This workflow follows the Agent → HTTP Request 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": "testmyagent_functions",
"nodes": [
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "c8938cc3-041f-48c7-9a65-3923b6372a1d",
"name": "conversations",
"value": "={{ $json.output.conversations }}",
"type": "array"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
540,
380
],
"id": "89087716-a06d-45f4-8f0f-23d5667ae0b2",
"name": "Set response"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "0049332b-3ed6-41a9-8b8e-25fe118c394f",
"name": "prompt",
"value": "={{ $json.body.prompt }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
-20,
380
],
"id": "49118a8b-1615-453f-94fd-9aaa9d455082",
"name": "Set variables"
},
{
"parameters": {
"jsonSchemaExample": "{\n \"conversations\": [\n {\n \"user\": \"Oi\",\n \"agent\": \"Ol\u00e1, tudo bem? Sou Adriana da Casa Sertanejo Eletro, como posso ajudar voc\u00ea hoje?\"\n },\n {\n \"user\": \"Estou interessado em comprar um notebook\",\n \"agent\": \"\u00d3timo! Voc\u00ea busca algum modelo espec\u00edfico ou precisa de recomenda\u00e7\u00e3o?\"\n }\n ]\n}"
},
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"typeVersion": 1.2,
"position": [
360,
600
],
"id": "3dd95f11-f1ef-45d4-8a88-95c1c16b7059",
"name": "Structured Output Parser"
},
{
"parameters": {
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"typeVersion": 1,
"position": [
180,
80
],
"id": "38a4b26d-fce6-40e1-8364-14dc78100a2e",
"name": "OpenAI Chat Model",
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"promptType": "define",
"text": "={{ $json.prompt }}",
"hasOutputParser": true,
"options": {
"systemMessage": "=Voc\u00ea \u00e9 um agente especializado em gerar conversas sint\u00e9ticas realistas entre um usu\u00e1rio e um agente de IA. Dado um texto curto ou descri\u00e7\u00e3o da conversa desejada, voc\u00ea deve retornar uma conversa estruturada no formato JSON, alternando as falas entre \\\"user\\\" e \\\"agent\\\".\n\n## Regras:\n- Sempre retorne um objeto JSON com uma \u00fanica chave chamada \\\"conversations\\\", que cont\u00e9m um array de objetos.\n- Cada objeto no array deve possuir exatamente duas chaves: \\\"user\\\" e \\\"agent\\\".\n- \\\"user\\\" cont\u00e9m a fala do usu\u00e1rio, e \\\"agent\\\" cont\u00e9m a fala correspondente do agente de IA.\n- As falas devem ser breves, naturais e coerentes com o contexto fornecido.\n\n## Exemplo 1:\n\nInput:\n\"Ol\u00e1\\nOl\u00e1, em que posso ajudar?\\nQuero reservar uma mesa para dois\\nClaro, qual hor\u00e1rio?\"\n\nOutput:\n```json\n{\n \\\"conversations\\\": [\n {\n \\\"user\\\": \\\"Ol\u00e1\\\",\n \\\"agent\\\": \\\"Ol\u00e1, em que posso ajudar?\\\"\n },\n {\n \\\"user\\\": \\\"Quero reservar uma mesa para dois\\\",\n \\\"agent\\\": \\\"Claro, qual hor\u00e1rio?\\\"\n }\n ]\n}\n```\n\n## Exemplo 2 (one-shot com descri\u00e7\u00e3o):\n\nInput:\n\"conversa entre cliente e vendedor de e-commerce\"\n\nOutput:\n```json\n{\n \\\"conversations\\\": [\n {\n \\\"user\\\": \\\"Oi\\\",\n \\\"agent\\\": \\\"Ol\u00e1, tudo bem? Sou Adriana da Casa Sertanejo Eletro, como posso ajudar voc\u00ea hoje?\\\"\n },\n {\n \\\"user\\\": \\\"Estou interessado em comprar um notebook\\\",\n \\\"agent\\\": \\\"\u00d3timo! Voc\u00ea busca algum modelo espec\u00edfico ou precisa de recomenda\u00e7\u00e3o?\\\"\n }\n ]\n}\n```\n\nAgora gere o JSON para o seguinte texto:\n\nInput:\n\"{{ $json.prompt }}\"\n\nOutput:"
