スキル skills-factory

📦

skills-factory

Name: skills-factory
Author: DaronVee

コンテンツリビジョン r1 重大 🌐 ネットワークアクセス📁 ファイルシステムへのアクセス🔑 環境変数⚙️ 外部コマンド

建置可用於生產環境的 Claude Code 技能

建立可靠的 AI 技能需要清楚的結構、驗證與部署指引。此技能提供範本、工作流程和指令碼，用於以可重複的品質檢查建置 Claude Code 技能。

対応: Claude Codex Code(CC)

⚠️ 38 不十分

Maintain this Skill? Claim it ⚠️ 38 不十分

この Skill を自動インストールしないでください。

基準ポリシーにより、インストール操作の前にオペレーターのレビューが必要です。

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 "skills-factory" from https://skillstore.io/skills/daronvee-skills-factory.md and its manifest at https://skillstore.io/api/skills/daronvee-skills-factory/manifest. Verify the artifact. Do not auto-install. Inspect the skill and report your findings, then wait for an operator or manual installation decision.

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/daronvee-skills-factory.md 署名済み manifest GET /api/skills/daronvee-skills-factory/manifest 署名済み lockfile GET /api/skills/daronvee-skills-factory/lockfile

テストする

「skills-factory」を使用しています。我需要一個協助工程師撰寫遷移計畫的技能。

期待される結果:

建議的技能大綱，包含聚焦的觸發條件、必要中繼資料、參考指南清單、驗證檢查清單和初始評估情境。

「skills-factory」を使用しています。檢查我的技能是否已準備好分享給我的團隊。

期待される結果:

Frontmatter 檢閱結果
缺少參考資料的警告
封裝準備狀態備註
執行前需檢閱的手動命令

「skills-factory」を使用しています。協助我評估此技能是否會在正確時機觸發。

期待される結果:

測試矩陣，包含黃金路徑、非觸發案例、邊界案例，以及修訂技能指示的準則。

セキュリティ監査

重大

v11 • 2026/7/9 バージョン別レポートを開く

Most static findings are documentation examples for Claude skill installation, validation commands, and official Anthropic API usage. Confirmed risks are limited to user-facing local command guidance, an example recursive delete command, and installer filesystem changes that can overwrite or remove an existing skill directory after confirmation.

スキャンされたファイル

5,359

解析済み行数

レビュー項目

誤検知を無視

確認済みのセキュリティ上の懸念 (1)

重大

references/DEPLOYMENT_GUIDE.md:67

Recursive delete on root/home

rm -rf ~/.claude/skills/my-skill

The documentation includes rm -rf against a home-directory skill path. Although scoped to a sample skill folder, recursive deletion in the user home area is dangerous if copied or modified.

機能レビュー項目 (29)

これらは、このスキルに期待される可能性のある実際のローカル機能であるため、レビューが必要ですが、確認済みの悪意ある動作としてはカウントされません。

重大

references/DEPLOYMENT_GUIDE.md:67

Hidden file in home directory

rm -rf ~/.claude/skills/my-skill

The documentation shows a recursive delete command under the user home skill directory. It is example cleanup, but copying it with the wrong target could cause data loss.

重大

references/DEPLOYMENT_GUIDE.md:67

Hidden file access

rm -rf ~/.claude/skills/my-skill

The documentation shows a recursive delete command under the user home skill directory. It is example cleanup, but copying it with the wrong target could cause data loss.

中

references/TROUBLESHOOTING.md:166-177

Ruby/shell backtick execution

Run: `grep -r "TODO: document" src/`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/TROUBLESHOOTING.md:177-182

Ruby/shell backtick execution

Run: `bash scripts/validate_docs.py`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/TROUBLESHOOTING.md:672-673

Ruby/shell backtick execution

Run: `python scripts/validate.py config.yaml`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/TROUBLESHOOTING.md:677-678

Ruby/shell backtick execution

Run: `python ~/.claude/skills/my-skill/scripts/validate.py config.yaml`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/VALIDATION_PATTERNS.md:472-474

Ruby/shell backtick execution

"Run: `npm run lint && npm test`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:60-66

