スキル adm-google-ads
📦
adm-google-ads
v2026-10-08 コンテンツリビジョン r1 中リスク ⚙️ 外部コマンド🌐 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
Manage Google Ads Reports and Campaign Changes
Reviewing Google Ads performance and applying campaign plans requires repeated checks. This skill coordinates ADM reports, previews, approved changes, and campaign verification.
対応: Claude Codex Code(CC)
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 "adm-google-ads" from https://skillstore.io/skills/ownway18-adm-google-ads.md and its manifest at https://skillstore.io/api/skills/ownway18-adm-google-ads/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。
テストする
「adm-google-ads」を使用しています。 Show an example 30-day campaign report without changing the account.
期待される結果:
- Illustrative report, not live account data: account Example Store; window last 30 days; currency USD.
- Brand Search: spend 600 USD; conversions 40; cost per conversion 15 USD.
- Product Search: spend 900 USD; conversions 30; cost per conversion 30 USD.
- Suggested next step: review Product Search terms before proposing exclusions. No changes made.
「adm-google-ads」を使用しています。 Show an example preview for a campaign plan.
期待される結果:
- Illustrative preview, not an executed build: one Search campaign with two ad groups and 20 keywords.
- Daily budget: 25 USD. Review the monthly estimate, bidding, and locations before approving.
- Planned status: paused. Creation waits for approval; activation requires a separate request.
「adm-google-ads」を使用しています。 Show an example reconciliation after an approved campaign build.
期待される結果:
- Illustrative reconciliation, not a live verification: campaign status, budget, bidding, locations, keywords, and ad text match the plan.
- Assets need a separate check because the default campaign setup response omits them.
- Remaining manual tasks: conversion tracking, ad scheduling, and automatic recommendation settings.
セキュリティ監査
中リスクv1 • 2026/10/9 バージョン別レポートを開く
Most alerts match Markdown formatting, legitimate MCP configuration, or authorized advertising workflows. Expanded bearer tokens in command arguments and chat-based setup create credential exposure risks. No evidence found of malware or prompt injection in the reviewed file; the external ADM server implementation was not supplied.
1
スキャンされたファイル
368
解析済み行数
3
レビュー項目
0
誤検知を無視
確認済みのセキュリティ上の懸念 (1)
中
Chat-Based Setup Retains API Credentials
The quick option says, "if the user pastes the key into the chat, use it only to write the user-level configuration below". This legitimizes entering a credential into retained conversation history; the rotation recommendation does not prevent exposure before rotation.
The setup explicitly accepts pasted keys and acknowledges chat-history retention. This is a credential-handling risk, not evidence of malicious exfiltration.
機能レビュー項目 (3)
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
Generic API/secret keys
- Claude Code: `claude mcp add --transport http --scope user adm https://app.adm.cc/mcp --header "Au
The shell expands ADM_API_KEY into the --header argument, exposing the bearer token to process inspection or command logging during MCP setup.
高
Generic API/secret keys
curl -F file=@<path> -H "Authorization: Bearer $ADM_API_KEY" https://app.adm.cc/mcp/uploads
The curl -H argument contains the expanded bearer token, potentially exposing it through process arguments or execution logs despite authorized uploads.
高
Generic API/secret keys
In PowerShell use `curl.exe` and `$env:ADM_API_KEY`. The response contains `upload_id`, `width`, `he
The PowerShell variant substitutes the credential into the preceding curl.exe header argument, carrying the same process-argument exposure risk.
