Adam Janes This workflow corresponds to n8n.io template #4712 — we link there as the canonical source.
This workflow follows the Chainllm → Google Drive 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": [
{
"id": "5c25edcc-987c-4b60-b472-6c476dc866c6",
"name": "When clicking \u2018Test workflow\u2019",
"type": "n8n-nodes-base.manualTrigger",
"position": [
-580,
440
],
"parameters": {},
"typeVersion": 1
},
{
"id": "60521a1e-3f00-4dd6-9651-a5f9ae1659ab",
"name": "Loop Over Items",
"type": "n8n-nodes-base.splitInBatches",
"position": [
-20,
-180
],
"parameters": {
"options": {
"reset": false
},
"batchSize": "=1"
},
"typeVersion": 3
},
{
"id": "8e0fd8c4-51a1-44fb-8524-034df4fd4fe5",
"name": "Wait",
"type": "n8n-nodes-base.wait",
"position": [
1520,
340
],
"parameters": {
"amount": 0.5
},
"typeVersion": 1.1
},
{
"id": "0e457acc-5047-43ed-977f-f08714498ffc",
"name": "OpenRouter Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenRouter",
"position": [
740,
320
],
"parameters": {
"model": "openai/gpt-4.1",
"options": {}
},
"credentials": {
"openRouterApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "46ce90ac-6b58-4770-a6bd-48fccefc8713",
"name": "Extract from File",
"type": "n8n-nodes-base.extractFromFile",
"position": [
500,
20
],
"parameters": {
"options": {},
"operation": "pdf"
},
"typeVersion": 1
},
{
"id": "9e76d679-6f4b-419b-af36-771aba5b9e98",
"name": "Google Drive",
"type": "n8n-nodes-base.googleDrive",
"onError": "continueRegularOutput",
"position": [
120,
20
],
"parameters": {
"fileId": {
"__rl": true,
"mode": "id",
"value": "={{ $json[\"URL\"].match(/[-\\w]{25,}/)[0] }}"
},
"options": {},
"operation": "download"
},
"credentials": {
"googleDriveOAuth2Api": {
"name": "<your credential>"
}
},
"typeVersion": 3,
"alwaysOutputData": true
},
{
"id": "8f2fc75f-ea5d-4700-b58b-b5f5e3e360e5",
"name": "Get Tests",
"type": "n8n-nodes-base.googleSheets",
"position": [
-420,
440
],
"parameters": {
"options": {},
"sheetName": {
"__rl": true,
"mode": "list",
"value": "gid=0",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/10l_gMtPsge00eTTltGrgvAo54qhh3_twEDsETrQLAGU/edit#gid=0",
"cachedResultName": "Tests"
},
"documentId": {
"__rl": true,
"mode": "list",
"value": "10l_gMtPsge00eTTltGrgvAo54qhh3_twEDsETrQLAGU",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/10l_gMtPsge00eTTltGrgvAo54qhh3_twEDsETrQLAGU/edit?usp=drivesdk",
"cachedResultName": "LLM Judge"
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"name": "<your credential>"
}
},
"typeVersion": 4.5
},
{
"id": "49b0cc43-30d3-4331-a608-4f51c9ac0efd",
"name": "Structured Output Parser",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
900,
320
],
"parameters": {
"jsonSchemaExample": "{\n \"reasoning\": \"The Assistant fabricated a $1 million figure and a 12-month provision that are not found in the source. This breaches factual correctness and completeness. The output would mislead business stakeholders if used without correction.\",\n \"decision\": \"Fail\"\n}"
},
"typeVersion": 1.2
},
{
"id": "cf3cce00-33c4-4bcd-a2a2-75d7e49439d3",
"name": "Update Results",
"type": "n8n-nodes-base.googleSheets",
"position": [
1220,
140
],
"parameters": {
"columns": {
"value": {
"ID": "={{ $('Loop Over Items').item.json[\"ID\"] }}",
"URL": "={{ $('Loop Over Items').item.json[\"URL\"] }}",
"Input": "={{ $('Loop Over Items').item.json[\"Input\"] }}",
"Output": "={{ $('Loop Over Items').item.json[\"Output\"] }}",
"Decision": "={{ $json.output.decision }}",
"Test No.": "={{ $('Loop Over Items').item.json[\"Test No.\"] }}",
"Reasoning": "={{ $json.output.reasoning }}",
"AI Platform": "={{ $('Loop Over Items').item.json[\"AI Platform\"] }}",
"Relevant Source Reference": "={{ $('Loop Over Items').item.json[\"Relevant Source Reference\"] }}"
},
"schema": [
{
