This workflow follows the Notion → OpenAI recipe pattern — see all workflows that pair these two integrations.
The workflow JSON
Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →
{
"name": "AI Innovation Feed \u2192 Notion (Top 3 every 3 days)",
"nodes": [
{
"id": "Cron_3Day",
"name": "Every 3 Days @ 9:00 ET",
"type": "n8n-nodes-base.cron",
"typeVersion": 2,
"position": [
200,
200
],
"parameters": {
"triggerTimes": {
"item": [
{
"mode": "everyX",
"everyX": 3,
"unit": "days"
}
]
}
},
"notesInFlow": true,
"notes": "Runs every 3 days. n8n server timezone should be America/New_York or adjust in your host."
},
{
"id": "RSS_OpenAI",
"name": "RSS: OpenAI",
"type": "n8n-nodes-base.rssFeedRead",
"typeVersion": 1,
"position": [
480,
40
],
"parameters": {
"url": "https://openai.com/blog/rss.xml"
}
},
{
"id": "RSS_Anthropic",
"name": "RSS: Anthropic",
"type": "n8n-nodes-base.rssFeedRead",
"typeVersion": 1,
"position": [
480,
160
],
"parameters": {
"url": "https://www.anthropic.com/news/rss.xml"
}
},
{
"id": "RSS_GoogleAI",
"name": "RSS: Google AI",
"type": "n8n-nodes-base.rssFeedRead",
"typeVersion": 1,
"position": [
480,
280
],
"parameters": {
"url": "https://ai.googleblog.com/atom.xml"
}
},
{
"id": "RSS_HF",
"name": "RSS: Hugging Face",
"type": "n8n-nodes-base.rssFeedRead",
"typeVersion": 1,
"position": [
480,
400
],
"parameters": {
"url": "https://huggingface.co/blog/rss.xml"
}
},
{
"id": "RSS_arXiv",
"name": "RSS: arXiv cs.AI",
"type": "n8n-nodes-base.rssFeedRead",
"typeVersion": 1,
"position": [
480,
520
],
"parameters": {
"url": "https://export.arxiv.org/rss/cs.AI"
}
},
{
"id": "Merge_1",
"name": "Merge Feeds (1)",
"type": "n8n-nodes-base.merge",
"typeVersion": 2,
"position": [
720,
120
],
"parameters": {
"mode": "append"
}
},
{
"id": "Merge_2",
"name": "Merge Feeds (2)",
"type": "n8n-nodes-base.merge",
"typeVersion": 2,
"position": [
920,
200
],
"parameters": {
"mode": "append"
}
},
{
"id": "Merge_3",
"name": "Merge Feeds (3)",
"type": "n8n-nodes-base.merge",
"typeVersion": 2,
"position": [
1120,
280
],
"parameters": {
"mode": "append"
}
},
{
"id": "Merge_4",
"name": "Merge Feeds (4)",
"type": "n8n-nodes-base.merge",
"typeVersion": 2,
"position": [
1320,
360
],
"parameters": {
"mode": "append"
}
},
{
"id": "Fn_Normalize",
"name": "Normalize Items",
"type": "n8n-nodes-base.function",
"typeVersion": 2,
"position": [
1540,
360
],
"parameters": {
"functionCode": "const out = [];\nfor (const item of items) {\n const i = item.json;\n // n8n RSS node outputs fields like: title, link, content, isoDate, pubDate\n out.push({\n json: {\n title: i.title || '',\n link: i.link || i.guid || '',\n content: i.content || i['content:encoded'] || i.summary || '',\n pubDate: i.isoDate || i.pubDate || null,\n source: i['feed']?.title || i.source || '',\n }\n })\n}\nreturn out;"
}
},
{
"id": "Fn_ScoreTop3",
"name": "Score + Pick Top 3",
"type": "n8n-nodes-base.function",
"typeVersion": 2,
"position": [
1740,
360
],
"parameters": {
"functionCode": "function daysAgo(d) {\n if (!d) return 9999;\n const dt = new Date(d);\n return Math.floor((Date.now() - dt.getTime()) / (1000*60*60*24));\n}\nconst KEYWORDS = [\n 'rag','retrieval','agent','agents','agentic','workflow','automation','prompt','prompting','guardrails','eval','evaluation','langchain','langgraph','n8n','zapier','make','vertex','watsonx','openai','anthropic','google','hugging face','vector','embedding','faiss','chroma','pinecone','mcp','connector','tool use','function calling'\n];\nconst seen = new Set();\nconst scored = items.map(item => {\n const j = item.json;\n const age = daysAgo(j.pubDate);\n const text = (j.title + ' ' + (j.content||'')).toLowerCase();\n let relevance = 0;\n for (const k of KEYWORDS) if (text.includes(k)) relevance += 1;\n const recency = Math.max(0, 10 - age); // fresher is higher\n const score = recency * 1.5 + relevance * 2.0;\n return { ...j, age, score };\n}).filter(x => x.age <= 5); // only last ~5 days\n\n// dedupe by link/title\nconst dedup = [];\nfor (const x of scored) {\n const key = (x.link||'') + '|' + (x.title||'');\n if (!seen.has(key)) { seen.add(key); dedup.push(x); }\n}\n\ndedup.sort((a,b)=> b.score - a.score);\nreturn dedup.slice(0,3).map(j=>({json:j}));"
