Automatisation de formulaire en article de blog avec Dumpling AI et Google Docs
Ceci est unContent Creation, Multimodal AIworkflow d'automatisation du domainecontenant 17 nœuds.Utilise principalement des nœuds comme Code, Wait, Limit, SplitOut, Aggregate. Utiliser Dumpling AI, OpenAI et Google Docs pour automatiser la création de blog à partir de mots-clés
- •Peut nécessiter les informations d'identification d'authentification de l'API cible
- •Clé API OpenAI
Nœuds utilisés (17)
Catégorie
{
"id": "hCUgw7o0NPlNWR0K",
"meta": {
"instanceId": "a1ae5c8dc6c65e674f9c3947d083abcc749ef2546dff9f4ff01de4d6a36ebfe6",
"templateCredsSetupCompleted": true
},
"name": "Form to Blog Automation using Dumpling AI and Google Docs",
"tags": [],
"nodes": [
{
"id": "5e40b2b3-8d03-46b0-b1da-a1111997da55",
"name": "Soumission de Formulaire (Mots-clés)",
"type": "n8n-nodes-base.formTrigger",
"position": [
-416,
32
],
"webhookId": "1efeb31a-b7f2-4750-8c8a-5aabbdafd57d",
"parameters": {
"options": {},
"formTitle": "Blog form",
"formFields": {
"values": [
{
"fieldLabel": "Keywords"
}
]
}
},
"typeVersion": 2.2
},
{
"id": "4e2c4c76-6d22-4a6b-abd1-8771b2ff21f7",
"name": "Dumpling AI Autocomplete",
"type": "n8n-nodes-base.httpRequest",
"position": [
-192,
32
],
"parameters": {
"url": "https://app.dumplingai.com/api/v1/get-autocomplete",
"method": "POST",
"options": {},
"sendBody": true,
"authentication": "genericCredentialType",
"bodyParameters": {
"parameters": [
{
"name": "query",
"value": "={{ $json.Keywords }}"
},
{
"name": "country",
"value": "US"
}
]
},
"genericAuthType": "httpHeaderAuth"
},
"credentials": {
"httpHeaderAuth": {
"id": "RLFzAcGRepr5eXZB",
"name": "Dumpling AI-n8n"
}
},
"typeVersion": 4.2
},
{
"id": "99e4b141-da82-4131-a38e-d65e14307049",
"name": "Diviser les Suggestions d'Auto-complétion",
"type": "n8n-nodes-base.splitOut",
"position": [
32,
32
],
"parameters": {
"options": {},
"fieldToSplitOut": "suggestions"
},
"typeVersion": 1
},
{
"id": "cc647e9e-fb05-4ea0-ac9c-c5c619ff0d7c",
"name": "Boucle sur les Suggestions",
"type": "n8n-nodes-base.splitInBatches",
"position": [
256,
32
],
"parameters": {
"options": {}
},
"typeVersion": 3
},
{
"id": "dd3edb69-ea7c-4d45-9fec-28400e9c16d7",
"name": "Délai entre les Requêtes",
"type": "n8n-nodes-base.wait",
"position": [
464,
32
],
"webhookId": "7ec6dbb9-5d73-43f1-aec9-ba1d6d6c7890",
"parameters": {},
"typeVersion": 1.1
},
{
"id": "2e768d0f-a2fa-4236-a25f-1ed2ae36a9f1",
"name": "Dumpling AI Actualités Google",
"type": "n8n-nodes-base.httpRequest",
"position": [
672,
32
],
"parameters": {
"url": " https://app.dumplingai.com/api/v1/search-news",
"method": "POST",
"options": {},
"sendBody": true,
"authentication": "genericCredentialType",
"bodyParameters": {
"parameters": [
{
"name": "query",
"value": "={{ $json.value }}"
},
{
"name": "country",
"value": "US"
}
]
},
"genericAuthType": "httpHeaderAuth"
},
"credentials": {
"httpHeaderAuth": {
"id": "RLFzAcGRepr5eXZB",
"name": "Dumpling AI-n8n"
}
},
"typeVersion": 4.2
},
{
"id": "4cc5e45e-3fd1-444b-b872-83ca62196b41",
"name": "Diviser les Articles d'Actualité",
"type": "n8n-nodes-base.splitOut",
"position": [
864,
32
],
"parameters": {
"options": {},
"fieldToSplitOut": "news"
},
"typeVersion": 1
},
{
"id": "3f03dc09-2cbc-4c6a-a798-1a61b81a8073",
"name": "Filtrer les Articles (1–2 Jours)",
