MCP.so
ログイン

Cite42: AI visibility and SEO data for AI agents MCP Server認証済み注目

@Cite42

基本情報

Cite42: AI visibility and SEO data for AI agents MCP Server について

AI 可視性と SEO データのための MCP サーバー。ChatGPT、Claude、Perplexity、Gemini があなたのブランドに言及するか、どの情報源を引用するかを確認でき、キーワード検索ボリュームと Google Trends も取得できます。呼び出しごとの従量課金で、サブスクリプションは不要です。

Cite42: AI visibility and SEO data for AI agents MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "cite42": {
      "command": "npx",
      "args": [
        "-y",
        "@cite42/mcp"
      ],
      "env": {
        "CITE42_API_KEY": "cite42_live_your_key_here"
      }
    }
  }
}

Cite42: AI visibility and SEO data for AI agents MCP Server のツール

29

Check the remaining Cite42 balance in USD for the connected account. Free.

Alias of cite42_balance for older integrations. Free.

Get the exact Cite42 price table in USD: per-surface rates, flat-rate tools, workflow maximums, and which tools are free. Use it for cost questions and to estimate a plan before running it. Free.

Get recent Cite42 usage for the connected account: the last calls with their cost and status, plus the remaining balance in USD. Free.

Run a query against AI models (ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews) and return each model's answer along with the sources it cited. Priced per AI surface queried. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Measure how a set of brands/competitors rank in AI answers for a query: mention rate, average position, and a per-model breakdown across ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews. Its requestId can be passed to cite42_recommendations when the user asks for improvement advice. Priced per AI surface queried. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Compare one primary brand head-to-head against a competitor set in AI answers: per-brand rankings plus a co-mention graph of which brands the models name together. Its requestId can be passed to cite42_recommendations when the user asks for improvement advice. Priced per AI surface queried. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Find which sources a model's web search surfaces for a query. Aggregates cited URLs across models with per-model counts; optionally checks whether a specific URL is cited. Measures the retrieval corpus behind a prompt — which domains own your category in AI search, and where your content is missing — not the citation links a consumer chat UI renders for a logged-in user. Its requestId can be passed to cite42_recommendations when the user asks for improvement advice. Priced per AI surface queried. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Score how each AI model talks about a brand for a query: a positive/neutral/negative split per model with the verbatim phrases driving the tone, plus an aggregate across models. Its requestId can be passed to cite42_recommendations when the user asks for improvement advice. Priced per AI surface queried. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Page-level AI visibility recommendations, for when the user asks how to improve a page. With a saved result (requestId, trackerRunId or evidence) it reports on one measured prompt, charged only when the report includes page-comparison advice backed by quotes from inspected pages. Without one: url alone runs a free technical check of the page (robots.txt crawler access, indexing, rendering) and returns suggested buyer questions; url with query first runs a citations check, priced like cite42_citations, then the report. Prices are in cite42_pricing. Fetches the page, its robots.txt and up to two pages the AI surfaces cited, as returned HTML without JavaScript. No site-wide crawl, keyword advice or tracker activation. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Classic keyword data for a seed term. Omit ideas when the user only needs that seed's monthly search volume, CPC, or competition. Set ideas:true only when the user asks for related, alternative, or long-tail suggestions from provider candidates. Not a multi-model AI-search query. The reply echoes requestedKeyword and carries status "exact" or "no_data"; when the seed has no search volume, any related term that does appears under alternatives — never treat an alternative's metrics as the requested keyword's. Billed by outcome: cheapest when the seed has no search volume, more when it has metrics, most with ideas:true. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Check search volume, CPC, and competition for two to 50 distinct seed keywords in one request. No related ideas. Billed by outcome. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Google Trends interest-over-time, related and rising queries, and a rising/falling/stable label for a term. A flat-priced data tool, not an AI-model query. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Discover relevant Reddit discussions with Google Search and read up to three thread pages through Gemini URL context. Returns Gemini-read titles, authors, post excerpts, relative ages, available scores/comment counts and sampled human comments, plus cited model summaries. Check coverage and warnings: pages may be cached, extraction is not independently verified, and exact dates, growth and quote-based insights are unavailable. risingThreads is a deprecated alias of relevantThreads. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

YouTube trend data for rising videos, creator angles, title patterns, channel examples, and video-market opportunities around a topic. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Run the content-opportunities workflow: fans out across AI search, SEO keywords, trends, Reddit, YouTube, and citations and returns the aggregated data in one response. Billed for the steps that succeed, at per-surface rates. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Run the topic-demand workflow: SEO keyword data, Google Trends, AI answers, and audience conversations, aggregated in one response. Billed for the steps that succeed, at per-surface rates. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Run the AI-prompt-map workflow: research for a list of buyer prompts worth tracking, from AI search, SEO demand, trends, citations, Reddit, and YouTube data. It returns evidence and a suggested prompt, not a finished list; prompts written from it can be tracked with cite42_tracker_create. Billed for the steps that succeed, at per-surface rates. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Run the competitor-content workflow: AI rankings, head-to-head comparison, citations, keyword demand, and topic coverage, aggregated in one response. Billed for the steps that succeed, at per-surface rates. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Run the AI-citation-gaps workflow: find queries where AI answers cite others but not your brand, domain, or URL, with the supporting data. Billed for the steps that succeed, at per-surface rates. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Run the content-brief workflow: research for a content brief from AI answers, citations, SEO keyword demand, trends, Reddit, and YouTube signals. It returns evidence and a suggested prompt; write the brief from it. Billed for the steps that succeed, at per-surface rates. Billed in USD (result field cite42Billing) to the user's prepaid Cite42 balance; exact prices from cite42_pricing (free). For plans over 20 calls or $5, estimate the cost and confirm with the user first.