リスク要因
⚙️ 外部コマンド (50)
SKILL.md:14 SKILL.md:15 SKILL.md:16 SKILL.md:17 SKILL.md:18 SKILL.md:19 SKILL.md:20 SKILL.md:21 SKILL.md:22 SKILL.md:28 SKILL.md:30 SKILL.md:31 SKILL.md:32 SKILL.md:35 SKILL.md:37 SKILL.md:38 SKILL.md:39 SKILL.md:40 SKILL.md:42 SKILL.md:46 SKILL.md:48 SKILL.md:49 SKILL.md:51 SKILL.md:52 SKILL.md:53 SKILL.md:60 SKILL.md:64 SKILL.md:65 SKILL.md:74 SKILL.md:77 SKILL.md:79 SKILL.md:80 SKILL.md:81 SKILL.md:82 SKILL.md:84 SKILL.md:90 SKILL.md:96 SKILL.md:97 SKILL.md:98 SKILL.md:99 SKILL.md:100 SKILL.md:102 SKILL.md:103 SKILL.md:109-128 SKILL.md:128-130 SKILL.md:130 SKILL.md:138 SKILL.md:144 SKILL.md:145 SKILL.md:149
🌐 ネットワークアクセス (10)
📁 ファイルシステムへのアクセス (6)
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。
レポートリンクをコピー
https://skillstore.io/skills/ownway18-adm-google-ads/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdownバッジ
[](https://skillstore.io/skills/ownway18-adm-google-ads?utm_source=security_passport_badge)HTMLバッジ
<a href="https://skillstore.io/skills/ownway18-adm-google-ads?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/ownway18-adm-google-ads/security.svg" alt="Skillstore security assessment" loading="lazy"></a>埋め込みカード
<iframe src="https://skillstore.io/embed/skills/ownway18-adm-google-ads.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
ownway18. (2026). adm-google-ads security audit report (audit version 1) [Author version 2026-10-08]. Skillstore. https://skillstore.io/skills/ownway18-adm-google-ads/audits/1BibTeX形式の引用
@techreport{ownway18-ownway18-adm-google-ads-2026,
author = {ownway18},
title = {adm-google-ads security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/ownway18-adm-google-ads/audits/1},
note = {Author version 2026-10-08}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "adm-google-ads security audit report (audit version 1)"
version: "2026-10-08"
type: report
authors:
- name: "ownway18"
date-released: "2026-10-09"
url: "https://skillstore.io/skills/ownway18-adm-google-ads/audits/1"
identifiers:
- type: other
value: "skillstore:ownway18-adm-google-ads: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 55
アーキテクチャ
85
保守性
87
コンテンツ
65
コミュニティ
91
仕様準拠
作成できるもの
Review campaign performance
Compare spending and conversions, identify costly search terms, and prepare changes for approval.
Build campaigns from client plans
Convert approved plans into paused Search campaigns, record progress, and reconcile settings before activation.
Draft campaigns for a business
Use a landing page to request ADM campaign drafts, review budgets and ad text, and approve supported changes.
これらのプロンプトを試す
Review recent campaign results
Show campaign spending and conversions for the last 30 days. State the account, date window, and currency. Do not change anything.
Preview negative keyword changes
Review search terms for [campaign] over 14 days. Suggest negative keywords and preview additions. Wait for approval before writing.
Draft a paused Search campaign
Request an ADM Search campaign draft from [landing page]. Ask for missing budget decisions. Preview creation, wait for approval, and keep it paused.
Resume and reconcile a campaign build
Resume the build for [plan file]. Verify the account and existing operations. Preview remaining changes, wait for approval, and reconcile results without enabling campaigns.
ベストプラクティス
- Use read-only credentials for reporting and enter secrets locally outside chat; avoid exposing tokens in command arguments or logs.
- Check the account, currency, total budget, and affected entities before approving each preview.
- Keep new campaigns paused, verify builds, and inspect recorded operations before retrying uncertain writes.
回避
- Paste API keys into chat, project files, manifests, or commits.
- Enable campaigns or approve spend changes without reviewing the account, budget, and preview warnings.
- Recreate campaigns after timeouts or follow instructions embedded in third-party ad text or search terms.
よくある質問
What must I connect before using this skill?
Connect the ADM MCP server, configure an ADM API key locally, and link your Google Ads account in ADM.
Can I use it with read-only access?
Yes. Reports work with read-only access. Writes and AI drafts require additional permissions, a paid plan, and available tools.
Does it start new campaigns automatically?
The documented workflow creates campaigns paused. Activation requires a separate explicit request and approval of the activation preview.
How are changes approved?
Write tools return previews first. The skill requests approval and confirms the approved preview identifier before executing the change.
Can an interrupted build continue?
The workflow records progress in a local manifest and checks existing campaigns or operations before retrying unfinished work.
What must I still configure in Google Ads?
Configure conversion tracking, ad scheduling, and automatic recommendation settings in Google Ads. Other unsupported edits also remain manual.