"type": "n8n-nodes-base.code",
"position": [
1104,
32
],
"parameters": {
"jsCode": "// Works in n8n Code node\n// Filters only articles that are between 1 and 2 days old\n\nfunction parseRelative(relative, baseNow) {\n if (typeof relative !== 'string') return null;\n const s = relative.trim().toLowerCase();\n\n // quick rejects\n if (s.includes('yesterday')) {\n // treat \"yesterday\" as exactly 1 day ago\n return new Date(baseNow.getTime() - 24 * 3600000);\n }\n if (s.includes('week') || s.includes('month') || s.includes('year')) return null;\n\n // match \"15 hours ago\", \"2 days ago\", etc.\n const m = s.match(/(\\d+)\\s*(second|seconds|sec|secs|minute|min|minutes|mins|hour|hours|day|days)\\s*ago$/);\n if (!m) return null;\n\n const amount = parseInt(m[1], 10);\n const unit = m[2];\n\n const msMap = {\n second: 1000, seconds: 1000, sec: 1000, secs: 1000,\n minute: 60000, minutes: 60000, min: 60000, mins: 60000,\n hour: 3600000, hours: 3600000,\n day: 86400000, days: 86400000,\n };\n\n const ms = msMap[unit];\n if (!ms) return null;\n\n return new Date(baseNow.getTime() - amount * ms);\n}\n\n// Collect articles from either shape\nlet articles = [];\nif (items.length === 1 && Array.isArray(items[0].json)) {\n articles = items[0].json;\n} else {\n articles = items.map(i => i.json);\n}\n\n// Determine \"now\" (or override with present_date if provided)\nlet baseNow = new Date();\nconst present = items?.[0]?.json?.present_date;\nif (typeof present === 'string' && /^\\d{4}-\\d{2}-\\d{2}$/.test(present)) {\n const [y, m, d] = present.split('-').map(n => parseInt(n, 10));\n baseNow = new Date(y, m - 1, d, 12, 0, 0, 0);\n}\n\n// Define boundaries\nconst twoDaysAgo = new Date(baseNow.getTime() - 2 * 24 * 3600000);\nconst oneDayAgo = new Date(baseNow.getTime() - 1 * 24 * 3600000);\n\nconst out = [];\nfor (const a of articles) {\n const rel = a.date || a.published_time || a.publishedAt || a.pubDate || a.time || '';\n const abs = parseRelative(rel, baseNow);\n if (!abs) continue;\n\n // Keep only if between 1 and 2 days ago\n if (abs >= twoDaysAgo && abs < oneDayAgo) {\n const url = a.link || a.url || a.uri;\n if (!url) continue;\n\n out.push({ json: { URL: String(url), score: 10 } });\n }\n}\n\nreturn out;\n"
},
"typeVersion": 2,
"alwaysOutputData": true
},
{
"id": "a62c05a4-a000-4e0a-af8d-834f0623802c",
"name": "Limiter les Articles",
"type": "n8n-nodes-base.limit",
"position": [
1328,
32
],
"parameters": {
"maxItems": 2
},
"typeVersion": 1
},
{
"id": "30ba63a3-5ac6-4a80-ab09-39ead3320774",
"name": "Dumpling AI Scraper",
"type": "n8n-nodes-base.httpRequest",
"onError": "continueRegularOutput",
"position": [
1568,
32
],
"parameters": {
"url": "https://app.dumplingai.com/api/v1/scrape",
"method": "POST",
"options": {},
"sendBody": true,
"authentication": "genericCredentialType",
"bodyParameters": {
"parameters": [
{
"name": "url",
"value": "={{ $json.URL }}"
}
]
},
"genericAuthType": "httpHeaderAuth"
},
"credentials": {
"httpHeaderAuth": {
"id": "RLFzAcGRepr5eXZB",
"name": "Dumpling AI-n8n"
}
},
"typeVersion": 4.2
},
{
"id": "f1ad7937-7877-4c35-9840-27b8ac21d447",
"name": "Nettoyer et Préparer le Contenu de l'Article",
"type": "n8n-nodes-base.code",
"position": [
1776,
32
],
"parameters": {
