Abdullah Alshiekh This workflow corresponds to n8n.io template #12075 — we link there as the canonical source.
This workflow follows the Agent → OpenAI Chat 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 →
{
"id": "rgpHmMpjYVZFi020",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "Price Monitor (Template)",
"tags": [],
"nodes": [
{
"id": "f9298f65-d6ca-49b5-8aa9-8949e3d300e6",
"name": "Main Sticky",
"type": "n8n-nodes-base.stickyNote",
"position": [
-2608,
1088
],
"parameters": {
"width": 450,
"height": 612,
"content": "### \ud83d\udecd\ufe0f AI Price Monitor & Aggregator\n\n#### How it works\nThis workflow takes a product name from Telegram, searches Google for purchase links (Amazon EG & Jumia), scrapes the prices, and uses AI to summarize the best deal.\n1. **Trigger:** Receive product name via Telegram.\n2. **Search:** Google Search finds relevant URLs.\n3. **Scrape:** Parallel processing scrapes Amazon and Jumia/Other sites separately.\n4. **AI Analysis:** GPT-4o extracts the product details and prices from the raw HTML.\n5. **Response:** Sends a clean summary back to Telegram.\n\n#### Setup steps\n1. **Credentials:** configure your `Telegram API`, `OpenAI`, and `Decodo` credentials.\n2. **Decodo:** Ensure your Decodo account allows Google Search operations.\n3. **Webhook:** Verify the Telegram webhook is active.\n\n#### Customization tips\n- **Sites:** Edit the `Build Queries` node to add more stores (e.g., noon.com).\n- **Filters:** Adjust the `Router` node if you want to add specific scraping logic for other domains."
},
"typeVersion": 1
},
{
"id": "51a537b5-3a24-4055-a3b4-4ad53ef52a42",
"name": "Section 1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-2144,
1232
],
"parameters": {
"color": 7,
"width": 900,
"height": 260,
"content": "## 1. Get Links & Route\nReceives the user request, searches Google, and separates Amazon links from others for specialized handling."
},
"typeVersion": 1
},
{
"id": "4b054f9d-c25c-4166-bb02-ac3257c1a0d3",
"name": "Section 2",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1296,
896
],
"parameters": {
"color": 7,
"width": 750,
"height": 300,
"content": "## 2. Amazon Scraping\nSpecialized HTML cleaning and scoring logic optimized for Amazon product pages."
},
"typeVersion": 1
},
{
"id": "4feef981-8b47-4b59-8066-028c6d6c5589",
"name": "Section 3",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1328,
1632
],
"parameters": {
"color": 7,
"width": 750,
"height": 300,
"content": "## 3. General Scraping\nGeneric scraping logic for Jumia and other supported e-commerce sites."
},
"typeVersion": 1
},
{
"id": "be4e928a-c8a9-45cf-9a32-7d95d27b6d03",
"name": "Section 4",
"type": "n8n-nodes-base.stickyNote",
"position": [
-800,
1232
],
"parameters": {
"color": 7,
"width": 600,
"height": 380,
"content": "## 4. AI Extraction & Reply\nMerges results and uses an AI Agent to extract structured price data and reply to the user."