Ruby/shell backtick execution

Run validation: `bash scripts/validate_format.py [document]`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:100-101

Ruby/shell backtick execution

- [ ] All tests passing (`npm test`)

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:101-102

Ruby/shell backtick execution

- [ ] No linting errors (`npm run lint`)

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:102-110

Ruby/shell backtick execution

- [ ] Code coverage ≥80% (`npm run coverage`)

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:110-111

Ruby/shell backtick execution

- [ ] Dependencies updated (`npm audit`)

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:111-116

Ruby/shell backtick execution

- [ ] No secrets in code (check with `git secrets --scan`)

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:116-119

Ruby/shell backtick execution

- [ ] Git tag created (`git tag v1.2.3`)

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:119-120

Ruby/shell backtick execution

**All items complete?** Run: `bash scripts/deploy.sh`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:178-180

Ruby/shell backtick execution

4. Run validation: `bash scripts/validate_openapi.py`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:192-194

Ruby/shell backtick execution

4. Validate schema: `bash scripts/validate_graphql_schema.py`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:206-208

Ruby/shell backtick execution

4. Validate: `protoc --lint_out=. *.proto`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:249-265

Ruby/shell backtick execution

1. Run test coverage: `npm run coverage`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:265-273

Ruby/shell backtick execution

4. Run new tests: `npm test [test-file]`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:273-274

Ruby/shell backtick execution

1. Run full coverage again: `npm run coverage`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:348-356

Ruby/shell backtick execution

Run: `bash scripts/validate_config.py`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:363-366

Ruby/shell backtick execution

Run: `openapi-generator generate -i spec.yaml -g typescript-fetch`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:370-371

Ruby/shell backtick execution

**If errors occur:** Check OpenAPI spec validity with `bash scripts/validate_spec.py`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

references/WORKFLOW_PATTERNS.md:457-460

Ruby/shell backtick execution

Run: `npm run build`

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

scripts/package_skill.py:76

Python shutil operations

shutil.rmtree(target_path)

The installer deletes an existing target skill directory before copying a replacement. The path is constrained to skill install locations and requires confirmation, but it can still remove user files under that target.

中

scripts/package_skill.py:81

Python shutil operations

shutil.copytree(skill_path, target_path)

The installer copies a user-selected skill directory into Claude skill locations. This is expected behavior, but it modifies filesystem state and should be disclosed.

中

SKILL.md:170-186

Ruby/shell backtick execution

- Load scripts when needed: "Run validation: `bash scripts/validate.py`"

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

中

SKILL.md:262-265

Ruby/shell backtick execution

Team members get skill automatically on `git pull`.

The skill documentation asks users to run local development commands. They are ordinary validation or packaging commands, but marketplace users should review them before execution.

リスク要因

🌐 ネットワークアクセス (8)

references/DEPLOYMENT_GUIDE.md:273 references/DEPLOYMENT_GUIDE.md:318 references/DEPLOYMENT_GUIDE.md:325 references/DEPLOYMENT_GUIDE.md:486 references/DEPLOYMENT_GUIDE.md:587 references/TROUBLESHOOTING.md:813 references/TROUBLESHOOTING.md:868 references/TROUBLESHOOTING.md:872

📁 ファイルシステムへのアクセス (50)

🔑 環境変数 (11)

references/DEPLOYMENT_GUIDE.md:274 references/DEPLOYMENT_GUIDE.md:293 references/DEPLOYMENT_GUIDE.md:319 references/DEPLOYMENT_GUIDE.md:326 references/DEPLOYMENT_GUIDE.md:487 references/DEPLOYMENT_GUIDE.md:588 references/TROUBLESHOOTING.md:814 references/TROUBLESHOOTING.md:869 references/TROUBLESHOOTING.md:873 references/VALIDATION_PATTERNS.md:81 references/VALIDATION_PATTERNS.md:85

⚙️ 外部コマンド (50)

検出されたパターン

Recursive delete on root/home

監査者: claude 監査履歴を表示 →

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/daronvee-skills-factory/audits/11?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdownバッジ

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

HTMLバッジ

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

埋め込みカード

<iframe src="https://skillstore.io/embed/skills/daronvee-skills-factory.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

