを開く

レポートリンクをコピー

https://skillstore.io/skills/jakubkrehel-interface-review/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdownバッジ

[![Skillstore security assessment](https://skillstore.io/badges/skills/jakubkrehel-interface-review/security.svg)](https://skillstore.io/skills/jakubkrehel-interface-review?utm_source=security_passport_badge)

HTMLバッジ

<a href="https://skillstore.io/skills/jakubkrehel-interface-review?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/jakubkrehel-interface-review/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

埋め込みカード

<iframe src="https://skillstore.io/embed/skills/jakubkrehel-interface-review.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

学術引用 (APA · BibTeX · CFF)

APA形式の引用

jakubkrehel. (2026). interface-review security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/jakubkrehel-interface-review/audits/1

BibTeX形式の引用

@techreport{jakubkrehel-jakubkrehel-interface-review-2026,
  author = {jakubkrehel},
  title = {interface-review security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/jakubkrehel-interface-review/audits/1},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "interface-review security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "jakubkrehel"
date-released: "2026-09-04"
url: "https://skillstore.io/skills/jakubkrehel-interface-review/audits/1"
identifiers:
  - type: other
    value: "skillstore:jakubkrehel-interface-review:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore スコア

このスコアの理由証拠の信頼度: 中

アーキテクチャ

保守性

コンテンツ

コミュニティ

仕様準拠

作成できるもの

Review a feature branch

Check interface changes across the branch and local edits before opening a pull request.

Assess a pull request

Resolve the pull request ref, inspect affected surfaces, and identify interface regressions.

Verify accessibility changes

Inspect removed semantics, focus behavior, labels, and motion preferences within the selected change.

これらのプロンプトを試す

Review my current changes

Review the current interface changes. Resolve the default scope and report only introduced or regressed interface issues.

Review a pull request

Review pull request 482 for interface regressions. Include the resolved scope, affected surfaces, exclusions, coverage, and status for each finding.

Review an explicit range

Review release..feature as written. Check removed lines, expand affected consumers, and separate introduced, regressed, and pre-existing findings.

Audit a shared design token change

Review this design token change across two consumer hops. Prioritize rendered entry points, explain coverage limits, and verify each claimed regression against the base.

ベストプラクティス

Name the target explicitly when you need a pull request, ref, branch, or range reviewed.
Provide the change intent so missing states and incomplete behavior can be evaluated.
Request rendered verification when findings depend on runtime behavior or visual output.

回避

Do not use this skill as a whole-repository audit without routing directly to better-interface.
Do not treat pre-existing issues as failures caused by the current change.
Do not assume runtime or visual behavior was verified when the report marks it unverified.