スキル skill-market-publisher

📦

skill-market-publisher

Name: skill-market-publisher
Author: zjsxply

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

複数のマーケットプレイスにスキルを公開する

多くのレジストリにAIスキルを公開するには、異なるメタデータ、URL、検証手順が必要です。このスキルはバンドルを準備し、マーケットのサポート状況を確認し、サポート対象の送信アダプターを実行します。

対応: Claude Codex Code(CC)

⚠️ 38 不十分

インストール前に停止して確認を求めてください。

計画を確認し、ファイルを変更する前にユーザーの明示的な同意を得てください。

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 "skill-market-publisher" from https://skillstore.io/skills/zjsxply-skill-market-publisher.md and its manifest at https://skillstore.io/api/skills/zjsxply-skill-market-publisher/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

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/zjsxply-skill-market-publisher.md 署名済み manifest GET /api/skills/zjsxply-skill-market-publisher/manifest 署名済み lockfile GET /api/skills/zjsxply-skill-market-publisher/lockfile

テストする

「skill-market-publisher」を使用しています。ローカルのスキルフォルダーとGitHubリポジトリURL。

期待される結果:

互換性のあるマーケット、不足項目、構成上の警告、推奨される次のアクションを一覧化した準備状況サマリー。

「skill-market-publisher」を使用しています。作成者、バージョン、カテゴリ、販売者メタデータを含むスキルフォルダー。

期待される結果:

マーケット固有のペイロードファイル、手動手順、検証ターゲットを含む送信バンドル。

「skill-market-publisher」を使用しています。サポート対象マーケットへの完了済みライブ送信。

期待される結果:

リスティングが公開済み、レビュー保留中、ブロック中、またはフォローアップが必要かを説明する検証結果。

セキュリティ監査

高リスク

v5 • 2026/7/8 バージョン別レポートを開く

Most static findings are false positives from Markdown backticks, documentation URLs, placeholder examples, and disclosed marketplace metadata. Confirmed risks are the ClawHub subprocess paths and explicit network egress used for recon and live publication. No prompt injection evidence was found, but this community skill should require operator confirmation before external execution or submission.

スキャンされたファイル

2,890

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 (2)

高

scripts/skill_market_publish.py:1008-1021 scripts/skill_market_publish.py:1031-1036 scripts/skill_market_publish.py:1079-1103

External CLI Execution During Bundle Generation

The bundle writer includes a ClawHub command payload builder that calls build_clawhub_command, which probes the local clawhub executable with subprocess.run. This can execute a PATH-resolved binary before an operator uses live publish.

The call chain from write_bundle to build_clawhub_command and detect_clawhub_publish_prefix is direct. The subprocess uses argument lists, but the local binary execution is real.

中

scripts/skill_market_publish.py:954-971 scripts/skill_market_publish.py:1126-1147 scripts/skill_market_publish.py:1163-1186 scripts/skill_market_publish.py:1195-1325

External Publication Of Skill Content And Metadata

Live adapters post repository URLs, skill descriptions, contact fields, seller metadata, and in one adapter the full SKILL.md content to third-party marketplaces. This is the stated purpose, but users need explicit consent before data leaves the machine.

The payload builders and submit functions clearly assemble local metadata and send it through post_json or post_form. The behavior is intentional but materially affects privacy and publication state.

機能レビュー項目 (38)

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。

高

scripts/skill_market_publish.py:1015

Python subprocess.run

result = subprocess.run(

The bundle and ClawHub command builder can execute a PATH-resolved clawhub binary while probing help output. Argument-list invocation prevents shell injection, but running an external binary is still a real local code execution risk.

高

scripts/skill_market_publish.py:1337

Python subprocess.run

result = subprocess.run(

Live ClawHub publishing runs an external command built from the selected CLI binary and skill path. It requires --execute, but a community skill invoking local CLIs is a real execution risk.

低

scripts/skill_market_publish.py:63

Hardcoded URL

"url": "https://raw.githubusercontent.com/openclaw/clawhub/main/docs/cli.md",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:80

Hardcoded URL

"url": "https://agentskillsindex.com/en/submit",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:92

Hardcoded URL

"submit_url": "https://agentskill.sh/api/skills/submit",

This endpoint is used by a supported live publish adapter when --execute is supplied. The behavior is disclosed and expected, but it can transmit skill metadata to an external marketplace.

低

scripts/skill_market_publish.py:97

Hardcoded URL

"url": "https://agentskill.sh/submit",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:109

Hardcoded URL

"submit_url": "https://www.skillz.directory/api/submit",

This endpoint is used by a supported live publish adapter when --execute is supplied. The behavior is disclosed and expected, but it can transmit skill metadata to an external marketplace.