},
"typeVersion": 1
},
{
"id": "6d1b1c0b-fdc6-4d34-b364-4bffca5ec4c3",
"name": "Telegram Trigger",
"type": "n8n-nodes-base.telegramTrigger",
"position": [
-2112,
1344
],
"parameters": {
"updates": [
"message"
],
"additionalFields": {}
},
"typeVersion": 1.2
},
{
"id": "b345f524-39b0-494d-aab3-cda1f8a886b4",
"name": "Build Queries",
"type": "n8n-nodes-base.code",
"position": [
-1904,
1344
],
"parameters": {
"jsCode": "const product = $json.message.text?.trim();\nif (!product) return [];\n\nconst stores = [\n { store: \"amazon\", site: \"amazon.eg\" },\n { store: \"jumia\", site: \"jumia.com.eg\" },\n];\n\nreturn stores.map(s => ({\n json: {\n store: s.store,\n product,\n query: `${product} site:${s.site}`\n }\n}));\n"
},
"typeVersion": 2
},
{
"id": "e899ff7e-84cf-43e8-93dc-03f035134f1c",
"name": "AI Extraction Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
-640,
1344
],
"parameters": {
"text": "=User search intent:\n{{ $('Telegram Trigger').first().json.message.text }}\n\nScraped data (multiple sources):\n{{ $json.text }}\n",
"options": {
"systemMessage": "You are a product comparison extraction agent.\n\nYou will receive scraped and cleaned HTML text from multiple e-commerce pages\n(Amazon, Jumia, or others), along with their URLs and relevance scores.\n\nYour job is to:\n\n1. Identify the SAME product across different platforms.\n2. For EACH platform extract:\n - Product name (clean, human readable)\n - Price (number + currency if present)\n - Platform name (Amazon, Jumia, etc.)\n - URL\n3. Compare the platforms.\n4. Prepare a final user-friendly message suitable for Telegram.\n\nSTRICT RULES:\n- Do NOT invent prices or product names.\n- If price is missing, write \"Price not found\".\n- If multiple prices exist on the same page, choose the most prominent product price.\n- Ignore ads, bundles, delivery fees, unrelated products.\n- Use ONLY the provided text.\n- Be concise and clear.\n- Use emojis moderately (Telegram friendly).\n\nOUTPUT FORMAT:\nReturn ONLY a single formatted message string.\nNo JSON.\nNo explanations.\nNo markdown blocks.\n"
},
"promptType": "define"
},
"typeVersion": 3
},
{
"id": "91e84561-d41a-464c-90eb-13174e459871",
"name": "Google Search",
"type": "@decodo/n8n-nodes-decodo.decodo",
"position": [
-1728,
1344
],
"parameters": {
"query": "={{ $json.query }}",
"headless": false,
"operation": "google_search",
"results_limit": 1
},
"typeVersion": 1
},
{
"id": "57f34ec5-1b04-4c7d-aa33-9d3c398f2bec",
"name": "Extract & Filter URLs",
"type": "n8n-nodes-base.code",
"position": [
-1520,
1344
],
"parameters": {
"jsCode": "// Collect URLs from Decodo Google Search output\nconst allItems = $input.all();\n\nconst urls = [];\n\nfunction collectUrls(obj) {\n if (!obj || typeof obj !== 'object') return;\n\n // Collect URL fields\n if (typeof obj.url === 'string' && obj.url.startsWith('http')) {\n urls.push(obj.url);\n }\n\n // Recurse into arrays\n if (Array.isArray(obj)) {\n for (const item of obj) {\n collectUrls(item);\n }\n }\n // Recurse into objects\n else {\n for (const key of Object.keys(obj)) {\n collectUrls(obj[key]);\n }\n }\n}\n\n// Walk through every Decodo item\nfor (const item of allItems) {\n collectUrls(item.json);\n}\n\n// Deduplicate + remove Google search URLs\nconst uniqueUrls = [...new Set(urls)]\n .filter(url => !url.includes('google.com/search'))\n .slice(0, 20); // \u2705 LIMIT TO 20 ONLY\n\n// Return one item per URL\nreturn uniqueUrls.map(url => ({\n json: { url }\n}));\n"