"jsCode": "// n8n Code node, JavaScript\n\n// If the incoming item is a single item whose .json is an array,\n// make it our working array. Otherwise assume multiple items.\nconst raw = Array.isArray(items[0]?.json) ? items[0].json : items.map(i => i.json);\n\nfunction cleanMarkdown(md) {\n if (!md || typeof md !== \"string\") return \"\";\n\n let text = md;\n\n // Remove markdown images  and bare images \n text = text.replace(/!\\[[^\\]]*?\\]\\([^)]+\\)/g, \"\");\n\n // Replace markdown links [text](url) with just \"text\"\n text = text.replace(/\\[([^\\]]+)\\]\\((?:https?:\\/\\/|mailto:)[^)]+\\)/g, \"$1\");\n\n // Remove bare URLs\n text = text.replace(/https?:\\/\\/\\S+/g, \"\");\n\n // Cut everything after non article sections\n const cutMarkers = [\n /^##\\s*More videos\\b/mi,\n /^##\\s*Related Articles\\b/mi,\n /^##\\s*More Regional News\\b/mi,\n /^\\s*Share This Article\\b/mi,\n /^##\\s*About Me\\b/mi,\n ];\n for (const rx of cutMarkers) {\n const idx = text.search(rx);\n if (idx !== -1) {\n text = text.slice(0, idx);\n break;\n }\n }\n\n // Drop common clutter lines\n const dropStarts = [\n \"Skip to content\",\n \"Skip to main content\",\n \"Share\",\n \"Watch later\",\n \"Copy link\",\n \"Include playlist\",\n \"More videos\",\n \"You are signed out\",\n \"CancelConfirm\",\n \"Search\",\n \"Shopping\",\n \"Info\",\n \"Tap to unmute\",\n \"If playback does not begin\",\n \"Video Player is loading\",\n \"Play Video\",\n \"Stream Type\",\n \"Seek to live\",\n \"Remaining Time\",\n \"Playback Rate\",\n \"Chapters\",\n \"Captions\",\n \"Audio Track\",\n \"Fullscreen\",\n \"Close Modal Dialog\",\n \"Topic:\",\n ];\n text = text\n .split(\"\\n\")\n .filter(line => {\n const L = line.trim();\n if (!L) return true;\n return !dropStarts.some(s => L.startsWith(s));\n })\n .join(\"\\n\");\n\n // Remove heading markers but keep their words\n text = text.replace(/^#{1,6}\\s*/gm, \"\");\n\n // Remove blockquote markers\n text = text.replace(/^\\s*>\\s?/gm, \"\");\n\n // Remove leftover image captions such as \"(Image by ...)\"\n text = text.replace(/\\((?:Image by|Photo|GIF|Video)[^)]+\\)\\s*/gi, \"\");\n\n // Compress extra blank lines\n text = text.replace(/\\n{3,}/g, \"\\n\\n\").trim();\n\n return text;\n}\n\nconst out = raw.map(a => {\n const title = a.title || \"\";\n const content = a.content || \"\";\n return {\n json: {\n title,\n article: cleanMarkdown(content),\n },\n };\n});\n\nreturn out;\n"
},
"typeVersion": 2
},
{
"id": "e122251c-5540-4dd3-be3f-59620c1aacf2",
"name": "Agréger les Articles",
"type": "n8n-nodes-base.aggregate",
"position": [
496,
-208
],
"parameters": {
"include": "specifiedFields",
"options": {},
"aggregate": "aggregateAllItemData",
"fieldsToInclude": "article",
"destinationFieldName": "article"
},
"typeVersion": 1
},
{
"id": "acf43fa6-f8d4-4533-8522-68640916877c",
"name": "OpenAI : Générer une Ébauche de Blog",
"type": "@n8n/n8n-nodes-langchain.openAi",
"position": [
752,
-208
],
"parameters": {
"modelId": {
"__rl": true,
"mode": "list",
"value": "gpt-4.1-mini",
"cachedResultName": "GPT-4.1-MINI"
},
"options": {},
"messages": {
"values": [
{
"role": "system",