}
},
"type": "@n8n/n8n-nodes-langchain.agent",
"typeVersion": 1.7,
"position": [
180,
380
],
"id": "9062f8bf-e9ca-45ba-8c35-187f78ec6990",
"name": "One-Scenario Generator"
},
{
"parameters": {
"httpMethod": "POST",
"path": "generateScenario",
"responseMode": "responseNode",
"options": {}
},
"type": "n8n-nodes-base.webhook",
"typeVersion": 2,
"position": [
-240,
380
],
"id": "bf1bff97-c4fb-4653-b1ec-f623f8b01463",
"name": "generateScenario"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "0049332b-3ed6-41a9-8b8e-25fe118c394f",
"name": "prompt",
"value": "={{ $json.body.prompt }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
-20,
-140
],
"id": "70cc1232-c703-4d5d-92a7-0414aab1bd63",
"name": "Set variables multi"
},
{
"parameters": {
"httpMethod": "POST",
"path": "generateMultiScenarios",
"responseMode": "responseNode",
"options": {}
},
"type": "n8n-nodes-base.webhook",
"typeVersion": 2,
"position": [
-240,
-140
],
"id": "86b10cac-40eb-465d-ba4c-f3143bc4c9e9",
"name": "generateMultiScenarios"
},
{
"parameters": {
"options": {}
},
"type": "n8n-nodes-base.respondToWebhook",
"typeVersion": 1.1,
"position": [
740,
-140
],
"id": "01e42d2a-eb52-41fc-96d0-1fdd2ccfb210",
"name": "Respond to generateMultiScenarios"
},
{
"parameters": {
"options": {}
},
"type": "n8n-nodes-base.respondToWebhook",
"typeVersion": 1.1,
"position": [
740,
380
],
"id": "6be52130-e7dd-4439-a9c8-3631509c8b76",
"name": "Respond to generateScenario"
},
{
"parameters": {
"promptType": "define",
"text": "={{ $json.prompt }}",
"hasOutputParser": true,
"options": {
"systemMessage": "=Voc\u00ea \u00e9 um agente especializado em gerar cen\u00e1rios m\u00faltiplos de teste para avaliar respostas de agentes de IA. Receber\u00e1 um input textual curto ou uma descri\u00e7\u00e3o breve do tipo de conversa desejada e dever\u00e1 retornar um JSON estruturado com cen\u00e1rios de testes individuais.\n\nCada cen\u00e1rio gerado deve seguir esta estrutura:\n\n```json\n{\n \"scenarios\": [\n {\n \"name\": \"(\u00cdndice). Nome descritivo do cen\u00e1rio\",\n \"folder\": \"(\u00cdndice). Etapa da conversa\",\n \"expected\": \"Descri\u00e7\u00e3o conceitual clara e espec\u00edfica sobre o que \u00e9 esperado na resposta do agente, contemplando poss\u00edveis varia\u00e7\u00f5es aceit\u00e1veis da resposta literal\",\n \"conversations\": [\n {\n \"user\": \"Input espec\u00edfico do usu\u00e1rio nesta etapa\",\n \"agent\": \"Resposta esperada do agente nesta etapa\"\n }\n ]\n }\n ]\n}\n```\n\nRegras para cria\u00e7\u00e3o dos cen\u00e1rios:\n- Cada cen\u00e1rio representa uma \u00fanica intera\u00e7\u00e3o entre usu\u00e1rio e agente.\n- O campo `expected` deve descrever de maneira clara, conceitual e abrangente o comportamento ou a resposta esperada do agente, permitindo varia\u00e7\u00f5es razo\u00e1veis devido \u00e0 natureza aleat\u00f3ria das respostas.