AIニュース要約ビルダー:Dumpling AIでサイトをスクラップし、GPT-4oで要約
中級
これはAI, Marketing分野の自動化ワークフローで、10個のノードを含みます。主にSet, Code, Gmail, SplitOut, HttpRequestなどのノードを使用、AI技術を活用したスマート自動化を実現。 AIニュースドットメ거ー:Dumpling AIによるサイト取得でGPT-4oサマリー
前提条件
- •Googleアカウント + Gmail API認証情報
- •ターゲットAPIの認証情報が必要な場合あり
- •Google Sheets API認証情報
- •OpenAI API Key
ワークフロープレビュー
ノード接続関係を可視化、ズームとパンをサポート
ワークフローをエクスポート
以下のJSON設定をn8nにインポートして、このワークフローを使用できます
{
"id": "HlVn82cNQLDWx73K",
"meta": {
"instanceId": "a1ae5c8dc6c65e674f9c3947d083abcc749ef2546dff9f4ff01de4d6a36ebfe6",
"templateCredsSetupCompleted": true
},
"name": "AI Newsletter Builder: Crawl Sites with Dumpling AI, Summarize with GPT-4o",
"tags": [],
"nodes": [
{
"id": "b607052f-8712-4cce-977e-6216da30a144",
"name": "Start Workflow Manually",
"type": "n8n-nodes-base.manualTrigger",
"position": [
-800,
-160
],
"parameters": {},
"typeVersion": 1
},
{
"id": "5bf110b3-a678-4a34-a4c9-3a87ab22774f",
"name": "Google スプレッドシートからウェブサイトURLを取得",
"type": "n8n-nodes-base.googleSheets",
"position": [
-580,
-160
],
"parameters": {
"options": {},
"sheetName": {
"__rl": true,
"mode": "list",
"value": "",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/10AQk_0kG8d0fa7gwbm0W8f9CyMZ9syPGksQoMNBkut4/edit#gid=0",
"cachedResultName": "Sheet1"
},
"documentId": {
"__rl": true,
"mode": "list",
"value": "",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/10AQk_0kG8d0fa7gwbm0W8f9CyMZ9syPGksQoMNBkut4/edit?usp=drivesdk",
"cachedResultName": "sites"
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "GaJqJHuS5mQxap7q",
"name": "Google Sheets account"
}
},
"typeVersion": 4.5
},
{
"id": "a996f64a-bb61-493d-90ca-dce9d465fd58",
"name": "Dumpling AIでサイトコンテンツをクロール・抽出",
"type": "n8n-nodes-base.httpRequest",
"position": [
-360,
-160
],
"parameters": {
"url": "https://app.dumplingai.com/api/v1/crawl",
"method": "POST",
"options": {},
"jsonBody": "={\n \"url\": \"{{ $json.websites }}\", \n \"limit\": \"5\",\n \"depth\": \"2\", \n \"format\": \"text\"\n}",
"sendBody": true,
"specifyBody": "json",
"authentication": "genericCredentialType",
"genericAuthType": "httpHeaderAuth"
},
"credentials": {
"httpHeaderAuth": {
"id": "5CBvzXCLjwWzCJRE",
"name": "n8n_integration"
}
},
"typeVersion": 4.2
},
{
"id": "01d0b913-cfdc-4144-8556-d5c18221cf7c",
"name": "抽出結果を個別アイテムに分割",
"type": "n8n-nodes-base.splitOut",
"position": [
-140,
-160
],
"parameters": {
"options": {},
"fieldToSplitOut": "results"
},
"typeVersion": 1
},
{
"id": "9d69abfa-c867-4fbc-83a2-58723f3c4dfb",
"name": "タイトル、URL、ページテキストをマッピング",
"type": "n8n-nodes-base.set",
"position": [
80,
-160
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "51ae9a51-06f4-46b1-979e-578eb5dc361c",
"name": "metadata.title",
"type": "string",
"value": "={{ $json.metadata.title }}"
},
{
"id": "2f16ef6f-cc2f-4a4d-a5e7-3cdde3465398",
"name": "content",
"type": "string",
"value": "={{ $json.content }}"
},
{
"id": "31ac95b7-beeb-4b3b-8e52-ed75027ae379",
"name": "metadata.original_url",
"type": "string",