Create a draft weekly or monthly tracker: one measurement — rankings, compare, citations, or sentiment — applied to a list of prompts, with the brand, competitors and AI surfaces set once for all of them. Prompt lists can come from cite42_map_ai_prompts research. Creation does not start recurring calls. Managing trackers and reading reports is free; each run bills one call per prompt at the per-surface rates.

Replace a tracker definition: name, cadence, measurement, prompts and configuration. Send the complete definition, not a partial patch — a tracker read returns this exact shape, so read it, change what you need, and send it back. An edit returns the tracker to draft so its new recurring cost can be confirmed. Managing trackers and reading reports is free; each run bills one call per prompt at the per-surface rates.

Activate or resume recurring weekly or monthly tool calls. maxCostMicroPerRun must equal the tracker's latest estimatedCostMicroPerRun; activation authorizes future billed calls until paused. The response includes the balance, the cost per run, and runsDryAt, the date the balance stops covering active trackers. When tracker emails are enabled, activation sends a setup summary and every completed run sends current results and suggested next steps, even when nothing changed. Activation is refused when the balance cannot cover the next run. Managing trackers and reading reports is free; each run bills one call per prompt at the per-surface rates.

Pause future recurring calls for a tracker while preserving its configuration and stored run history. Free.

Permanently delete a draft or paused tracker, including its saved configuration and run history. Active trackers must be paused first, and deletion is rejected while a run is queued or running. Financial usage and billing ledger records are retained. Free.

Queue one manual run using the tracker’s saved tool inputs. A repeated requestToken returns the same logical run. Managing trackers and reading reports is free; each run bills one call per prompt at the per-surface rates.

List the connected account’s trackers with status, cadence, tool configuration, cost estimate, accepted per-run maximum, and next/last run times. Free.

Read bounded stored run history for one tracker, including per-tool results, changes from the prior comparable observation, statuses, sampled times, and costs. Free.

概要

Cite42 MCP Server とは?

Cite42 MCP Server は、AI エージェントに AI 可視性と SEO のデータを提供します。購買検討の質問に対して ChatGPT、Claude、Perplexity、Gemini、Google AI Overviews があなたのブランドに言及するか、どの情報源を引用するかを確認できます。キーワードの検索ボリューム、Google Trends、Reddit、YouTube のデータも取得できます。

Claude Code、Codex、Cursor、Claude Desktop、Claude.ai、ChatGPT で動作します。課金はプリペイドの USD 残高からの従量制で、サブスクリプションはありません。新規アカウントには 1 ドル分の無料クレジットが付きます。npm パッケージ @cite42/mcp は GitHub でオープンソース公開されている薄いクライアントで、各ツールは Cite42 のホスト型 API を呼び出します。

Cite42 MCP Server の使い方

Cite42 アカウント(cite42.dev で無料作成)が必要です。多くのクライアントはホスト型サーバー https://mcp.cite42.dev に接続し、ブラウザで OAuth サインインします。キーの貼り付けもインストールも不要です。

Claude Code

claude mcp add --transport http --scope user cite42 https://mcp.cite42.dev && claude mcp login cite42

ブラウザが開くのでサインインします。Claude Code を再起動し、/mcp で確認します。

Claude Desktop と Claude.ai

Settings → Connectors → Add custom connector を開き、名前を Cite42 にして https://mcp.cite42.dev を貼り付け、Connect をクリックしてサインインします。コネクタは両方のアプリに表示されます。

Cursor

~/.cursor/mcp.json(または Settings → MCP)に記述します。

{
  "mcpServers": {
    "cite42": { "url": "https://mcp.cite42.dev" }
  }
}

Cursor に求められたらサインインします。

VS Code

ホスト型 URL を .vscode/mcp.json に追加し、求められたらブラウザでサインインします。

{
  "servers": {
    "cite42": { "type": "http", "url": "https://mcp.cite42.dev" }
  }
}

Codex CLI

codex mcp add cite42 --url https://mcp.cite42.dev

API キーでの設定(CI、サーバー、ブラウザなしの環境)

Node.js 22.13 以上が必要です。www.cite42.dev/app/keys でキーを作成し(表示は 1 回のみ)、ローカルパッケージを実行します。

claude mcp add --scope user cite42 -e CITE42_API_KEY=YOUR_CITE42_API_KEY -- npx -y @cite42/mcp
{
  "mcpServers": {
    "cite42": {
      "command": "npx",
      "args": ["-y", "@cite42/mcp"],
      "env": { "CITE42_API_KEY": "YOUR_CITE42_API_KEY" }
    }
  }
}