\n- O campo `folder` deve indicar qual etapa da conversa o cen\u00e1rio representa, por exemplo:\n - `1. Sauda\u00e7\u00e3o Inicial`, `2. Solicita\u00e7\u00e3o de Informa\u00e7\u00f5es`\n- O campo `name` deve resumir brevemente o objetivo ou contexto do cen\u00e1rio. Ele tamb\u00e9m deve ser hierarquicamente organizado, de acordo com o campo `folder`, por exemplo:\n - `1.1. Sauda\u00e7\u00e3o Inicial Simples`, `2.1. Solicita\u00e7\u00e3o de Informa\u00e7\u00f5es B\u00e1sicas`\n\nEssa organiza\u00e7\u00e3o hier\u00e1rquica ajuda a desenhar a \u00e1rvore de di\u00e1logo esperada e a identificar rapidamente a etapa da conversa que cada cen\u00e1rio representa.\n\n\nExemplos:\n\nInput:\n\"conversa entre usu\u00e1rio e agente para reserva de restaurante\"\n\nOutput:\n```json\n{\n \"scenarios\": [\n {\n \"name\": \"1.1. Sauda\u00e7\u00e3o Inicial Simples\",\n \"folder\": \"1. Sauda\u00e7\u00e3o Inicial\",\n \"expected\": \"O agente deve saudar o usu\u00e1rio e perguntar detalhes necess\u00e1rios para realizar uma reserva, como data, hor\u00e1rio ou n\u00famero de pessoas.\",\n \"conversations\": [\n {\n \"user\": \"Gostaria de fazer uma reserva\",\n \"agent\": \"Claro! Para quando seria a reserva e quantas pessoas participar\u00e3o?\"\n }\n ]\n },\n {\n \"name\": \"2.1. Confirma\u00e7\u00e3o de Reserva Simples\",\n \"folder\": \"2. Confirma\u00e7\u00e3o de Reserva\",\n \"expected\": \"Ap\u00f3s o usu\u00e1rio fornecer detalhes da reserva, o agente deve confirmar claramente as informa\u00e7\u00f5es fornecidas e verificar se est\u00e3o corretas.\",\n \"conversations\": [\n {\n \"user\": \"Para amanh\u00e3 \u00e0s 20h, duas pessoas.\",\n \"agent\": \"Perfeito, confirmando: reserva para duas pessoas amanh\u00e3 \u00e0s 20h, correto?\"\n }\n ]\n }\n ]\n}\n\nOutro exemplo de input:\n\"oi\\nOl\u00e1, como posso lhe ajudar?\\nQuero informa\u00e7\u00f5es sobre o produto X\\nClaro, o que voc\u00ea gostaria de saber?\"\n\nNeste caso, voc\u00ea deve gerar cen\u00e1rios de teste para cada intera\u00e7\u00e3o da conversa, com descri\u00e7\u00f5es claras e objetivas do comportamento esperado do agente em cada etapa.\n\nAgora gere o JSON para o seguinte texto:\n\nInput:\n\"{{$json.prompt}}\"\n\nOutput:"
}
},
"type": "@n8n/n8n-nodes-langchain.agent",
"typeVersion": 1.7,
"position": [
180,
-140
],
"id": "2d9397fc-a861-40f1-8f68-1260a418efab",
"name": "Multi-Scenario Generator"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "c8938cc3-041f-48c7-9a65-3923b6372a1d",
"name": "scenarios",
"value": "={{ $json.output.scenarios }}",
"type": "array"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
540,
-140
],
"id": "070a8798-3232-419f-84ec-7a8b59ed2f3b",
"name": "Set response multi"
},
{
"parameters": {},
"type": "n8n-nodes-base.noOp",
"typeVersion": 1,
"position": [
1080,
-40
],
"id": "04a267ef-31f1-4db0-90d3-2e83170041ba",
"name": "End"
},
{
"parameters": {
"jsonSchemaExample": "{\n \"scenarios\": [\n {\n \"name\": \"Pedido de Informa\u00e7\u00f5es sobre Produto\",\n \"folder\": \"informacoes_produto\",\n \"expected\": \"O agente deve fornecer informa\u00e7\u00f5es detalhadas sobre o produto.\",\n \"conversations\": [\n {\n \"user\": \"Quais s\u00e3o as especifica\u00e7\u00f5es do smartphone modelo X?\",\n \"agent\": \"O smartphone modelo X possui uma tela de 6.5 polegadas. Ele est\u00e1 dispon\u00edvel por R$ 2.500,00.\"\n }\n ]\n },\n {\n \"nome\": \"Suporte T\u00e9cnico para Conex\u00e3o \u00e0 Internet\",\n \"folder\": \"suporte_internet\",\n \"expected\": \"O agente deve guiar o usu\u00e1rio por um processo de solu\u00e7\u00e3o de problemas.\",\n \"conversations\": [\n {\n \"user\": \"Minha internet caiu, o que posso fazer?\",\n \"agent\": \"Tente reiniciar seu roteador e modem. Se o problema persistir, verifique se h\u00e1 cabos soltos.\"\n }\n ]\n }\n ]\n}\n"