DaronVee. (2026). skills-factory security audit report (audit version 11) [Author version unspecified]. Skillstore. https://skillstore.io/skills/daronvee-skills-factory/audits/11

BibTeX形式の引用

@techreport{daronvee-daronvee-skills-factory-2026,
  author = {DaronVee},
  title = {skills-factory security audit report (audit version 11)},
  institution = {Skillstore},
  year = {2026},
  number = {11},
  url = {https://skillstore.io/skills/daronvee-skills-factory/audits/11},
  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: "skills-factory security audit report (audit version 11)"
version: "unspecified"
type: report
authors:
  - name: "DaronVee"
date-released: "2026-07-09"
url: "https://skillstore.io/skills/daronvee-skills-factory/audits/11"
identifiers:
  - type: other
    value: "skillstore:daronvee-skills-factory:audit:11"
    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

建立新的團隊技能

產生結構化技能，包含清楚的指示、參考資料與驗證步驟，之後再分享給團隊。

驗證現有技能

在發布前檢查技能的 frontmatter、參考資料、命名與封裝問題。

封裝技能以便部署

使用可重複的封裝指引，為 Claude Code、Claude Desktop 或 API 型發布準備技能資料夾。

これらのプロンプトを試す

開始建立基本技能

協助我為 [task] 建立一個簡單的 Claude Code 技能。在起草結構之前，請先詢問我目標、目標使用者和工具。

改善技能結構

檢閱我的 SKILL.md，檢查清晰度、漸進式揭露、必要中繼資料和缺少的參考檔案。提出具體修改建議。

驗證並封裝

帶我完成此技能的驗證與封裝。在我執行每個命令之前先說明，並標示任何檔案系統變更。

設計評估工作流程

使用實際成功案例、邊界案例、失敗案例和第二代理審查循環，為此技能建立評估計畫。

ベストプラクティス

在執行每個產生的命令前先檢閱，尤其是會複製、刪除或安裝檔案的命令。
保持 SKILL.md 精簡，並將詳細工作流程放在參考檔案中。
在分享技能之前，使用成功、邊界與失敗情境進行測試。

回避

不要建立會因不相關開發任務而觸發的過於寬泛技能。
不要將機密、私人設定或本機電腦路徑封裝到共享技能中。
不要僅依賴封裝結果作為技能安全或實用的證明。

よくある質問

這個技能可以協助我建立什麼？

它協助建立 Claude Code 技能，包含結構化指示、參考資料、驗證檢查和封裝指引。

它會自動執行命令嗎？

不會。它提供文件與指令碼，但使用者必須自行選擇何時執行本機命令。

它可以將技能安裝到 Claude Code 嗎？

可以。封裝指令碼可在使用者操作後，將已驗證的技能安裝到個人或專案技能目錄。

它只適用於 Claude Code 嗎？

此工作流程著重於 Claude Code，但封裝參考資料也提到 Claude Desktop、Claude.ai 和 API 發布。

使用前我應該檢閱什麼？

在執行指令碼或分享 zip 之前，請檢閱 shell 命令、檔案系統安裝路徑和所有已封裝檔案。

它可以取代安全審查嗎？

不會。它能改善結構與驗證，但在 marketplace 發布前仍需要另外進行安全審查。

開発者情報

作成者

DaronVee

ライセンス

MIT

Skillstore リビジョン

バージョンに関する注意

作者はバージョンを宣言していません。

リポジトリ

https://github.com/DaronVee/ccgg-skills-factory/tree/master/skills-factory

参照

1ffa7643651792ccb4bd3b15d924d2c97edff755

メンテナンスの新しさ

2026/7/18

利用状況

7 ダウンロード · 201 閲覧

ファイル構成

📁 references/

📄 DEPLOYMENT_GUIDE.md

📄 EVALUATION_GUIDE.md

📄 TROUBLESHOOTING.md

📄 TWO_CLAUDE_METHODOLOGY.md

📄 VALIDATION_PATTERNS.md

📄 WORKFLOW_PATTERNS.md

📁 scripts/

📄 comprehensive_validate.py

📄 init_skill.py

📄 package_skill.py

📄 quick_validate.py

📄 SKILL.md