"value": "={{ $json.metadata.original_url }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "eaaa909d-92df-4837-a474-d533577c5621",
"name": "複数の記事を単一のプロンプト形式に結合",
"type": "n8n-nodes-base.code",
"position": [
300,
-160
],
"parameters": {
"jsCode": "let output = '';\nitems.forEach((item, index) => {\n const title = item.json.metadata?.title || 'No title';\n const url = item.json.metadata?.original_url || 'No URL';\n const content = item.json.content || 'No content';\n\n output += `${index + 1}. ${title}\\n${url}\\n${content}\\n\\n`;\n});\n\nreturn [{ json: { aggregatedArticles: output } }];\n"
},
"typeVersion": 2
},
{
"id": "b2691018-fa64-4a55-a5e9-e68733ab9a7a",
"name": "GPT-4oで件名付きのHTMLニュースレターを生成",
"type": "@n8n/n8n-nodes-langchain.openAi",
"position": [
520,
-160
],
"parameters": {
"modelId": {
"__rl": true,
"mode": "list",
"value": "gpt-4o",
"cachedResultName": "GPT-4O"
},
"options": {},
"messages": {
"values": [
{
"content": "=You are a newsletter assistant that creates a compelling subject line and summarizes a list of articles in engaging, well-structured HTML format.\n\nInstructions:\n\n1. Read the list of articles. Each article includes a title, content, and original URL.\n2. First, generate a short, catchy subject line for the newsletter based on the overall theme of the articles.\n3. Then, for each article:\n - Write the title inside an <h3> tag.\n - Summarize the article in 2 to 3 engaging sentences inside a <p> tag.\n - Below the summary, write: “To read more, click the link:” and include the original URL as a clickable link inside another <p> tag.\n - Add a <br/> to separate entries.\n\nReturn the result in this JSON format:\n\n{\n \"subject\": \"[Newsletter subject line]\",\n \"body\": \"[HTML body with all the articles]\"\n}\n\nOnly output the JSON. Do not explain anything else.\n\n\n\nHere is the input:{{ $json.aggregatedArticles }}\n\n"
}
]
},
"jsonOutput": true
},
"credentials": {
"openAiApi": {
"id": "fdhWALG84tBLgSZT",
"name": "OpenAi account"
}
},
"typeVersion": 1.8
},
{
"id": "c1266ac2-e2c0-4258-be9d-08c1ae6f3727",
"name": "Gmailでニュースレターを送信",
"type": "n8n-nodes-base.gmail",
"position": [
896,
-160
],
"webhookId": "d883af34-9a6e-406e-8eee-42d501bc5791",
"parameters": {
"sendTo": "",
"message": "={{ $json.message.content.body }}",
"options": {
"appendAttribution": false
},
"subject": "={{ $json.message.content.subject }}"
},
"credentials": {
"gmailOAuth2": {
"id": "j70r3RTMED1pgN3R",
"name": "Gmail account 2"
}
},
"typeVersion": 2.1
},
{
"id": "25e0a2b5-5c27-4176-b9d8-4e91d15d86cc",
"name": "付箋",
"type": "n8n-nodes-base.stickyNote",
"position": [
-800,
-360
],
"parameters": {
"width": 680,
"height": 200,
"content": "### 🌐 Get Website Content Using Dumpling AI\n\nThis section starts with a manual trigger, then fetches a list of website URLs from a connected Google Sheet. Each URL is sent to Dumpling AI’s crawl endpoint, which extracts readable content up to a set depth. The response is an array of article-like results. These are then split one by one for individual processing. The `Set` node maps key fields such as the original URL, article title, and cleaned content for each page.\n"