},
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"typeVersion": 1.2,
"position": [
360,
80
],
"id": "bdf2240c-46dd-4add-b5a7-a89ad9557f1f",
"name": "Structured Output Parser multi"
},
{
"parameters": {
"modelId": {
"__rl": true,
"value": "gpt-4o-mini-2024-07-18",
"mode": "list",
"cachedResultName": "GPT-4O-MINI-2024-07-18"
},
"messages": {
"values": [
{
"content": "You are an impartial judge tasked with evaluating the relevance of responses. For each query, compare the \u201cAgent Response\u201d with the \u201cExpected Response\u201d and determine if the agent\u2019s response is relevant or irrelevant. Provide your judgment as two structured values:\n1. Relevance: \u201cRelevant\u201d or \u201cIrrelevant.\u201d\n2. Explanation: A brief justification for your judgment.\n\nFormat:\nQuery: [Insert query here]\nAgent Response: [Insert agent response here]\nExpected Response: [Insert expected response here]\n\nYour Evaluation (as JSON string):\n- \"relevance\": \"[relevant/irrelevant]\"\n- \"explanation\": \"[Short string explaining the judgment]\"\n\nExample:\n\nQuery: What is the capital of France?\nAgent Response: Paris.\nExpected Response: The capital of France is Paris.\n\nYour Evaluation:\n`{{\"relevance\": \"relevant\", \"explanation\": \"The agent\u2019s response correctly identifies the capital of France as Paris.\"}}`\n\nIf the agent\u2019s response does not align with the expected response or fails to address the query accurately, mark it as \u201cIrrelevant\u201d with a reason why.",
"role": "system"
},
{
"content": "=Query: \"{{ $('Set variables1').item.json.inputQuery }}\"\nAgent Response: \"{{ $('Execute Agent Workflow').item.json.message }}\"\nExpected Response: \"{{ $('Set variables1').item.json.outputExpected }}\""
}
]
},
"jsonOutput": true,
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.openAi",
"typeVersion": 1.7,
"position": [
1640,
860
],
"id": "4976873c-25f9-4db6-a85d-247aa98deb65",
"name": "Judge Agent",
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"jsCode": "function parseStringToArray(inputString) {\n try {\n const result = JSON.parse(inputString);\n\n return result;\n } catch (error) {\n console.error(\"Erro ao parsear a string:\", error);\n return [];\n }\n}\n\nconst context = $('Set variables1').first().json.inputContext;\nconst contextArray = parseStringToArray(context);\n\nreturn {\n contextArray: contextArray,\n};"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
660,
860
],
"id": "aa8859cf-cba7-419a-b29e-5ef85d3f25b5",
"name": "Parse context (if needed)"
},
{
"parameters": {
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict",
"version": 2
},
"conditions": [
{
"id": "bee0a8c6-577a-41b4-8926-0e3e30c74f82",
"leftValue": "={{ $json.contextArray }}",
"rightValue": 0,
"operator": {
"type": "array",
"operation": "lengthGt",
"rightType": "number"
}
}
],
"combinator": "and"
},
"options": {}
},
"type": "n8n-nodes-base.if",
"typeVersion": 2.2,
"position": [
880,
860
],
"id": "db636afe-9176-4a6d-9e28-53c1bdd18a78",
"name": "Has context?"