}
},
{
"id": "OpenAI_Researcher",
"name": "OpenAI: Researcher \u2192 Outline",
"type": "n8n-nodes-base.openAi",
"typeVersion": 4,
"position": [
1960,
300
],
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"parameters": {
"resource": "chat",
"operation": "createChatCompletion",
"model": "gpt-4o-mini",
"messages": "={{[\n { role: 'system', content: 'You are a strict research analyst for an AI Automation Engineer. Produce a fact-grounded outline with citations. Do not hallucinate.' },\n { role: 'user', content: `Create a structured outline for a short blog about this item. Use ONLY the info below and the linked page. Include: TL;DR bullets; What happened; Why it matters to automation/prompt engineers; How to use it (actionable steps/prompts); Risks/limits; Citations.\\n\\nTitle: ${$json.title}\\nURL: ${$json.link}\\nSnippet: ${($json.content||'').slice(0,1500)}` }\n]}}",
"temperature": 0.2,
"maxTokens": 800
}
},
{
"id": "OpenAI_Writer",
"name": "OpenAI: Writer \u2192 Full Blog",
"type": "n8n-nodes-base.openAi",
"typeVersion": 4,
"position": [
2160,
300
],
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"parameters": {
"resource": "chat",
"operation": "createChatCompletion",
"model": "gpt-4o",
"messages": "={{[\n { role: 'system', content: 'You are a technical writer for Synth Alchemy. Write concise, actionable posts for AI automation & prompt engineers. Use short sentences. No fluff. Include citations with the source URL provided. American English.' },\n { role: 'user', content: `Turn this outline into a 700\u20131000 word blog post. Use sections: Title, TL;DR, What happened, Why it matters for automation/prompt engineers, How to use it (steps/prompts), Caveats & limits, Sources. Add 1\u20132 specific prompts or workflow steps tailored to n8n/automation.\\n\\nOutline:\\n${$json.choices[0].message.content}` }\n]}}",
"temperature": 0.3,
"maxTokens": 1400
}
},
{
"id": "Fn_PackForNotion",
"name": "Pack Fields for Notion",
"type": "n8n-nodes-base.function",
"typeVersion": 2,
"position": [
2360,
300
],
"parameters": {
"functionCode": "const results = [];\nfor (const item of items) {\n const blog = item.json.choices?.[0]?.message?.content || '';\n const titleMatch = blog.match(/^\\s*#?\\s*(.+?)\\n/);\n const title = titleMatch ? titleMatch[1].trim().slice(0, 200) : this.getNode().name + ' Post';\n const sourceUrl = $item(0).$node[\"Fn_ScoreTop3\"].json.link || '';\n const dateFound = new Date().toISOString();\n // crude tags\n const t = ($item(0).$node[\"Fn_ScoreTop3\"].json.title + ' ' + $item(0).$node[\"Fn_ScoreTop3\"].json.content).toLowerCase();\n const tags = [];\n const tagMap = ['RAG','Agents','Prompting','Automation','LangChain','n8n','Zapier','Make','Guardrails','Evals','LLM'];\n const kwMap = ['rag','agent','prompt','automation','langchain','n8n','zapier','make','guardrail','eval','llm'];\n kwMap.forEach((kw, idx)=>{ if (t.includes(kw)) tags.push(tagMap[idx]); });\n results.push({ json: { title, blog, sourceUrl, dateFound, tags, relevance: 0, link: sourceUrl } });\n}\nreturn results;"
}
},
{
"id": "Notion_Create",
"name": "Notion: Create Page",
"type": "n8n-nodes-base.notion",
"typeVersion": 2,
"position": [
2580,
300
],
"credentials": {
"notionApi": {
"name": "<your credential>"
}
},
"parameters": {
"resource": "databasePage",
"operation": "create",
"databaseId": "{{NOTION_DATABASE_ID}}",
"propertiesUi": {
"property": [
{
"key": "Title",
"type": "title",
"title": {
"title": [
{
"text": {
"content": "={{$json.title}}"
}
}
]
}
},
{
"key": "Date Found",
"type": "date",
"date": {
"date": {
"start": "={{$json.dateFound}}"
}
}
},
{
"key": "Source URL",
"type": "url",
"url": "={{$json.sourceUrl}}"
},
{
"key": "Tags",
"type": "multiSelect",
"multiSelect": {
"multi_select": "={{$json.tags.map(t=>({name:t}))}}"
}
},
{
"key": "AI Relevance Score",
"type": "number",
"number": "={{$item(0).$node[\"Fn_ScoreTop3\"].json.score || 0}}"
}