},
"typeVersion": 2
},
{
"id": "98f99c7f-c4a1-4599-8d55-43ff58dff4f0",
"name": "Router (Amazon vs. Other)",
"type": "n8n-nodes-base.if",
"position": [
-1344,
1344
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 3,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "c783147f-8282-43f6-9b12-8e2526f35be9",
"operator": {
"type": "string",
"operation": "contains"
},
"leftValue": "={{ $json.url }}",
"rightValue": "amazon"
}
]
}
},
"typeVersion": 2.3
},
{
"id": "b88b3738-780f-4d91-8071-06739ede596f",
"name": "Scrape Amazon",
"type": "@decodo/n8n-nodes-decodo.decodo",
"position": [
-1072,
1008
],
"parameters": {
"url": "={{ $('Extract & Filter URLs').item.json.url }}",
"parse": false,
"operation": "amazon"
},
"typeVersion": 1
},
{
"id": "719695c2-8b79-49bb-9637-43abece23440",
"name": "Loop (Amazon)",
"type": "n8n-nodes-base.splitInBatches",
"position": [
-1232,
992
],
"parameters": {
"options": {}
},
"typeVersion": 3
},
{
"id": "886ec5de-c344-490f-a8aa-70909b6b60d4",
"name": "Clean HTML (Amazon)",
"type": "n8n-nodes-base.code",
"position": [
-912,
1008
],
"parameters": {
"mode": "runOnceForEachItem",
"jsCode": "const item = $input.item.json;\n\nlet html = \"\";\n\ntry {\n html = item.results[0].content || \"\";\n} catch (e) {\n html = \"\";\n}\n\nlet cleaned = html\n .replace(/<script[^>]*>[\\s\\S]*?<\\/script>/gi, \"\")\n .replace(/<style[^>]*>[\\s\\S]*?<\\/style>/gi, \"\")\n .replace(/<noscript[^>]*>[\\s\\S]*?<\\/noscript>/gi, \"\")\n .replace(/<\\/?[^>]+>/g, \" \")\n .replace(/ /gi, \" \")\n .replace(/&/gi, \"&\")\n .replace(/"/gi, '\"')\n .replace(/'/gi, \"'\")\n .replace(/</gi, \"<\")\n .replace(/>/gi, \">\")\n .replace(/\\s+/g, \" \")\n .trim();\n\nreturn {\n json: {\n text_clean: cleaned,\n url: $('Loop (Amazon)').first().json.url\n }\n};\n"
},
"typeVersion": 2
},
{
"id": "28e8d953-d8f5-4946-b2d5-a09a156cc63e",
"name": "Scrape Other",
"type": "@decodo/n8n-nodes-decodo.decodo",
"position": [
-1072,
1728
],
"parameters": {
"url": "={{ $json.url }}"
},
"typeVersion": 1
},
{
"id": "ca82c8e1-bd5e-4432-84d9-8bf30b45b098",
"name": "Loop (Other)",
"type": "n8n-nodes-base.splitInBatches",
"position": [
-1232,
1712
],
"parameters": {
"options": {}
},
"typeVersion": 3
},
{
"id": "08182aff-c250-4133-8dea-45dba46d1c68",
"name": "Clean HTML (Other)",
"type": "n8n-nodes-base.code",
"position": [
-896,
1728
],
"parameters": {
"mode": "runOnceForEachItem",
"jsCode": "const item = $input.item.json;\n\nlet html = \"\";\n\ntry {\n html = item.results[0].content || \"\";\n} catch (e) {\n html = \"\";\n}\n\nlet cleaned = html\n .replace(/<script[^>]*>[\\s\\S]*?<\\/script>/gi, \"\")\n .replace(/<style[^>]*>[\\s\\S]*?<\\/style>/gi, \"\")\n .replace(/<noscript[^>]*>[\\s\\S]*?<\\/noscript>/gi, \"\")\n .replace(/<\\/?[^>]+>/g, \" \")\n .replace(/ /gi, \" \")\n .replace(/&/gi, \"&\")\n .replace(/"/gi, '\"')\n .replace(/'/gi, \"'\")\n .replace(/</gi, \"<\")\n .replace(/>/gi, \">\")\n .replace(/\\s+/g, \" \")\n .trim();\n\nreturn {\n json: {\n text_clean: cleaned,\n url: $('Loop (Other)').first().json.url\n }\n};\n"