},
{
"parameters": {
"options": {}
},
"type": "n8n-nodes-base.respondToWebhook",
"typeVersion": 1.1,
"position": [
2180,
860
],
"id": "17f60667-1164-44ea-8389-8969aea48159",
"name": "Respond to Webhook"
},
{
"parameters": {
"method": "POST",
"url": "={{ $('Set variables1').item.json.agentWorkflowUrl }}",
"sendBody": true,
"bodyParameters": {
"parameters": [
{
"name": "query",
"value": "={{ $('Set variables1').item.json.inputQuery }}"
},
{
"name": "user",
"value": "={{ $('Set variables1').item.json.sessionId }}"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
1400,
860
],
"id": "d1b38e3d-1663-4e00-a5f6-bf329ad01643",
"name": "Execute Agent Workflow"
},
{
"parameters": {
"method": "POST",
"url": "={{ $('Set variables1').item.json.agentWorkflowUrl }}",
"sendBody": true,
"bodyParameters": {
"parameters": [
{
"name": "query",
"value": "=@reset"
},
{
"name": "user",
"value": "={{ $('Set variables1').item.json.sessionId }}"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
440,
860
],
"id": "a110aa25-e66e-4ea2-b0e2-343eabf6abb5",
"name": "Clear context"
},
{
"parameters": {
"method": "POST",
"url": "={{ $('Set variables1').item.json.setContextWorkflowUrl }}",
"sendBody": true,
"bodyParameters": {
"parameters": [
{
"name": "contextArray",
"value": "={{ $json.contextArray }}"
},
{
"name": "sessionId",
"value": "={{ $('Set variables1').item.json.sessionId }}"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
1140,
840
],
"id": "21124846-0db6-413f-98a7-e0964ae8f63b",
"name": "Set context"
},
{
"parameters": {},
"id": "f937adf0-42f0-4c01-9658-8ed8507046ec",
"name": "End1",
"type": "n8n-nodes-base.noOp",
"typeVersion": 1,
"position": [
2360,
860
]
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "a9d33576-c812-485f-8000-e047fa022bf5",
"name": "inputContext",
"value": "={{ $json.body.context || null }}",
"type": "string"
},
{
"id": "6216ab60-f4ce-4c5b-9a6a-8fb06be9978a",
"name": "outputExpected",
"value": "={{ $json.body.output_expected }}",
"type": "string"
},
{
"id": "ef24c3c0-0e06-4f74-9c32-3bfaa964edca",
"name": "inputQuery",
"value": "={{ $json.body.input_query }}",
"type": "string"
},
{
"id": "91f4f31a-798b-4788-ac64-c58e4651ae0b",
"name": "scenarioId",
"value": "={{ $json.body.id }}",
"type": "string"
},
{
"id": "a88ab22d-10d7-4513-a775-a2c706d15891",
"name": "sessionId",
"value": "={{ 'testmyagent@s.whatsapp.net' }}",
"type": "string"
},
{
"id": "ab7cb549-b1ac-4c09-a36c-d71068cface6",
"name": "agentWorkflowUrl",
"value": "={{ $json.body.n8n_workflow_base_url }}",
"type": "string"
},
{
"id": "66816cb1-5216-477b-a150-62fb4b8a2468",
"name": "setContextWorkflowUrl",
"value": "={{ $json.body.n8n_workflow_base_url }}_set_context",
"type": "string"
}
]
},
"options": {}
},
"id": "b90f2da5-298a-48c7-99d6-f08181c8d991",
"name": "Set variables1",
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
-20,
860
]
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "b3c84eda-e252-4936-b593-e54af8c71a65",
"name": "input_query",
"value": "={{ $('Set variables1').item.json.inputQuery }}",
"type": "string"
},
{
"id": "44758a21-d132-4e0b-b96c-c708e8eb350f",
"name": "output_content",