},
"typeVersion": 1
},
{
"id": "9ca5931f-6637-4371-8830-d73801381276",
"name": "付箋1",
"type": "n8n-nodes-base.stickyNote",
"position": [
300,
-360
],
"parameters": {
"color": 4,
"width": 660,
"height": 220,
"content": "### 🧠 Summarize and Email AI-Generated Newsletter\n\nAfter extracting and mapping each article’s data, the `Code` node formats all the articles into a readable string prompt. GPT-4o is used to create a newsletter-style HTML body and catchy subject line based on the content. Finally, the result is sent using the Gmail node, delivering a ready-to-send AI-generated newsletter straight to the recipient’s inbox.\n"
},
"typeVersion": 1
}
],
"active": false,
"pinData": {},
"settings": {
"executionOrder": "v1"
},
"versionId": "a86325fc-4f26-4052-8883-ec983049f51d",
"connections": {
"b607052f-8712-4cce-977e-6216da30a144": {
"main": [
[
{
"node": "5bf110b3-a678-4a34-a4c9-3a87ab22774f",
"type": "main",
"index": 0
}
]
]
},
"9d69abfa-c867-4fbc-83a2-58723f3c4dfb": {
"main": [
[
{
"node": "eaaa909d-92df-4837-a474-d533577c5621",
"type": "main",
"index": 0
}
]
]
},
"5bf110b3-a678-4a34-a4c9-3a87ab22774f": {
"main": [
[
{
"node": "a996f64a-bb61-493d-90ca-dce9d465fd58",
"type": "main",
"index": 0
}
]
]
},
"eaaa909d-92df-4837-a474-d533577c5621": {
"main": [
[
{
"node": "b2691018-fa64-4a55-a5e9-e68733ab9a7a",
"type": "main",
"index": 0
}
]
]
},
"01d0b913-cfdc-4144-8556-d5c18221cf7c": {
"main": [
[
{
"node": "9d69abfa-c867-4fbc-83a2-58723f3c4dfb",
"type": "main",
"index": 0
}
]
]
},
"a996f64a-bb61-493d-90ca-dce9d465fd58": {
"main": [
[
{
"node": "01d0b913-cfdc-4144-8556-d5c18221cf7c",
"type": "main",
"index": 0
}
]
]
},
"b2691018-fa64-4a55-a5e9-e68733ab9a7a": {
"main": [
[
{
"node": "c1266ac2-e2c0-4258-be9d-08c1ae6f3727",
"type": "main",
"index": 0
}
]
]
}
}
}よくある質問
このワークフローの使い方は?
上記のJSON設定コードをコピーし、n8nインスタンスで新しいワークフローを作成して「JSONからインポート」を選択、設定を貼り付けて認証情報を必要に応じて変更してください。
このワークフローはどんな場面に適していますか?
中級 - 人工知能, マーケティング
有料ですか?
このワークフローは完全無料です。ただし、ワークフローで使用するサードパーティサービス(OpenAI APIなど)は別途料金が発生する場合があります。
関連ワークフロー
Dumpling AIとGPT-4oを使ってAIニュースの自動レビューを生成
Dumpling AI、GPT-4oを使って自動生成してAIニュースのレビューを投稿する
Code
Wait
Split Out
+
Code
Wait
Split Out
14 ノードYang
人工知能
YouTube動画をSEOブログ記事に変換する
YouTube動画をSEOブログ記事に変換するには、GPT-4o、Dumpling AI、Fluxを使用します
Set
Gmail
Markdown
+
Set
Gmail
Markdown
15 ノードYang
人工知能
Dumpling AI + GPT-4oで autocomplete の提案に基づいてタイトルを生成
Dumpling AI+GPT-4oで自動補完提案からタイトルを生成
Set
Split Out
Http Request
+
Set
Split Out
Http Request
8 ノードYang
デザイン
ブログ投稿自動化とSNSプロモーションエージェント
GPT-4、Perplexity、WordPressを使用したSEOブログ作成の自動化+ソーシャルメディア
Set
Code
Gmail
+
Set
Code
Gmail
79 ノードLukaszB
デザイン
私のワークフロー5
AIメールパーソナライゼーションを使用した自動LinkedInリード生成(未完了)
If
Set
Code
+
If
Set
Code
37 ノードMatthieu
営業
Slack 上にユーザーに通知しながら、AI を使って LinkedIn への貢献を作成する
AIを使ってLinkedInでの貢献を作成し、Slackでユーザーに通知する
Set
Code
Html
+
Set
Code
Html
33 ノードDarryn
人工知能