"id": "ID",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "ID",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "Test No.",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "Test No.",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "AI Platform",
"type": "string",
"display": true,
"required": false,
"displayName": "AI Platform",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "Relevant Source Reference",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "Relevant Source Reference",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "URL",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "URL",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "Input",
"type": "string",
"display": true,
"required": false,
"displayName": "Input",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "Output",
"type": "string",
"display": true,
"required": false,
"displayName": "Output",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "Decision",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "Decision",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "Reasoning",
"type": "string",
"display": true,
"required": false,
"displayName": "Reasoning",
"defaultMatch": false,
"canBeUsedToMatch": true
}
],
"mappingMode": "defineBelow",
"matchingColumns": [],
"attemptToConvertTypes": false,
"convertFieldsToString": false
},
"options": {},
"operation": "append",
"sheetName": {
"__rl": true,
"mode": "list",
"value": 537199982,
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/10l_gMtPsge00eTTltGrgvAo54qhh3_twEDsETrQLAGU/edit#gid=537199982",
"cachedResultName": "Results"
},
"documentId": {
"__rl": true,
"mode": "list",
"value": "10l_gMtPsge00eTTltGrgvAo54qhh3_twEDsETrQLAGU",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/10l_gMtPsge00eTTltGrgvAo54qhh3_twEDsETrQLAGU/edit?usp=drivesdk",
"cachedResultName": "Info Extraction Tasks (LLM Judge)"
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"name": "<your credential>"
}
},
"typeVersion": 4.5
},
{
"id": "80a9da50-93df-476a-a39f-ffece642a934",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-540,
220
],
"parameters": {
"color": 6,
"width": 360,
"height": 180,
"content": "## 1. Fetch test cases\nWe start by grabbing our list of test cases stored in a Google Sheet [here](https://docs.google.com/spreadsheets/d/10l_gMtPsge00eTTltGrgvAo54qhh3_twEDsETrQLAGU/edit?usp=sharing).\n\nWe only want the rows that connect to a PDF document, as DOCX downloads will need to be handled separately."
},
"typeVersion": 1
},
{
"id": "2bba0cc9-460d-48c3-bcc6-a79c2b00b23e",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-80,
-400
],
"parameters": {
"color": 6,
"width": 260,
"height": 200,
"content": "## 2. Loop through our data\nWe use a loop here to visually see our workflow going through each item one-by-one."
},
"typeVersion": 1
},
{
"id": "7ccb6426-e79e-4cdc-ac5f-59c022f8978d",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
200,
-180
],
"parameters": {
"color": 6,
"width": 360,
"content": "## 3. Grab the PDF as text\nWe download the PDF from the Google Drive link in the Google Sheet, extracting the file as text for the next step. We filter out any files that do not return data."
},
"typeVersion": 1
},
{
"id": "6dc17e29-ea6c-460b-b0e0-99cc41977891",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
700,
-180
],
"parameters": {
"color": 6,
"width": 360,
"height": 280,
"content": "## 4. Judge LLM outputs\nOur prompt judges the LLM input/output and decides if the LLM passed the test. We also ask for a reason why the judge made its decision, which we can use to refine our eval later.\n\nWe're using OpenRouter here, which lets us easily tweak which LLM we want to use.\n\nThe output parser makes sure that the output is in JSON format, making the data easy to parse in the next step."