"value": "={{ $('Execute Agent Workflow').item.json.message }}",
"type": "string"
},
{
"id": "0e89b823-e640-4712-8278-b5ee8ef0105d",
"name": "output_expected",
"value": "={{ $('Set variables1').item.json.outputExpected }}",
"type": "string"
},
{
"id": "a52bbd13-c328-4a9e-9ae1-08f58ed671a2",
"name": "eval_relevance",
"value": "={{ $json.message.content.relevance }}",
"type": "string"
},
{
"id": "e5b67922-be75-4b2c-aa79-d336e3d35f30",
"name": "eval_explanation",
"value": "={{ $json.message.content.explanation }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
2000,
860
],
"id": "4f3fc8ee-19bf-4e92-b3f2-2a0140bae55e",
"name": "Set response1"
},
{
"parameters": {
"httpMethod": "POST",
"path": "runTestExecution",
"responseMode": "responseNode",
"options": {}
},
"type": "n8n-nodes-base.webhook",
"typeVersion": 2,
"position": [
-260,
860
],
"id": "38292c25-0b44-4d28-ba98-6cc3b9dd9eeb",
"name": "runTestExecution"
},
{
"parameters": {
"content": "# runTestExecution",
"height": 80,
"width": 2740
},
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-280,
760
],
"id": "4d8dfd8f-33e3-4600-92a8-7a7150bdfb6b",
"name": "Sticky Note"
},
{
"parameters": {
"content": "# generateScenario",
"height": 80,
"width": 1480
},
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-260,
260
],
"id": "36ebd9dd-a172-43ab-a8f8-d39b290e7874",
"name": "Sticky Note1"
},
{
"parameters": {
"content": "# generateMultiScenarios",
"height": 80,
"width": 1460
},
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-260,
-260
],
"id": "a1576f59-3b61-404f-b6e5-230e98e713bc",
"name": "Sticky Note2"
},
{
"parameters": {
"jsCode": "const data = $input.first().json;\nconsole.log(data);\n\nconst keys = Object.keys(data);\nconsole.log(keys);\n\nkeys.forEach((key) => {\n if (key === 'inputContext') { return false; } \n if (!data[key]) {\n throw new Error(`Field ${key} cannot be empty.`)\n }\n})\n\nreturn $input.all();"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
200,
860
],
"id": "c99fc180-cd74-4e1f-8a59-77ef0a27f246",
"name": "Do Validation"
}
],
"connections": {
"Set response": {
"main": [
[
{
"node": "Respond to generateScenario",
"type": "main",
"index": 0
}
]
]
},
"Set variables": {
"main": [
[
{
"node": "One-Scenario Generator",
"type": "main",
"index": 0
}
]
]
},
"Structured Output Parser": {
"ai_outputParser": [
[
{
"node": "One-Scenario Generator",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "One-Scenario Generator",
"type": "ai_languageModel",
"index": 0
},
{
"node": "Multi-Scenario Generator",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"One-Scenario Generator": {
"main": [
[
{
"node": "Set response",
"type": "main",
"index": 0
}
]
]
},
"generateScenario": {
"main": [
[
{
"node": "Set variables",
"type": "main",
"index": 0
}
]
]
},
"generateMultiScenarios": {
"main": [
[
{
"node": "Set variables multi",
"type": "main",
"index": 0
}
]
]
},
"Set variables multi": {
"main": [
[
{
"node": "Multi-Scenario Generator",
"type": "main",
"index": 0
}
]
]
},
"Multi-Scenario Generator": {
"main": [
[
{
"node": "Set response multi",
"type": "main",
"index": 0
}
]
]
},
"Set response multi": {