},
"typeVersion": 2
},
{
"id": "95becc17-a85a-42b3-8c7f-41707045543d",
"name": "Score Match (Other)",
"type": "n8n-nodes-base.code",
"position": [
-736,
1728
],
"parameters": {
"jsCode": "// 1\ufe0f\u20e3 User message (dynamic, from Telegram)\nconst userMessage =\n $('Telegram Trigger').first().json.message.text\n .toLowerCase()\n .replace(/[^\\w\\s\\u0600-\\u06FF]/g, ' ');\n\n// 2\ufe0f\u20e3 Tokenize user message dynamically\nconst userTokens = userMessage\n .split(/\\s+/)\n .filter(w => w.length >= 3);\n\n// 3\ufe0f\u20e3 Cleaned page text\nconst pageText = ($json.text_clean || '')\n .toLowerCase()\n .replace(/[^\\w\\s\\u0600-\\u06FF]/g, ' ');\n\n// 4\ufe0f\u20e3 URL (pass-through from earlier nodes)\nconst url = $json.url || null;\n\n// Guards\nif (!pageText || userTokens.length === 0) {\n return [{\n json: {\n match: false,\n score: 0,\n url\n }\n }];\n}\n\n// 5\ufe0f\u20e3 Dynamic scoring\nlet score = 0;\nlet matchedTokens = [];\n\nfor (const token of userTokens) {\n const occurrences = pageText.split(token).length - 1;\n if (occurrences > 0) {\n score += Math.min(occurrences * 3, 15); // density-aware\n matchedTokens.push(token);\n }\n}\n\n// 6\ufe0f\u20e3 Structural bonuses (language-agnostic)\nif (pageText.length > 1500) score += 5; // real product page\nif (pageText.length < 400) score -= 10; // weak page\nif (\n pageText.includes('add to cart') ||\n pageText.includes('\u0627\u0636\u0641') ||\n pageText.includes('\u0623\u0636\u0641')\n) score += 5;\n\n// 7\ufe0f\u20e3 Dynamic threshold (based on user input length)\nconst threshold = Math.max(12, userTokens.length * 4);\nconst isMatch = score >= threshold;\n\n// 8\ufe0f\u20e3 Output EVERYTHING needed downstream\nreturn [{\n json: {\n match: isMatch,\n score,\n threshold,\n url,\n matched_tokens: matchedTokens,\n text_clean: pageText\n }\n}];\n"
},
"typeVersion": 2
},
{
"id": "502a928b-2091-4b63-9716-651337efa735",
"name": "Score Match (Amazon)",
"type": "n8n-nodes-base.code",
"position": [
-736,
1008
],
"parameters": {
"jsCode": "// 1\ufe0f\u20e3 User message (dynamic, from Telegram)\nconst userMessage =\n $('Telegram Trigger').first().json.message.text\n .toLowerCase()\n .replace(/[^\\w\\s\\u0600-\\u06FF]/g, ' ');\n\n// 2\ufe0f\u20e3 Tokenize user message dynamically\nconst userTokens = userMessage\n .split(/\\s+/)\n .filter(w => w.length >= 3);\n\n// 3\ufe0f\u20e3 Cleaned page text\nconst pageText = ($json.text_clean || '')\n .toLowerCase()\n .replace(/[^\\w\\s\\u0600-\\u06FF]/g, ' ');\n\n// 4\ufe0f\u20e3 URL (pass-through from earlier nodes)\nconst url = $json.url || null;\n\n// Guards\nif (!pageText || userTokens.length === 0) {\n return [{\n json: {\n match: false,\n score: 0,\n url\n }\n }];\n}\n\n// 5\ufe0f\u20e3 Dynamic scoring\nlet score = 0;\nlet matchedTokens = [];\n\nfor (const token of userTokens) {\n const occurrences = pageText.split(token).length - 1;\n if (occurrences > 0) {\n score += Math.min(occurrences * 3, 15); // density-aware\n matchedTokens.push(token);\n }\n}\n\n// 6\ufe0f\u20e3 Structural bonuses (language-agnostic)\nif (pageText.length > 1500) score += 5; // real product page\nif (pageText.length < 400) score -= 10; // weak page\nif (\n pageText.includes('add to cart') ||\n pageText.includes('\u0627\u0636\u0641') ||\n pageText.includes('\u0623\u0636\u0641')\n) score += 5;\n\n// 7\ufe0f\u20e3 Dynamic threshold (based on user input length)\nconst threshold = Math.max(12, userTokens.length * 4);\nconst isMatch = score >= threshold;\n\n// 8\ufe0f\u20e3 Output EVERYTHING needed downstream\nreturn [{\n json: {\n match: isMatch,\n score,\n threshold,\n url,\n matched_tokens: matchedTokens,\n text_clean: pageText\n }\n}];\n"
},
"typeVersion": 2
},
{
"id": "d373cd04-5d7b-42fb-bb8e-bff1e13bf9e8",
"name": "Pick Best (Amazon)",
"type": "n8n-nodes-base.code",
"position": [
-1120,
1280
],
"parameters": {
"jsCode": "const items = $input.all().map(i => i.json);\n\n// Keep only valid matches\nconst valid = items.filter(i => i.match === true);\n\n// If none matched, return nothing\nif (!valid.length) return [];\n\n// Pick highest score\nvalid.sort((a, b) => b.score - a.score);\n\nreturn [{\n json: valid[0]\n}];\n"
},
"typeVersion": 2
},
{
"id": "dd39e9a7-5491-487a-a4d2-9bfb636f645b",
"name": "Pick Best (Other)",
"type": "n8n-nodes-base.code",
"position": [
-1120,
1424
],
"parameters": {
"jsCode": "const items = $input.all().map(i => i.json);\n\n// Keep only valid matches\nconst valid = items.filter(i => i.match === true);\n\n// If none matched, return nothing\nif (!valid.length) return [];\n\n// Pick highest score\nvalid.sort((a, b) => b.score - a.score);\n\nreturn [{\n json: valid[0]\n}];\n"
},
"typeVersion": 2
},
{
"id": "d8c9cb70-d67b-401c-8191-474efc31eaac",
"name": "Merge Results",
"type": "n8n-nodes-base.merge",
"position": [
-944,
1344
],
"parameters": {},
"typeVersion": 3.2
},
{
"id": "c863240d-717b-4557-9aa9-cdbe7944c45c",
"name": "GPT Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
-640,
1504
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4o-mini"
},
"options": {}
},
"typeVersion": 1.2
},
{
"id": "a3b4c774-7d06-4188-9503-8735319cbae8",
"name": "Format Context",
"type": "n8n-nodes-base.code",
"position": [
-800,
1344
],
"parameters": {
"jsCode": "const items = $input.all().map(i => i.json);\n\nconst combinedText = items.map(item => `\nSOURCE URL:\n${item.url}\n\nRELEVANCE SCORE:\n${item.score}\n\nCONTENT:\n${item.text_clean.slice(0, 5000)}\n`).join('\\n\\n====================\\n\\n');\n\nreturn [{\n json: {\n text: combinedText // \ud83d\udd34 THIS NAME IS CRITICAL\n }\n}];\n"
},
"typeVersion": 2
},
{
"id": "fb221ed8-fa1c-4483-b9d2-38289e5a21df",
"name": "Send Reply",
"type": "n8n-nodes-base.telegram",
"position": [
-336,
1344
],
"parameters": {
"text": "={{ $json.output }}",
"chatId": "={{ $('Telegram Trigger').item.json.message.chat.id }}",
"additionalFields": {
"appendAttribution": false
}
},
"typeVersion": 1.2
}
],
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "9d7a4e7f-c857-4178-a66d-490c22015ce6",
"connections": {
"GPT Model": {
"ai_languageModel": [
[
{
"node": "AI Extraction Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Loop (Other)": {
"main": [
[
{
"node": "Pick Best (Other)",
"type": "main",
"index": 0
}
],
[
{
"node": "Scrape Other",
"type": "main",
"index": 0
}
]
]
},
"Scrape Other": {
"main": [
[
{
