🌲 Agent IA pour l'audit de rapports de durabilité en utilisant Gmail et GPT-40
Ceci est unOther, AIworkflow d'automatisation du domainecontenant 14 nœuds.Utilise principalement des nœuds comme If, Code, Gmail, GmailTrigger, Agent, combinant la technologie d'intelligence artificielle pour une automatisation intelligente. 🌲 utilisationGmailetGPT-40进行可持续发展rapport审计deAIagent
- •Compte Google et informations d'identification Gmail API
- •Clé API OpenAI
Nœuds utilisés (14)
Catégorie
{
"meta": {
"instanceId": "="
},
"nodes": [
{
"id": "a2d54127-d1d1-44d2-859e-b89e2e6c3b4d",
"name": "If",
"type": "n8n-nodes-base.if",
"position": [
260,
260
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "=",
"operator": {
"type": "string",
"operation": "contains"
},
"leftValue": "={{ $json.subject }}",
"rightValue": "CSRD Reporting"
}
]
}
},
"typeVersion": 2.2
},
{
"id": "6a664023-ea8c-4973-b3ac-13a9e0664a58",
"name": "Vérifier le format",
"type": "n8n-nodes-base.code",
"position": [
960,
260
],
"parameters": {
"jsCode": "const content = $input.first().json.xhtml_content;\n\n// Helper to extract tags\nfunction extractTags(tagName) {\n const regex = new RegExp(`<${tagName}[^>]*>(.*?)<\\\\/${tagName}>`, 'gs');\n let matches = [];\n let match;\n while ((match = regex.exec(content)) !== null) {\n matches.push(match[1].trim());\n }\n return matches;\n}\n\n// Basic Tests\nconst headerPresent = /<ix:header>/i.test(content);\nconst governanceTag = /<ix:nonNumeric[^>]*name=\"esrs:SustainabilityGovernance\"/i.test(content);\nconst strategyTag = /<ix:nonNumeric[^>]*name=\"esrs:StrategySustainability\"/i.test(content);\n\n// KPI Tags\nconst kpiTags = [\"esrs:GHGScope1Emissions\", \"esrs:GHGScope2Emissions\", \"esrs:GHGScope3Emissions\"];\nconst kpiMatches = kpiTags.filter(tag => content.includes(tag));\n\n// Check for empty tags\nconst emptyNonNumeric = (content.match(/<ix:nonNumeric[^>]*>\\s*<\\/ix:nonNumeric>/g) || []).length;\n\n// Check duplicate text\nconst nonNumericValues = extractTags(\"ix:nonNumeric\");\nconst duplicates = [...new Set(nonNumericValues.filter((v, i, arr) => arr.indexOf(v) !== i))];\n\n// Final Result\nreturn [\n {\n json: {\n audit_results:{\n total_nonNumeric_tags: nonNumericValues.length,\n total_kpis_found: kpiMatches.length,\n empty_disclosures: emptyNonNumeric,\n governance_check: governanceTag ? \"PASS\" : \"MISSING\",\n strategy_check: strategyTag ? \"PASS\" : \"MISSING\",\n header_check: headerPresent ? \"PASS\" : \"MISSING\",\n duplicate_disclosures: duplicates,\n }\n\n }\n }\n];\n"
},
"typeVersion": 2
},
{
"id": "a16b613e-a7c2-4079-9ff9-46c485019ca3",
"name": "Agent IA",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
1240,
260
],
"parameters": {
"text": "=Generate an email to the sustainability team summarizing this CSRD XHTML report audit:\n\n{{JSON.stringify($json.audit_results, null, 2)}}\n\nReturn the output in the following JSON format:\n\n{\n \"subject\": \"...\",\n \"body\": \"...\"\n}",
"options": {