},
"typeVersion": 1
},
{
"id": "1fd001a6-e67a-406c-b81c-c0a454ecd73e",
"name": "LLM Judge",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
740,
140
],
"parameters": {
"text": "=INPUT:\n\n{\n \"task\": {{ $('Loop Over Items').item.json['Input '] }},\n \"source\": {{ $json.text }},\n \"output\": {{ $('Loop Over Items').item.json['Output '] }}\n}\n\nOUTPUT:",
"messages": {
"messageValues": [
{
"message": "=You are an expert legal evaluator. Your role is to decide if each AI output for an information extraction task is fully accurate.\nInstructions:\nRead the extraction task and the source material carefully.\n\n\nExtract the correct answer for yourself, based only on the source material. (Do not include your answer in your output.)\n\n\nFor each AI output provided, do the following:\n\n\nCompare it against your own extraction.\n\n\nJudge \u201caccuracy\u201d as follows:\n\n\nPass: The output is complete (all required info present), faithful (statements fully supported by the source), and contains no hallucinated or omitted content.\n\n\nFail: Any omission, misstatement, unsupported addition, or distortion.\n\n\nIgnore differences in style, order, or formatting. Judge on factual substance only.\n\n\nFor each output, give:\n\n\nAccuracy: Pass or Fail\n\n\nReasoning: In 1-2 sentences, cite the specific content that was included, missed, or misstated. Be as objective and specific as possible.\n\n\nOutput Format:\n\nGive your answer in a JSON format. Don't include any other text than the JSON.\n\n{\n \"decision\": \"Pass/Fail\",\n \"reasoning\": \"[Concise explanation referencing the substance of the source]\"\n}\n\nGuidelines:\nUse your internal extraction for comparison, but do not include it in your output.\n\n\nFocus only on whether the output accurately and fully reflects the required information from the source.\n\n\nDo not consider writing style or format.\n\n\nJustify your decision with evidence from the source.\n\n\nExamples:\n\nAI Output 1:\n\n{\n \"decision\": \"Pass\",\n \"reasoning\": \"Correctly included the requirements for advance notice and refund as set out in the source, with no omissions.\"\n}\n\n\nAI Output 2:\n\n{\n \"decision\": \"Fail\",\n \"reasoning\": \"Missed the requirement that refunds are provided only as credit, not cash, making the output incomplete.\"\n}"
}
]
},
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 1.4
},
{
"id": "2e94c252-9973-4915-ab21-838681e7ea5c",
"name": "Sticky Note4",
"type": "n8n-nodes-base.stickyNote",
"position": [
1160,
-80
],
"parameters": {
"color": 6,
"width": 260,
"height": 180,
"content": "## 5. Update results\nWe create a new row in our output sheet, containing our original data together with the judge decision/reasoning."
},
"typeVersion": 1
},
{
"id": "05d7a5c0-3ad5-4ec1-ae3f-f533e16f8027",
"name": "Sticky Note5",
"type": "n8n-nodes-base.stickyNote",
"position": [
1460,
140
],
"parameters": {
"color": 6,
"width": 220,
"height": 180,
"content": "## 6. Pause between runs\nWe wait half a second, so that we don't hit the OpenAI API rate limit and get an error."
},
"typeVersion": 1
},
{
"id": "099b10f3-c9cf-4fa0-9a07-ccb73ea11542",
"name": "Sticky Note6",
"type": "n8n-nodes-base.stickyNote",
"position": [
-600,
640
],
"parameters": {
"width": 460,
"height": 280,
"content": "## Data format\nOur Tests Sheet contains the following columns:\n- ID: A unique identifier for each row\n- Test No.: The test that the LLM was given\n- AI Platform: The LLM that was given the test.\n- Relevant Source: The file name of the source document that was given to the LLM.\n- URL: The Google Drive URL where the file is stored.\n- Input: The input prompt that the LLM was given.\n- Output: The response that the LLM gave."