"node": "Clean HTML (Other)",
"type": "main",
"index": 0
}
]
]
},
"Build Queries": {
"main": [
[
{
"node": "Google Search",
"type": "main",
"index": 0
}
]
]
},
"Google Search": {
"main": [
[
{
"node": "Extract & Filter URLs",
"type": "main",
"index": 0
}
]
]
},
"Loop (Amazon)": {
"main": [
[
{
"node": "Pick Best (Amazon)",
"type": "main",
"index": 0
}
],
[
{
"node": "Scrape Amazon",
"type": "main",
"index": 0
}
]
]
},
"Merge Results": {
"main": [
[
{
"node": "Format Context",
"type": "main",
"index": 0
}
]
]
},
"Scrape Amazon": {
"main": [
[
{
"node": "Clean HTML (Amazon)",
"type": "main",
"index": 0
}
]
]
},
"Format Context": {
"main": [
[
{
"node": "AI Extraction Agent",
"type": "main",
"index": 0
}
]
]
},
"Telegram Trigger": {
"main": [
[
{
"node": "Build Queries",
"type": "main",
"index": 0
}
]
]
},
"Pick Best (Other)": {
"main": [
[
{
"node": "Merge Results",
"type": "main",
"index": 1
}
]
]
},
"Clean HTML (Other)": {
"main": [
[
{
"node": "Score Match (Other)",
"type": "main",
"index": 0
}
]
]
},
"Pick Best (Amazon)": {
"main": [
[
{
"node": "Merge Results",
"type": "main",
"index": 0
}
]
]
},
"AI Extraction Agent": {
"main": [
[
{
"node": "Send Reply",
"type": "main",
"index": 0
}
]
]
},
"Clean HTML (Amazon)": {
"main": [
[
{
"node": "Score Match (Amazon)",
"type": "main",
"index": 0
}
]
]
},
"Score Match (Other)": {
"main": [
[
{
"node": "Loop (Other)",
"type": "main",
"index": 0
}
]
]
},
"Score Match (Amazon)": {
"main": [
[
{
"node": "Loop (Amazon)",
"type": "main",
"index": 0
}
]
]
},
"Extract & Filter URLs": {
"main": [
[
{
"node": "Router (Amazon vs. Other)",
"type": "main",
"index": 0
}
]
]
},
"Router (Amazon vs. Other)": {
"main": [
[
{
"node": "Loop (Amazon)",
"type": "main",
"index": 0
}
],
[
{
"node": "Loop (Other)",
"type": "main",
"index": 0
}
]
]
}
}
}
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
We’ve all been there: you want to check if a product is cheaper on Amazon or Jumia, but opening a dozen tabs is a pain. Building a bot to do this usually fails because big e-commerce sites love to block scrapers with CAPTCHAs. This workflow fixes that headache by: Taking a…
Source: https://n8n.io/workflows/12075/ — 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.
Stop watching long videos, start listening to concise summaries.
Telegram, Output Parser Structured, Telegram Trigger +4
Generate AI viral videos with NanoBanana & VEO3, shared on socials via Blotato 2. Uses @blotato/n8n-nodes-blotato, googleSheets, lmChatOpenAi, toolThink. Event-driven trigger; 94 nodes.
@Blotato/N8N Nodes Blotato, Google Sheets, OpenAI Chat +9
RAG CHATBOT Main. Uses telegram, telegramTrigger, lmChatOpenAi, n8n-nodes-mcp. Event-driven trigger; 87 nodes.
Telegram, Telegram Trigger, OpenAI Chat +8
Digital marketers, content creators, social media managers, and businesses who want to use AI marketing automation for YouTube Shorts without spending hours on production. This AI workflow helps anyon
OpenAI, HTTP Request, OpenAI Chat +7
Transform your salon/service business with this streamlined Telegram automation system featuring Claude integration, zero-setup database management, and intelligent conversation handling. Claude MCP I
Redis, Agent Tool, Google Calendar +10