"main": [
[
{
"node": "Respond to generateMultiScenarios",
"type": "main",
"index": 0
}
]
]
},
"Respond to generateMultiScenarios": {
"main": [
[
{
"node": "End",
"type": "main",
"index": 0
}
]
]
},
"Respond to generateScenario": {
"main": [
[
{
"node": "End",
"type": "main",
"index": 0
}
]
]
},
"Structured Output Parser multi": {
"ai_outputParser": [
[
{
"node": "Multi-Scenario Generator",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"Judge Agent": {
"main": [
[
{
"node": "Set response1",
"type": "main",
"index": 0
}
]
]
},
"Parse context (if needed)": {
"main": [
[
{
"node": "Has context?",
"type": "main",
"index": 0
}
]
]
},
"Has context?": {
"main": [
[
{
"node": "Set context",
"type": "main",
"index": 0
}
],
[
{
"node": "Execute Agent Workflow",
"type": "main",
"index": 0
}
]
]
},
"Respond to Webhook": {
"main": [
[
{
"node": "End1",
"type": "main",
"index": 0
}
]
]
},
"Execute Agent Workflow": {
"main": [
[
{
"node": "Judge Agent",
"type": "main",
"index": 0
}
]
]
},
"Clear context": {
"main": [
[
{
"node": "Parse context (if needed)",
"type": "main",
"index": 0
}
]
]
},
"Set context": {
"main": [
[
{
"node": "Execute Agent Workflow",
"type": "main",
"index": 0
}
]
]
},
"Set variables1": {
"main": [
[
{
"node": "Do Validation",
"type": "main",
"index": 0
}
]
]
},
"Set response1": {
"main": [
[
{
"node": "Respond to Webhook",
"type": "main",
"index": 0
}
]
]
},
"runTestExecution": {
"main": [
[
{
"node": "Set variables1",
"type": "main",
"index": 0
}
]
]
},
"Do Validation": {
"main": [
[
{
"node": "Clear context",
"type": "main",
"index": 0
}
]
]
}
},
"active": true,
"settings": {
"executionOrder": "v1"
},
"versionId": "a01e9dc4-e3d7-40de-8e68-26936842655e",
"id": "FtGyhAVW41oOWPSG",
"tags": []
}
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.
openAiApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
testmyagent_functions. Uses outputParserStructured, lmChatOpenAi, agent, openAi. Webhook trigger; 29 nodes.
Source: https://github.com/taigfs/public-assets/blob/d92517913eff620f6e585e4616ca5b9508eba4b1/dev/testmyagent_functions_n8n_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.
leads. Uses supabase, gmail, formTrigger, httpRequest. Webhook trigger; 62 nodes.
Supabase, Gmail, Form Trigger +13
This workflow automatically processes new free-trial / lead sign-ups in real time: Catches a webhook from any source (Webflow form, Intercom, custom agent, etc.) Filters out personal / disposable / .e
Output Parser Structured, Agent, HTTP Request +7
This workflow turns raw product inputs into a complete, launch-ready AI-generated social media campaign package. It accepts product details via webhook, sanitizes messy fields, generates a strategic c
Agent, Output Parser Structured, OpenAI Chat +4
Logo Animator - Supabase. Uses postgres, openAi, agent, toolThink. Webhook trigger; 26 nodes.
Postgres, OpenAI, Agent +4
🧾 An intelligent automation system that turns Google Meet recordings into structured meeting notes — integrating Fireflies.ai, OpenAI GPT-4.1-mini, Notion, Slack, Google Drive, and Gmail via n8n.
Google Drive, OpenAI Chat, Output Parser Structured +8