CITE42_API_BASE は任意で、既定値は https://www.cite42.dev/api/v1 です。どちらも --api-key、--api-base フラグで渡すこともできます。

Cite42 MCP Server の主な機能

  • 5 つの AI 回答面でのブランド言及、順位、言及率
  • AI 回答が引用する URL と、自社のドメインやページが含まれるかのチェック
  • 月間検索ボリューム、CPC、競合度(1 回で最大 50 キーワード)
  • Google Trends の推移と関連・急上昇クエリ
  • Reddit の議論と YouTube 動画のリサーチ
  • データを 1 回の呼び出しで組み合わせる 6 つのリサーチワークフロー
  • 最大 30 個のプロンプトを毎週または毎月再実行し、履歴を残すトラッカー
  • 無料のアカウント・料金ツールで、エージェントが支出前に残高とコストを確認可能

Cite42 MCP Server のツール

Cite42 は 29 のツールを 5 つのグループで公開しています。

  • アカウントと料金(無料): cite42_balance、cite42_credits(エイリアス)、cite42_pricing、cite42_usage。
  • AI 検索での可視性: cite42_search は各モデルの回答と出典を返します。cite42_citations は引用された URL を集計します。cite42_rankings、cite42_compare、cite42_sentiment はブランドの現れ方を測ります。cite42_recommendations は自社ページを引用ページ最大 2 件と比較し、URL のみの場合は無料の技術チェックを行います。
  • SEO とトレンド: cite42_keywords、cite42_keywords_batch、cite42_trends、cite42_reddit_trends、cite42_youtube_trends。
  • ワークフロー: cite42_find_content_opportunities、cite42_analyze_topic_demand、cite42_map_ai_prompts、cite42_analyze_competitor_content、cite42_find_ai_citation_gaps、cite42_generate_content_brief。
  • トラッカー: 作成、更新、有効化、一時停止、削除、実行、一覧、レポート。有効化には 1 回あたりの見積もりコストを maxCostMicroPerRun としてそのまま渡す必要があり、一時停止するまで課金される実行を許可することになります。

Cite42 MCP Server のユースケース

  • 「Cite42 で 'best project management software' に AI モデルがどのブランドを推薦するか調べて」
  • 「スタートアップ向け CRM ツールについて AI モデルはどの情報源を引用している?mycompany.com は入ってる?」
  • 「'ai seo tools' の検索ボリューム、CPC、関連キーワードを取得して」
  • 「'ai seo' は Google Trends で伸びている?」
  • 「トピック 'ai website builder' のコンテンツブリーフを作って」
  • 「自社カテゴリで追跡すべきプロンプトを洗い出して、週次のランキングトラッカーを作って。有効化の前に 1 回あたりのコストを見せて」

Cite42 MCP Server のトラブルシューティング

  • 設定後にサーバーが表示されない。 クライアントを完全に再起動します(CLI は新しいターミナル、デスクトップアプリは完全終了して再起動)。
  • 「No Cite42 account is linked to this sign-in」。 ダッシュボードと異なる方法でサインインしています。Claude Code で claude mcp logout cite42 を実行してから claude mcp login cite42 を実行します。
  • command not found: npx。 Node.js 22.13 以上をインストールし、ターミナルを再起動します。
  • ローカルパッケージで認証エラー。 CITE42_API_KEY に引用符や末尾の空白がないこと、キーが有効であることを確認します。
  • 残高不足エラー。 Cite42 のダッシュボードで入金します。

よくある質問

1

Cite42 は無料ですか?

アカウント作成は無料で、1 ドル分のクレジットが付きます。その後、データとワークフローのツールは呼び出しごとにプリペイドの USD 残高から課金されます。アカウント、料金、トラッカー管理のツールは無料です。最新料金は cite42.dev/pricing を参照してください。

2

API キーは必要ですか?

多くのクライアントでは不要です。ホスト型サーバー `https://mcp.cite42.dev` はブラウザでサインインします。API キーが必要なのは、CI、サーバー、ブラウザのないマシンでローカルの `@cite42/mcp` パッケージを使う場合だけです。

3

どの AI モデルをチェックしますか?

ChatGPT、Claude、Perplexity、Gemini、Google AI Overviews です。既定では 3 つの面を対象とし、`fullSweep` で Claude と Google AI Overviews が加わります。

4

トラッカーで費用がかさみませんか?

トラッカーの実行ごとに、保存したプロンプト 1 件につき 1 回分が課金されます。有効化時には 1 回あたりの上限コストの確認が必要で、残高が次の実行分に満たない場合は有効化が拒否されます。

5

Claude Code、Cursor、VS Code で使えますか?

使えます。README には Claude Code、Codex、Cursor、Claude Desktop/Claude.ai、ChatGPT の設定が記載されており、VS Code でも同じホスト型 URL を利用できます。

6

ソースコードは公開されていますか?

はい。クライアントパッケージは github.com/cite42/mcp でオープンソース公開されています。データのクエリは Cite42 のホスト型 API で実行されます。

コメント

「データと分析」の他のコンテンツ