]
},
"blocksUi": {
"block": [
{
"type": "heading_2",
"heading_2": {
"text": [
{
"type": "text",
"text": {
"content": "Blog Post"
}
}
]
}
},
{
"type": "paragraph",
"paragraph": {
"text": [
{
"type": "text",
"text": {
"content": "={{$json.blog}}"
}
}
]
}
}
]
}
}
},
{
"id": "Split_Top3",
"name": "Split Top 3 (Iterate)",
"type": "n8n-nodes-base.splitInBatches",
"typeVersion": 1,
"position": [
1880,
360
],
"parameters": {
"batchSize": 1
}
}
],
"connections": {
"Every 3 Days @ 9:00 ET": {
"main": [
[
{
"node": "RSS: OpenAI",
"type": "main",
"index": 0
},
{
"node": "RSS: Anthropic",
"type": "main",
"index": 0
},
{
"node": "RSS: Google AI",
"type": "main",
"index": 0
},
{
"node": "RSS: Hugging Face",
"type": "main",
"index": 0
},
{
"node": "RSS: arXiv cs.AI",
"type": "main",
"index": 0
}
]
]
},
"RSS: OpenAI": {
"main": [
[
{
"node": "Merge Feeds (1)",
"type": "main",
"index": 0
}
]
]
},
"RSS: Anthropic": {
"main": [
[
{
"node": "Merge Feeds (1)",
"type": "main",
"index": 1
}
]
]
},
"Merge Feeds (1)": {
"main": [
[
{
"node": "Merge Feeds (2)",
"type": "main",
"index": 0
}
]
]
},
"RSS: Google AI": {
"main": [
[
{
"node": "Merge Feeds (2)",
"type": "main",
"index": 1
}
]
]
},
"Merge Feeds (2)": {
"main": [
[
{
"node": "Merge Feeds (3)",
"type": "main",
"index": 0
}
]
]
},
"RSS: Hugging Face": {
"main": [
[
{
"node": "Merge Feeds (3)",
"type": "main",
"index": 1
}
]
]
},
"Merge Feeds (3)": {
"main": [
[
{
"node": "Merge Feeds (4)",
"type": "main",
"index": 0
}
]
]
},
"RSS: arXiv cs.AI": {
"main": [
[
{
"node": "Merge Feeds (4)",
"type": "main",
"index": 1
}
]
]
},
"Merge Feeds (4)": {
"main": [
[
{
"node": "Normalize Items",
"type": "main",
"index": 0
}
]
]
},
"Normalize Items": {
"main": [
[
{
"node": "Score + Pick Top 3",
"type": "main",
"index": 0
}
]
]
},
"Score + Pick Top 3": {
"main": [
[
{
"node": "Split Top 3 (Iterate)",
"type": "main",
"index": 0
}
]
]
},
"Split Top 3 (Iterate)": {
"main": [
[
{
"node": "OpenAI: Researcher \u2192 Outline",
"type": "main",
"index": 0
}
]
]
},
"OpenAI: Researcher \u2192 Outline": {
"main": [
[
{
"node": "OpenAI: Writer \u2192 Full Blog",
"type": "main",
"index": 0
}
]
]
},
"OpenAI: Writer \u2192 Full Blog": {
"main": [
[
{
"node": "Pack Fields for Notion",
"type": "main",
"index": 0
}
]
]
},
"Pack Fields for Notion": {
"main": [
[
{
"node": "Notion: Create Page",
"type": "main",
"index": 0
}
]
]
},
"Notion: Create Page": {
"main": [
[
{
"node": "Split Top 3 (Iterate)",
"type": "main",
"index": 1
}
]
]
}
},
"active": false,
"settings": {
"saveExecutionProgress": true,
"executionOrder": "v1"
},
"version": 1
}
Credentials you'll need
Each integration node will prompt for credentials when you import. We strip credential IDs before publishing — you'll add your own.
notionApiopenAiApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
AI Innovation Feed → Notion (Top 3 every 3 days). Uses rssFeedRead, openAi, notion. Scheduled trigger; 17 nodes.
Source: https://gist.github.com/zcharlesholland-netizen/3ab3ed9b5d1ab9473a2f173af79dd1f9 — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
Categories: AI, Content Creation, Marketing Automation, Productivity
RSS Feed Read, OpenAI, Notion
AI-powered priority re-evaluation every 2 hours. Analyzes new signals, meeting decisions, emails, and blockers, then runs 3 AI passes (Impact, Urgency, Final Ranking) to suggest re-ranking. Only updat
Notion, OpenAI, Slack
YouTube Automation Pipeline - Notion + Gemini + CometAPI + JSON2Video. Uses notion, httpRequest, googleDrive, writeBinaryFile. Scheduled trigger; 43 nodes.
Notion, HTTP Request, Google Drive +3
This workflow automatically collects the latest technology news, filters for emerging topics, and uses AI to score relevance and generate clean, ready-to-share content. It helps you focus on high-impa
RSS Feed Read, OpenAI, Google Sheets +1
Daily trigger scans your Notion database for unpublished blog ideas AI generates complete blog posts + engaging LinkedIn content using OpenAI (Blog Posting is not implemented yet) Creates custom image
Notion, Error Trigger, Gmail +3