This workflow follows the Agent → Groq 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 →
{
"nodes": [
{
"parameters": {
"httpMethod": "POST",
"path": "5d37cc73-0a60-4da7-b0df-a330b4b638a1",
"responseMode": "responseNode",
"options": {}
},
"type": "n8n-nodes-base.webhook",
"typeVersion": 2.1,
"position": [
-160,
-252
],
"id": "fa15b370-b80b-4033-8dd9-83b4779795e5",
"name": "Webhook"
},
{
"parameters": {
"promptType": "define",
"text": "={{ $json.body.text }}",
"options": {
"systemMessage": "You are an AI Claim Extraction Agent.\n\nYour task is to extract only verifiable factual claims from the provided text.\n\nFocus ONLY on:\n- statistics\n- percentages\n- dates\n- company facts\n- technical claims\n- financial numbers\n- measurable statements\n\nIgnore:\n- opinions\n- marketing language\n- generic descriptions\n- motivational statements\n- non-verifiable text\n\nReturn ONLY valid JSON in this format:\n\n{\n \"claims\": [\n {\n \"claim\": \"...\"\n }\n ]\n}\n\nDo not include explanations.\nDo not include markdown.\nDo not include extra text."
}
},
"type": "@n8n/n8n-nodes-langchain.agent",
"typeVersion": 3.1,
"position": [
64,
-252
],
"id": "0567988b-56e4-4234-a36e-16bc2d7d3644",
"name": "AI Agent1"
},
{
"parameters": {
"model": "openai/gpt-oss-120b",
"options": {
"temperature": 0
}
},
"type": "@n8n/n8n-nodes-langchain.lmChatGroq",
"typeVersion": 1,
"position": [
136,
-28
],
"id": "8b7e0711-d4a4-4d04-a6d0-c97a4b1c1b39",
"name": "Groq Chat Model1"
},
{
"parameters": {
"jsCode": "const parsed = JSON.parse($json.output);\n\nreturn parsed.claims.map(claim => ({\n json: {\n claim: claim.claim\n }\n}));"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
416,
-252
],
"id": "a04b552b-68e1-4912-bd96-4b66d2078e1e",
"name": "Code in JavaScript"
},
{
"parameters": {
"query": "={{ $json.claim }}",
"options": {
"max_results": 2
}
},
"type": "@tavily/n8n-nodes-tavily.tavily",
"typeVersion": 1,
"position": [
864,
-252
],
"id": "a6b10145-f534-468f-9e91-9c32e1fc9ab9",
"name": "Search"
},
{
"parameters": {
"promptType": "define",
"text": "={{ $json.claim }}\n{{ $json.evidence }}",
"options": {
"systemMessage": "You are an AI Fact-Checking Agent.\n\nYour task is to verify claims using ONLY the provided evidence.\n\nVerification Rules:\n- Compare the claim carefully with the evidence\n- Do not use outside knowledge\n- Use ONLY the provided sources\n- Prefer reputable sources over social media posts\n- If evidence is weak or incomplete, lower confidence\n\nClassification Rules:\n- VERIFIED \u2192 evidence directly supports the claim\n- INACCURATE \u2192 partially correct, misleading, outdated, or numerically incorrect\n- FALSE \u2192 contradicted by evidence or unsupported\n\nConfidence Rules:\n- 90\u2013100 \u2192 strong direct evidence from reliable sources\n- 70\u201389 \u2192 mostly supported but minor uncertainty\n- 40\u201369 \u2192 weak or incomplete evidence\n- 0\u201339 \u2192 unsupported or unreliable evidence\n\nAdditional Rules:\n- Keep explanations concise\n- Mention the most relevant source/domain in the explanation\n- If the claim is incorrect, provide the corrected fact\n- Return ONLY raw valid JSON\n- Do not include markdown\n- Do not include code blocks\n- Do not say \"Here is the response\"\n- Do not include extra text before or after JSON\n\nResponse format:\n{\n \"claim\": \"\",\n \"status\": \"\",\n \"confidence\": 0,\n \"correct_fact\": \"\",\n \"source_url\": \"\",\n \"explanation\": \"\"\n}"
}
},
"type": "@n8n/n8n-nodes-langchain.agent",
"typeVersion": 3.1,
"position": [
1312,
-180
],
"id": "190339e2-b622-425e-9922-11a05c1282b2",
"name": "AI Agent"
},
{
"parameters": {
"model": "meta-llama/llama-4-scout-17b-16e-instruct",
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatGroq",
"typeVersion": 1,
"position": [
1384,
44
],
"id": "eeca88a4-a5f0-40b9-bc80-5625edc606e9",
"name": "Groq Chat Model"
},
{
"parameters": {
"jsCode": "const results = [];\n\nfor (const item of $input.all()) {\n let output = item.json.output;\n\n const match = output.match(/\\{[\\s\\S]*\\}/);\n\n if (!match) {\n results.push({\n error: \"No valid JSON found\",\n raw_output: output\n });\n continue;\n }\n\n try {\n const parsed = JSON.parse(match[0]);\n results.push(parsed);\n } catch (err) {\n results.push({\n error: err.message,\n raw_output: output\n });\n }\n}\n\nreturn results.map(r => ({\n json: r\n}));"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