"systemMessage": "=You are LogiGreen CSRD Audit Bot, an ESG compliance assistant writing professional email summaries based on automated XHTML audits for CSRD compliance. Your role is to translate JSON audit results into clear, actionable summaries. Keep a neutral, helpful tone and highlight any risks or missing disclosures. Include key findings and suggest next steps if needed.\n\nWrite emails in plain English with no markdown (avoid **, #, ##, etc.).\nFormat your message with proper line breaks for readability.\nAlways sign with:\nBest regards,\nLogiGreen CSRD Audit Bot"
},
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 1.8
},
{
"id": "3dcbaf39-58be-465e-9ec2-0b2a9a8c8fe3",
"name": "OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
1200,
420
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4o-mini"
},
"options": {}
},
"typeVersion": 1.2
},
{
"id": "6e742627-f315-4ee2-be1b-023b38103978",
"name": "Structured Output Parser",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
1500,
440
],
"parameters": {
"jsonSchemaExample": "{\n \"subject\": \"CSRD XHTML Report Audit – Key Findings and Next Steps\",\n \"body\": \"Content of the email\"\n}"
},
"typeVersion": 1.2
},
{
"id": "994e5b98-5bda-4a4f-a3eb-cb521de9d88a",
"name": "Répondre",
"type": "n8n-nodes-base.gmail",
"position": [
1620,
260
],
"webhookId": "=",
"parameters": {
"message": "={{ $json.output.body }}",
"options": {},
"emailType": "text",
"messageId": "={{ $('Gmail').item.json.id }}",
"operation": "reply"
},
"notesInFlow": true,
"typeVersion": 2.1
},
{
"id": "8a7fbdcb-2197-437e-b3ba-126c7942ba4d",
"name": "Extraire le HTML",
"type": "n8n-nodes-base.code",
"position": [
800,
260
],
"parameters": {
"jsCode": "return [\n {\n json: {\n xhtml_content:$input.first().json.data \n }\n }\n];\n"
},
"typeVersion": 2
},
{
"id": "90f271b9-4b8b-49ef-90cc-d10d8e22a203",
"name": "Note adhésive1",
"type": "n8n-nodes-base.stickyNote",
"position": [
20,
-140
],
"parameters": {
"color": 7,
"width": 380,
"height": 680,
"content": "### 1. Workflow Trigger with Gmail Trigger\nThe workflow is triggered by a new email received in your Gmail mailbox. \nIf the subject includes the string \"CSRD Reporting\" we proceed, if not we do nothing.\n\n#### How to setup?\n- **Gmail Trigger Node:** set up your Gmail API credentials\n[Learn more about the Gmail Trigger Node](https://docs.n8n.io/integrations/builtin/trigger-nodes/n8n-nodes-base.gmailtrigger)\n"
},
"typeVersion": 1
},
{
"id": "803a758c-fba4-4f48-818b-1272c4509e81",
"name": "Note adhésive",
"type": "n8n-nodes-base.stickyNote",
"position": [
440,
-140
],
"parameters": {
"color": 7,
"width": 640,
"height": 680,
"content": "### 2. Extract and Process the xHTML report\nThis block extract the attachment file from the email, process the xHTML and perform the audit of the content.\n\n#### How to setup?\n- **Gmail Node:** set up your Gmail API credentials\n[Learn more about the Gmail Trigger Node](https://docs.n8n.io/integrations/builtin/trigger-nodes/n8n-nodes-base.gmailtrigger)\n"
},
"typeVersion": 1
},
{
"id": "0b72f7d8-23ce-4243-b2e5-e3ff5c7f163e",
"name": "Note adhésive2",
"type": "n8n-nodes-base.stickyNote",
"position": [
1120,
-140
],
"parameters": {
"color": 7,
"width": 640,
"height": 680,