"content": "=You are a professional content writer. I will provide you with raw scraped content from different articles. Your job is to create a polished, original blog post based on the ideas and insights in that content.\n\nOutput Requirements:\n- Return ONLY valid JSON (no explanations, no markdown formatting outside JSON, no comments).\n- JSON must have exactly these keys:\n {\n \"Blog_post\": \"Markdown formatted blog post\",\n \"title\": \"A clear, engaging blog post title\"\n }\n\nWriting Instructions:\n1. Write the blog post in engaging, professional, and conversational language.\n2. Use Markdown formatting: \n - Headings (#, ##, ###) for sections\n - Bold or italic where necessary\n - Lists where it improves readability\n3. Do not copy the scraped text directly. Rewrite it in your own words while keeping the meaning.\n4. Exclude clutter such as “related articles,” “share this,” or video references.\n5. Organize with a strong introduction, clear body sections, and a compelling conclusion.\n6. Add smooth transitions between sections so the blog post flows naturally.\n\n"
},
{
"content": "=---\nHere is the scraped content to base the blog post on:\n{{ JSON.stringify($json.article) }}\n---\n"
}
]
},
"jsonOutput": true
},
"credentials": {
"openAiApi": {
"id": "dd8NvMC6rvx8RITo",
"name": "OpenAi account 2"
}
},
"typeVersion": 1.8
},
{
"id": "04ee4323-3527-401c-a963-a726c7f1dad6",
"name": "Docs Google : Créer un Fichier de Blog",
"type": "n8n-nodes-base.googleDocs",
"position": [
1104,
-208
],
"parameters": {
"title": "={{ $json.message.content.title }}",
"folderId": "1NU00YbKNiHJptNuQZH6kgVUhLvDzE0ka"
},
"credentials": {
"googleDocsOAuth2Api": {
"id": "Ofx9WFXGfyxTK2fk",
"name": "Google Docs account"
}
},
"typeVersion": 2
},
{
"id": "bacf9175-722d-4f93-a571-4a5e4e555995",
"name": "Docs Google : Insérer le Contenu du Blog",
"type": "n8n-nodes-base.googleDocs",
"position": [
1312,
-208
],
"parameters": {
"actionsUi": {
"actionFields": [
{
"text": "={{ $('OpenAI: Generate Blog Draft').item.json.message.content.Blog_post }}",
"action": "insert"
}
]
},
"operation": "update",
"documentURL": "={{ $json.id }}"
},
"credentials": {
"googleDocsOAuth2Api": {
"id": "Ofx9WFXGfyxTK2fk",
"name": "Google Docs account"
}
},
"typeVersion": 2
},
{
"id": "cac11f57-cf15-40ec-9c0a-49ad02b48564",
"name": "Note Adhésive",
"type": "n8n-nodes-base.stickyNote",
"position": [
-400,
-208
],
"parameters": {
"width": 624,
"height": 368,
"content": "## Agent Branch\n\nThis branch starts with a form submission. \nThe entered keyword is expanded using **Dumpling AI Autocomplete**, then news articles are fetched with **Dumpling AI Google News**. \n\nThe workflow splits suggestions and articles, filters them to keep only fresh items (1–2 days old), and prepares them for content creation.\n"
},
"typeVersion": 1
},
{
"id": "d382d3f0-74fc-47d3-8b32-8e3c48f5bf40",
"name": "Note Adhésive 1",
"type": "n8n-nodes-base.stickyNote",
"position": [
704,
-384
],
"parameters": {
"width": 752,
"height": 256,
"content": "## Blog Branch\n\nFrom each filtered article, the workflow scrapes and cleans content with **Dumpling AI Scraper**. \nThe articles are aggregated and passed to **OpenAI**, which generates a polished blog draft. \n\nThe final post is saved into **Google Docs**, creating and updating a document with the finished blog content.\n"
},
"typeVersion": 1
}
],
"active": false,
"pinData": {
"Dumpling AI Autocomplete": [
{
"json": {
"credits": 1,
"suggestions": [
{
"value": "ai agents"
},
{
"value": "ai agents explained"
},
{
