AI Agent with Persistent Memory (mengram)

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": "AI Agent with Persistent Memory (Mengram)",
  "nodes": [
    {
      "parameters": {
        "httpMethod": "POST",
        "path": "chat",
        "responseMode": "responseNode",
        "options": {}
      },
      "id": "a1b2c3d4-0001-4000-8000-000000000001",
      "name": "Webhook",
      "type": "n8n-nodes-base.webhook",
      "typeVersion": 2,
      "position": [
        240,
        300
      ]
    },
    {
      "parameters": {
        "method": "POST",
        "url": "https://mengram.io/v1/search",
        "authentication": "genericCredentialType",
        "genericAuthType": "httpHeaderAuth",
        "sendBody": true,
        "specifyBody": "json",
        "jsonBody": "={{ JSON.stringify({ query: $json.body.message, user_id: $json.body.user_id || 'default', limit: 5 }) }}",
        "options": {}
      },
      "id": "a1b2c3d4-0002-4000-8000-000000000002",
      "name": "Search Memories",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        480,
        300
      ],
      "credentials": {
        "httpHeaderAuth": {
          "name": "<your credential>"
        }
      }
    },
    {
      "parameters": {
        "jsCode": "const input = $input.first().json;\nconst userMessage = $('Webhook').first().json.body.message;\nconst userId = $('Webhook').first().json.body.user_id || 'default';\n\n// Format retrieved memories into context\nlet memoryContext = '';\nif (input.results && input.results.length > 0) {\n  const memories = input.results.map(r => {\n    let parts = [r.entity];\n    if (r.facts && r.facts.length) parts.push(r.facts.join('; '));\n    return parts.join(': ');\n  });\n  memoryContext = 'What you know about this user:\\n' + memories.join('\\n');\n}\n\nconst systemPrompt = `You are a helpful assistant with persistent memory.\n\n${memoryContext ? memoryContext + '\\n\\nUse this context to personalize your response. Reference past interactions naturally.' : 'No prior memories found for this user.'}`;\n\nreturn [{\n  json: {\n    systemPrompt,\n    userMessage,\n    userId,\n    memoryContext\n  }\n}];"
      },
      "id": "a1b2c3d4-0003-4000-8000-000000000003",
      "name": "Build Prompt",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        720,
        300
      ]
    },
    {
      "parameters": {
        "method": "POST",
        "url": "https://api.openai.com/v1/chat/completions",
        "authentication": "genericCredentialType",
        "genericAuthType": "httpHeaderAuth",
        "sendBody": true,
        "specifyBody": "json",
        "jsonBody": "={{ JSON.stringify({ model: 'gpt-4o-mini', messages: [ { role: 'system', content: $json.systemPrompt }, { role: 'user', content: $json.userMessage } ], temperature: 0.7 }) }}",
        "options": {}
      },
      "id": "a1b2c3d4-0004-4000-8000-000000000004",
      "name": "AI Response",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        960,
        300
      ],
      "credentials": {
        "httpHeaderAuth": {
          "name": "<your credential>"
        }
      }
    },
    {
      "parameters": {
        "jsCode": "const aiResponse = $input.first().json;\nconst userMessage = $('Build Prompt').first().json.userMessage;\nconst userId = $('Build Prompt').first().json.userId;\n\nconst assistantMessage = aiResponse.choices[0].message.content;\n\nreturn [{\n  json: {\n    assistantMessage,\n    userMessage,\n    userId\n  }\n}];"
      },
      "id": "a1b2c3d4-0005-4000-8000-000000000005",
      "name": "Extract Response",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        1200,
        300
      ]
    },
    {
      "parameters": {
        "method": "POST",
        "url": "https://mengram.io/v1/add",
        "authentication": "genericCredentialType",
        "genericAuthType": "httpHeaderAuth",
        "sendBody": true,
        "specifyBody": "json",
        "jsonBody": "={{ JSON.stringify({ messages: [ { role: 'user', content: $json.userMessage }, { role: 'assistant', content: $json.assistantMessage } ], user_id: $json.userId }) }}",
        "options": {}
      },
      "id": "a1b2c3d4-0006-4000-8000-000000000006",
      "name": "Save to Memory",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        1440,
        300
      ],
      "credentials": {
        "httpHeaderAuth": {
          "name": "<your credential>"
        }
      }
    },
    {
      "parameters": {
        "respondWith": "json",
        "responseBody": "={{ JSON.stringify({ response: $('Extract Response').first().json.assistantMessage, user_id: $('Extract Response').first().json.userId, memory_saved: true }) }}",
        "options": {}
      },
      "id": "a1b2c3d4-0007-4000-8000-000000000007",
      "name": "Respond",
      "type": "n8n-nodes-base.respondToWebhook",
      "typeVersion": 1.1,
      "position": [
        1680,
        300
      ]
    }
  ],
  "connections": {
    "Webhook": {
      "main": [
        [
          {
            "node": "Search Memories",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Search Memories": {
      "main": [
        [
          {
            "node": "Build Prompt",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Build Prompt": {
      "main": [
        [
          {
            "node": "AI Response",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "AI Response": {
      "main": [
        [
          {
            "node": "Extract Response",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Extract Response": {
      "main": [
        [
          {
            "node": "Save to Memory",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Save to Memory": {
      "main": [
        [
          {
            "node": "Respond",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  },
  "settings": {
    "executionOrder": "v1"
  },
  "staticData": null,
  "tags": [],
  "triggerCount": 0,
  "updatedAt": "2026-02-27T00:00:00.000Z",
  "versionId": "1"
}

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.

• httpHeaderAuth

About this workflow

AI Agent with Persistent Memory (Mengram). Uses httpRequest. Webhook trigger; 7 nodes.

Source: https://github.com/alibaizhanov/mengram/blob/31b7ed19f960f6c575c4b4307c82897dc89586e1/examples/n8n/mengram-memory-agent.json — original creator credit. Request a take-down →

More AI & RAG workflows → · Browse all categories →

Related workflows

Workflows that share integrations, category, or trigger type with this one. All free to copy and import.