864,
-496
],
"id": "5dff2990-6bbd-4b1d-bd51-1ed0cf03504a",
"name": "Code in JavaScript2"
},
{
"parameters": {
"jsCode": "const results = $input.all().map(item => item.json);\n\nconst summary = {\n total_claims: results.length,\n verified: results.filter(r => r.status === \"VERIFIED\").length,\n inaccurate: results.filter(r => r.status === \"INACCURATE\").length,\n false: results.filter(r => r.status === \"FALSE\").length\n};\n\nreturn [\n {\n json: {\n success: true,\n summary,\n results\n }\n }\n];"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
1088,
-496
],
"id": "da06160c-4725-4033-b548-b54b979dd7b8",
"name": "Code in JavaScript3",
"executeOnce": false,
"alwaysOutputData": false
},
{
"parameters": {
"respondWith": "json",
"responseBody": "={{ $json }}",
"options": {}
},
"type": "n8n-nodes-base.respondToWebhook",
"typeVersion": 1.5,
"position": [
1376,
-496
],
"id": "ca75fa9a-cfea-4940-b2f9-2f635cd3ccd9",
"name": "Respond to Webhook",
"alwaysOutputData": false,
"executeOnce": false
},
{
"parameters": {
"jsCode": "const evidence = $json.results\n ?.map((r) => {\n\n const domain = r.url\n .replace(/^https?:\\/\\//, \"\")\n .replace(/^www\\./, \"\")\n .split(\"/\")[0];\n\n const cleanedContent = r.content\n .replace(/[#*_`>-]/g, \"\")\n .replace(/\\s+/g, \" \")\n .trim()\n .slice(0, 300);\n\n return `\nSource: ${domain}\n\nEvidence:\n${cleanedContent}\n\nURL:\n${r.url}\n`;\n })\n .join(\"\\n\\n\");\n\nreturn [\n {\n json: {\n claim: $json.query,\n evidence\n }\n }\n];"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
1088,
-252
],
"id": "4f23a4b8-85d8-4097-94b0-9cea2ea23d1a",
"name": "Code in JavaScript1"
},
{
"parameters": {
"options": {}
},
"type": "n8n-nodes-base.splitInBatches",
"typeVersion": 3,
"position": [
640,
-252
],
"id": "0a7f18a3-0e37-42e6-8a06-94d3a57c8192",
"name": "Loop Over Items"
}
],
"connections": {
"Webhook": {
"main": [
[
{
"node": "AI Agent1",
"type": "main",
"index": 0
}
]
]
},
"AI Agent1": {
"main": [
[
{
"node": "Code in JavaScript",
"type": "main",
"index": 0
}
]
]
},
"Groq Chat Model1": {
"ai_languageModel": [
[
{
"node": "AI Agent1",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Code in JavaScript": {
"main": [
[
{
"node": "Loop Over Items",
"type": "main",
"index": 0
}
]
]
},
"Search": {
"main": [
[
{
"node": "Code in JavaScript1",
"type": "main",
"index": 0
}
]
]
},
"AI Agent": {
"main": [
[
{
"node": "Loop Over Items",
"type": "main",
"index": 0
}
]
]
},
"Groq Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Code in JavaScript2": {
"main": [
[
{
"node": "Code in JavaScript3",
"type": "main",
"index": 0
}
]
]
},
"Code in JavaScript3": {
"main": [
[
{
"node": "Respond to Webhook",
"type": "main",
"index": 0
}
]
]
},
"Code in JavaScript1": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
},
"Loop Over Items": {
"main": [
[
{
"node": "Code in JavaScript2",
"type": "main",
"index": 0
}
],
[
{
"node": "Search",
"type": "main",
"index": 0
}
]
]
}
},
"meta": {
"templateCredsSetupCompleted": true
}
}
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
Fact-Check-Workflow. Uses agent, lmChatGroq, @tavily/n8n-nodes-tavily. Webhook trigger; 12 nodes.
Source: https://github.com/ABHICHAUHAN1480/cog-culture_assessment/blob/ad0781404507176a762b8069c4150e32c6169af3/n8n/fact-check-workflow.json — 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.
Hi! I'm Bruno — I’ve been building AI-powered workflows for n8n and Make for 2+ years, focused on smart automation and real conversational agents.
OpenAI, Agent, OpenAI Chat +5
Know that feeling when a "low priority" ticket turns into a production fire? Or when your on-call rotation starts showing signs of serious burnout from alert overload?
Google Sheets, Agent, Google Gemini Chat +2
B.IA-v.0.3.0. Uses agent, lmChatGroq, n8n-nodes-evolution-api, httpRequest. Webhook trigger; 37 nodes.
Agent, Groq Chat, N8N Nodes Evolution Api +7
Whatsapp Lead Agent. Uses httpRequest, hunter, @tavily/n8n-nodes-tavily, @mendable/n8n-nodes-firecrawl. Webhook trigger; 35 nodes.
HTTP Request, Hunter, @Tavily/N8N Nodes Tavily +11
<< What Does This Flow Do? >>
@Tavily/N8N Nodes Tavily, Agent, Groq Chat +2