"content": "### 3. AI Agent write and sends an audit report to the send\nThis summarize the results of the analysis in an email sent as a reply to the sender.\n\n#### How to setup?\n- **Gmail Node:** set up your Gmail API credentials\n[Learn more about the Gmail Trigger Node](https://docs.n8n.io/integrations/builtin/trigger-nodes/n8n-nodes-base.gmailtrigger)\n- **AI Agent with the Chat Model**:\n 1. Add a **chat model** with the required credentials *(Example: Open AI 4o-mini)*\n 2. Adapt the system prompt to the format of emails you want to send\n [Learn more about the AI Agent Node](https://docs.n8n.io/integrations/builtin/cluster-nodes/root-nodes/n8n-nodes-langchain.agent)\n"
},
"typeVersion": 1
},
{
"id": "18103fec-6761-4604-872e-dab251211ba0",
"name": "HTML à partir de binaire",
"type": "n8n-nodes-base.extractFromFile",
"position": [
660,
260
],
"parameters": {
"options": {},
"operation": "text",
"binaryPropertyName": "attachment_0"
},
"notesInFlow": true,
"typeVersion": 1
},
{
"id": "5c31c49d-2324-4d08-a5b5-309925266517",
"name": "Déclencheur Email",
"type": "n8n-nodes-base.gmailTrigger",
"position": [
40,
260
],
"parameters": {
"simple": false,
"filters": {},
"options": {},
"pollTimes": {
"item": [
{
"mode": "everyMinute"
}
]
}
},
"notesInFlow": true,
"typeVersion": 1.2
},
{
"id": "bacbd57d-af9b-49c8-82ae-c74aa2898fc8",
"name": "Télécharger la pièce jointe",
"type": "n8n-nodes-base.gmail",
"position": [
480,
260
],
"webhookId": "=",
"parameters": {
"simple": false,
"options": {
"downloadAttachments": true
},
"messageId": "={{ $json.id }}",
"operation": "get"
},
"notesInFlow": true,
"typeVersion": 2.1
},
{
"id": "af087293-0c3c-4c96-9523-ddb9ed238e00",
"name": "Note adhésive3",
"type": "n8n-nodes-base.stickyNote",
"position": [
1780,
-140
],
"parameters": {
"width": 780,
"height": 540,
"content": "### 4. Do you need more details?\nFind a step-by-step guide in this tutorial\n\n[🎥 Watch My Tutorial](https://www.youtube.com/watch?v=npeJZv5U7og)"
},
"typeVersion": 1
}
],
"pinData": {},
"connections": {
"a16b613e-a7c2-4079-9ff9-46c485019ca3": {
"main": [
[
{
"node": "994e5b98-5bda-4a4f-a3eb-cb521de9d88a",
"type": "main",
"index": 0
}
]
]
},
"5c31c49d-2324-4d08-a5b5-309925266517": {
"main": [
[
{
"node": "a2d54127-d1d1-44d2-859e-b89e2e6c3b4d",
"type": "main",
"index": 0
}
]
]
},
"6a664023-ea8c-4973-b3ac-13a9e0664a58": {
"main": [
[
{
"node": "a16b613e-a7c2-4079-9ff9-46c485019ca3",
"type": "main",
"index": 0
}
]
]
},
"8a7fbdcb-2197-437e-b3ba-126c7942ba4d": {
"main": [
[
{
"node": "6a664023-ea8c-4973-b3ac-13a9e0664a58",
"type": "main",
"index": 0
}
]
]
},
"18103fec-6761-4604-872e-dab251211ba0": {
"main": [
[
{
"node": "8a7fbdcb-2197-437e-b3ba-126c7942ba4d",
"type": "main",
"index": 0
}
]
]
},
"3dcbaf39-58be-465e-9ec2-0b2a9a8c8fe3": {
"ai_languageModel": [
[
{
"node": "a16b613e-a7c2-4079-9ff9-46c485019ca3",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"6e742627-f315-4ee2-be1b-023b38103978": {
"ai_outputParser": [
[
{
"node": "a16b613e-a7c2-4079-9ff9-46c485019ca3",
"type": "ai_outputParser",
"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 ?
Intermédiaire - Autres, Intelligence Artificielle
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
Samir Saci
@samirsaciAutomation, AI and Analytics for Supply Chain & Business Optimization Helping businesses streamline operations using n8n, AI agents, and data science to enhance efficiency and sustainability. Linkedin: www.linkedin.com/in/samir-saci
Partager ce workflow