低

scripts/skill_market_publish.py:114

Hardcoded URL

"url": "https://www.skillz.directory/submit",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:126

Hardcoded URL

"submit_url": "https://skillstore.io/api/submit",

This endpoint is used by a supported live publish adapter when --execute is supplied. The behavior is disclosed and expected, but it can transmit skill metadata to an external marketplace.

低

scripts/skill_market_publish.py:131

Hardcoded URL

"url": "https://skillstore.io/submit",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:143

Hardcoded URL

"submit_url": "https://api.a2amarket.live/v1/listings",

This endpoint is used by a supported live publish adapter when --execute is supplied. The behavior is disclosed and expected, but it can transmit skill metadata to an external marketplace.

低

scripts/skill_market_publish.py:148

Hardcoded URL

"url": "https://a2amarket.live/skill.md",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:165

Hardcoded URL

"url": "https://github.com/zjunlp/SkillNet",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:180

Hardcoded URL

"submit_url": "https://agent-skills.md/rpc/repos/submit",

This endpoint is used by a supported live publish adapter when --execute is supplied. The behavior is disclosed and expected, but it can transmit skill metadata to an external marketplace.

低

scripts/skill_market_publish.py:185

Hardcoded URL

"url": "https://agent-skills.md/submit",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:197

Hardcoded URL

"submit_url": "https://skillsmd.dev/api/submit",

This endpoint is used by a supported live publish adapter when --execute is supplied. The behavior is disclosed and expected, but it can transmit skill metadata to an external marketplace.

低

scripts/skill_market_publish.py:202

Hardcoded URL

"url": "https://skillsmd.dev/",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:219

Hardcoded URL

"url": "https://agentskillsrepo.com/",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:231

Hardcoded URL

"submit_url": "https://skills.re/api/rpc/skills/submitGithubRepoPublic",

This endpoint is used by a supported live publish adapter when --execute is supplied. The behavior is disclosed and expected, but it can transmit skill metadata to an external marketplace.

低

scripts/skill_market_publish.py:236

Hardcoded URL

"url": "https://skills.re/submit",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:253

Hardcoded URL

"url": "https://www.skillsdirectory.com/submit",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:257

Hardcoded URL

"url": "https://www.skillsdirectory.com/docs/registry-api",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:274

Hardcoded URL

"url": "https://www.skillsdir.dev/add",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:291

Hardcoded URL

"url": "https://skills.sh/",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:308

Hardcoded URL

"url": "https://skillsfor.ai/",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:323

Hardcoded URL

"recon": [{"url": "https://skillscatalog.ai/submit", "markers": ["submit", "GitHub", "skill"]}],

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:337

Hardcoded URL

"url": "https://www.mdskills.ai/submit",

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:352

Hardcoded URL

"recon": [{"url": "https://skillery.dev/submit", "markers": ["submit", "login", "skill"]}],

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:364

Hardcoded URL

"recon": [{"url": "https://www.skillszoo.com/submit", "markers": ["Skill Name", "GitHub Repository U

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:376

Hardcoded URL

"recon": [{"url": "https://www.skillhub.club/docs/cli", "markers": ["npx @skill-hub/cli login", "npx

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:388

Hardcoded URL

"recon": [{"url": "https://skillshop.sh/agent.md", "markers": ["there is no upload API on SkillShop

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:400

Hardcoded URL

"recon": [{"url": "https://skillhq.dev/become-seller", "markers": ["skillhq validate ./my-skill", "s

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:409

Hardcoded URL

"submit_url": "https://www.bogen.ai/api/skills/submit",

This endpoint is used by a supported live publish adapter when --execute is supplied. The behavior is disclosed and expected, but it can transmit skill metadata to an external marketplace.

低

scripts/skill_market_publish.py:412

Hardcoded URL

"recon": [{"url": "https://www.bogen.ai/skills/submit", "markers": ["SKILL.md Content", "Submit Skil

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:421

Hardcoded URL

"submit_url": "https://skillsrep.com/submit",

This endpoint is used by a supported live publish adapter when --execute is supplied. The behavior is disclosed and expected, but it can transmit skill metadata to an external marketplace.

低

scripts/skill_market_publish.py:424

Hardcoded URL

"recon": [{"url": "https://skillsrep.com/submit", "markers": ["Code / Skill Definition", "CLAUDE.md

This URL is fetched by the recon flow to check marketplace page markers. It is an explicit feature, but still creates third-party network egress.

低

scripts/skill_market_publish.py:1254

Hardcoded URL

fetch_result = post_json("https://skills.re/api/rpc/github/fetchRepo", fetch_payload)

The publish path posts a repository preview payload to the skills.re fetchRepo endpoint before submitting. That external request is expected but security-relevant network egress.

低

scripts/skill_market_publish.py:1367

Hardcoded URL

base_url = "https://www.skillz.directory"

This base URL is used during Skillz Directory recon to fetch static chunks and inspect expected markers. It is deliberate network activity against a third-party site.