"value": "ai agents vs agentic ai"
},
{
"value": "ai agents course"
},
{
"value": "ai agents examples"
},
{
"value": "ai agents startup science machine"
},
{
"value": "ai agents reddit"
},
{
"value": "ai agents directory"
},
{
"value": "ai agents tools"
},
{
"value": "ai agents in action"
}
],
"searchParameters": {
"q": "AI agents",
"gl": "us",
"type": "autocomplete",
"engine": "google"
}
}
}
],
"Form Submission (Keywords)": [
{
"json": {
"Keywords": "AI agents",
"formMode": "test",
"submittedAt": "2025-08-23T10:19:28.477+10:00"
}
}
]
},
"settings": {
"executionOrder": "v1"
},
"versionId": "34bb81b5-d086-4f95-b702-3f42cb606281",
"connections": {
"a62c05a4-a000-4e0a-af8d-834f0623802c": {
"main": [
[
{
"node": "30ba63a3-5ac6-4a80-ab09-39ead3320774",
"type": "main",
"index": 0
}
]
]
},
"cc647e9e-fb05-4ea0-ac9c-c5c619ff0d7c": {
"main": [
[
{
"node": "e122251c-5540-4dd3-be3f-59620c1aacf2",
"type": "main",
"index": 0
}
],
[
{
"node": "dd3edb69-ea7c-4d45-9fec-28400e9c16d7",
"type": "main",
"index": 0
}
]
]
},
"e122251c-5540-4dd3-be3f-59620c1aacf2": {
"main": [
[
{
"node": "acf43fa6-f8d4-4533-8522-68640916877c",
"type": "main",
"index": 0
}
]
]
},
"30ba63a3-5ac6-4a80-ab09-39ead3320774": {
"main": [
[
{
"node": "f1ad7937-7877-4c35-9840-27b8ac21d447",
"type": "main",
"index": 0
}
]
]
},
"4cc5e45e-3fd1-444b-b872-83ca62196b41": {
"main": [
[
{
"node": "3f03dc09-2cbc-4c6a-a798-1a61b81a8073",
"type": "main",
"index": 0
}
]
]
},
"dd3edb69-ea7c-4d45-9fec-28400e9c16d7": {
"main": [
[
{
"node": "2e768d0f-a2fa-4236-a25f-1ed2ae36a9f1",
"type": "main",
"index": 0
}
]
]
},
"2e768d0f-a2fa-4236-a25f-1ed2ae36a9f1": {
"main": [
[
{
"node": "4cc5e45e-3fd1-444b-b872-83ca62196b41",
"type": "main",
"index": 0
}
]
]
},
"4e2c4c76-6d22-4a6b-abd1-8771b2ff21f7": {
"main": [
[
{
"node": "99e4b141-da82-4131-a38e-d65e14307049",
"type": "main",
"index": 0
}
]
]
},
"5e40b2b3-8d03-46b0-b1da-a1111997da55": {
"main": [
[
{
"node": "4e2c4c76-6d22-4a6b-abd1-8771b2ff21f7",
"type": "main",
"index": 0
}
]
]
},
"acf43fa6-f8d4-4533-8522-68640916877c": {
"main": [
[
{
"node": "04ee4323-3527-401c-a963-a726c7f1dad6",
"type": "main",
"index": 0
}
]
]
},
"04ee4323-3527-401c-a963-a726c7f1dad6": {
"main": [
[
{
"node": "bacf9175-722d-4f93-a571-4a5e4e555995",
"type": "main",
"index": 0
}
]
]
},
"99e4b141-da82-4131-a38e-d65e14307049": {
"main": [
[
{
"node": "cc647e9e-fb05-4ea0-ac9c-c5c619ff0d7c",
"type": "main",
"index": 0
}
]
]
},
"f1ad7937-7877-4c35-9840-27b8ac21d447": {
"main": [
[
{
"node": "cc647e9e-fb05-4ea0-ac9c-c5c619ff0d7c",
"type": "main",
"index": 0
}
]
]
},
"3f03dc09-2cbc-4c6a-a798-1a61b81a8073": {
"main": [
[
{
"node": "a62c05a4-a000-4e0a-af8d-834f0623802c",
"type": "main",
"index": 0
}
]
]
}
}
}Comment utiliser ce workflow ?
Copiez le code de configuration JSON ci-dessus, créez un nouveau workflow dans votre instance n8n et sélectionnez "Importer depuis le JSON", collez la configuration et modifiez les paramètres d'authentification selon vos besoins.
Dans quelles scénarios ce workflow est-il adapté ?
Avancé - Création de contenu, IA Multimodale
Est-ce payant ?
Ce workflow est entièrement gratuit et peut être utilisé directement. Veuillez noter que les services tiers utilisés dans le workflow (comme l'API OpenAI) peuvent nécessiter un paiement de votre part.
Workflows recommandés
Yang
@yangPartager ce workflow