},
"typeVersion": 1
},
{
"id": "65532c6d-5638-432f-8f6a-a5a02b1aaf08",
"name": "Is PDF?",
"type": "n8n-nodes-base.if",
"position": [
-260,
440
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "loose"
},
"combinator": "and",
"conditions": [
{
"id": "1609d1f6-2142-4965-8d6b-01cfa53251c4",
"operator": {
"type": "string",
"operation": "contains"
},
"leftValue": "={{ $json['Relevant Source Reference'] }}",
"rightValue": ".pdf"
}
]
},
"looseTypeValidation": true
},
"typeVersion": 2.2
},
{
"id": "17aa2c76-7ea7-41fe-84ed-4e485ef4db5e",
"name": "Is a file?",
"type": "n8n-nodes-base.if",
"position": [
300,
20
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "ad7272d4-0840-4c71-bee4-64faf944e3ca",
"operator": {
"type": "object",
"operation": "notEmpty",
"singleValue": true
},
"leftValue": "={{ $binary }}",
"rightValue": ""
}
]
}
},
"typeVersion": 2.2
}
],
"connections": {
"Wait": {
"main": [
[
{
"node": "Loop Over Items",
"type": "main",
"index": 0
}
]
]
},
"Is PDF?": {
"main": [
[
{
"node": "Loop Over Items",
"type": "main",
"index": 0
}
]
]
},
"Get Tests": {
"main": [
[
{
"node": "Is PDF?",
"type": "main",
"index": 0
}
]
]
},
"LLM Judge": {
"main": [
[
{
"node": "Update Results",
"type": "main",
"index": 0
}
]
]
},
"Is a file?": {
"main": [
[
{
"node": "Extract from File",
"type": "main",
"index": 0
}
],
[
{
"node": "Loop Over Items",
"type": "main",
"index": 0
}
]
]
},
"Google Drive": {
"main": [
[
{
"node": "Is a file?",
"type": "main",
"index": 0
}
]
]
},
"Update Results": {
"main": [
[
{
"node": "Wait",
"type": "main",
"index": 0
}
]
]
},
"Loop Over Items": {
"main": [
[],
[
{
"node": "Google Drive",
"type": "main",
"index": 0
}
]
]
},
"Extract from File": {
"main": [
[
{
"node": "LLM Judge",
"type": "main",
"index": 0
}
]
]
},
"OpenRouter Chat Model": {
"ai_languageModel": [
[
{
"node": "LLM Judge",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Structured Output Parser": {
"ai_outputParser": [
[
{
"node": "LLM Judge",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"When clicking \u2018Test workflow\u2019": {
"main": [
[
{
"node": "Get Tests",
"type": "main",
"index": 0
}
]
]
}
}
}
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.
googleDriveOAuth2ApigoogleSheetsOAuth2ApiopenRouterApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
This workflow demonstrates a simple way to run evals on a set of test cases stored in a Google Sheet.
Source: https://n8n.io/workflows/4712/ — 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.
This workflow demonstrates a simple way to run evals on a set of test cases stored in a Google Sheet.
Google Sheets, Output Parser Structured, Chain Llm +3
This workflow is designed to automate the generation and updating of SEO meta titles and descriptions for WooCommerce products using n8n. It leverages Google Sheets for data input, a FREE language mod
Output Parser Structured, Google Sheets, WooCommerce +2
Splitout Code. Uses lmChatOpenRouter, outputParserStructured, httpRequest, stickyNote. Event-driven trigger; 11 nodes.
OpenRouter Chat, Output Parser Structured, HTTP Request +2
This workflow automates web scraping of Amazon search result pages by retrieving raw HTML, cleaning it to retain only the relevant product elements, and then using an LLM to extract structured product
OpenRouter Chat, Output Parser Structured, HTTP Request +2
This template is for educators, parents, or self-learners who want to gamify their study routines. It turns boring study logs into an interactive RPG math battle using AI.
Chain Llm, OpenRouter Chat, Output Parser Structured +2