リスク要因

📁 ファイルシステムへのアクセス (3)

references/market-matrix.md:431 scripts/skill_market_publish.py:1514 SKILL.md:43

🌐 ネットワークアクセス (110)

⚙️ 外部コマンド (75)

監査者: codex 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/zjsxply-skill-market-publisher/audits/5?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdownバッジ

[![Skillstore security assessment](https://skillstore.io/badges/skills/zjsxply-skill-market-publisher/security.svg)](https://skillstore.io/skills/zjsxply-skill-market-publisher?utm_source=security_passport_badge)

HTMLバッジ

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

埋め込みカード

<iframe src="https://skillstore.io/embed/skills/zjsxply-skill-market-publisher.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

zjsxply. (2026). skill-market-publisher security audit report (audit version 5) [Author version unspecified]. Skillstore. https://skillstore.io/skills/zjsxply-skill-market-publisher/audits/5

BibTeX形式の引用

@techreport{zjsxply-zjsxply-skill-market-publisher-2026,
  author = {zjsxply},
  title = {skill-market-publisher security audit report (audit version 5)},
  institution = {Skillstore},
  year = {2026},
  number = {5},
  url = {https://skillstore.io/skills/zjsxply-skill-market-publisher/audits/5},
  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: "skill-market-publisher security audit report (audit version 5)"
version: "unspecified"
type: report
authors:
  - name: "zjsxply"
date-released: "2026-07-08"
url: "https://skillstore.io/skills/zjsxply-skill-market-publisher/audits/5"
identifiers:
  - type: other
    value: "skillstore:zjsxply-skill-market-publisher:audit:5"
    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

新しいスキルリリースを準備する

1つのスキルフォルダーを検査し、不足しているマーケット項目を特定して、レビュー可能なリリース計画を作成します。

マーケットプレイス送信を調整する

オペレーターが送信する前に、複数の公開レジストリ向けのペイロードと手動メモを生成します。

公開済みリスティングを検証する

送信後にマーケット固有の検証画面を確認し、公開、保留、またはブロック状態を確認します。

これらのプロンプトを試す

1つのスキルを検査する

このスキルフォルダーを検査し、リポジトリ構成、メタデータ、明らかに不足しているマーケットプレイス入力項目を教えてください。

マーケットプランを作成する

このリポジトリURLとgit refを使って、このスキルのマーケット準備状況プランを作成してください。不足している項目と手動ターゲットを強調してください。

送信バンドルを作成する

このスキルについて、ペイロード、手動メモ、検証ターゲット、必須のマーケット専用項目を含む送信バンドルを生成してください。

安全に公開して検証する

サポート対象の公開ターゲットをドライランし、送信ペイロードをレビューし、承認後にのみ実行してから、公開リスティングの状態を検証してください。

ベストプラクティス

バンドル作成やライブ送信の前に、inspectとplanを実行してください。
ライブマーケットで --execute を使用する前に、生成されたすべてのペイロードをレビューしてください。
マーケットアダプターや公開サポートの主張に依存する前に、reconを再実行してください。

回避

マーケットの検証パスを調べるために偽データを送信しないでください。
manual、index-only、incompatible、needs-recon とマークされたマーケットに対して自動化を主張しないでください。
単発の送信IDや日付付きの実行ログを、再利用可能なスキル内に保存しないでください。

よくある質問

このスキルは自動的に公開しますか？

サポート対象のターゲットのみ自動公開できます。また、ライブ送信には --execute フラグが必要です。

このスキルはどのツールで使用できますか？

レポートにはClaude、Codex、Claude Codeとの互換性が一覧表示されます。

手動のマーケットプレイスに対応していますか？

はい。安定した自動化パスが利用できない場合、追跡対象の送信ページを開き、チェックリストを表示します。

リスティングが公開されていることを検証できますか？

マーケット固有の検証ガイダンスとreconチェックを提供しますが、最終的なリスティング状態は各マーケットに依存します。

どのデータが自分のマシンの外に出る可能性がありますか？

ライブ公開とreconはマーケットプレイスサイトに接続する場合があります。一部の送信には、スキルの内容、リポジトリURL、メール、販売者メタデータが含まれます。

非公開スキルに使用すべきですか？

いいえ。このワークフローは公開リポジトリと公開マーケットプレイスリスティング向けに設計されています。

開発者情報

作成者

zjsxply

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/zjsxply/skills/tree/main/skill-market-publisher/

参照

88a205c7f635a966e31156313b590d59007c5caa

メンテナンスの新しさ

2026/7/20

利用状況

45 ダウンロード · 118 閲覧

ファイル構成

📁 references/

📄 market-matrix.md

📄 publish-playbook.md

📄 recon-playbook.md

📄 verification-playbook.md

📁 scripts/

📄 open_manual_submit_pages.py

📄 skill_market_publish.py

📄 SKILL.md