Reply Detection & Follow-up

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": "04 - Reply Detection & Follow-up",
  "nodes": [
    {
      "parameters": {
        "rule": {
          "interval": [
            {
              "field": "hours",
              "hoursInterval": 2
            }
          ]
        }
      },
      "id": "reply-trigger",
      "name": "Check Every 2 Hours",
      "type": "n8n-nodes-base.scheduleTrigger",
      "typeVersion": 1.2,
      "position": [
        240,
        300
      ]
    },
    {
      "id": "imap-trigger-node",
      "name": "Fetch Unread Email (IMAP)",
      "type": "n8n-nodes-base.emailReadImap",
      "typeVersion": 2,
      "position": [
        460,
        300
      ],
      "parameters": {
        "mailbox": "INBOX",
        "action": "read",
        "markSeen": true,
        "download": false,
        "options": {}
      },
      "credentials": {
        "imap": {
          "name": "<your credential>"
        }
      }
    },
    {
      "parameters": {
        "jsCode": "// Filter replies from target emails in outreach_log\nconst emails = $input.all();\nconst replies = emails.filter(item => {\n  const from    = item.json.from || '';\n  const subject = (item.json.subject || '').toLowerCase();\n  const body    = (item.json.text || item.json.snippet || '').toLowerCase();\n  \n  // Exclude auto-replies\n  const isAutoReply = [\n    'out of office', 'auto-reply', 'automatic reply',\n    'vacation', 'away from', 'noreply',\n  ].some(kw => subject.includes(kw) || body.includes(kw));\n  \n  if (isAutoReply) return false;\n  \n  // Detect positive reply intent\n  const positiveSignals = [\n    'interested', 'yes', 'sure', 'please', 'tell me more',\n    'call', 'meeting', 'schedule', 'demo', 'price', 'cost',\n    'how much', 'magkano', 'tara', 'ok', 'pwede',\n  ];\n  const hasPositive = positiveSignals.some(kw => body.includes(kw));\n  return true; // Return all unread emails, classify in next node\n});\n\nconsole.log(`Emails to process: ${replies.length}`);\nreturn replies.slice(0, 20);"
      },
      "id": "filter-replies",
      "name": "Filter & Parse Replies",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        680,
        300
      ]
    },
    {
      "parameters": {
        "model": "gemini-2.0-flash",
        "messages": {
          "values": [
            {
              "role": "user",
              "content": "Classify this email reply from a Philippine business:\n\nFrom: {{ $json.from }}\nSubject: {{ $json.subject }}\nBody: {{ $json.text || $json.snippet }}\n\nReturn JSON only:\n{\"intent\": \"interested|not_interested|auto_reply|unsubscribe|other\", \"sentiment\": \"positive|neutral|negative\", \"suggested_action\": \"schedule_call|send_followup|remove_from_list|no_action\", \"urgency\": \"high|medium|low\", \"summary\": \"one line summary\"}"
            }
          ]
        },
        "options": {
          "temperature": 0.1
        }
      },
      "id": "classify-intent",
      "name": "Classify Intent (Gemini)",
      "type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
      "typeVersion": 1,
      "position": [
        900,
        300
      ],
      "credentials": {
        "googlePalmApi": {
          "name": "<your credential>"
        }
      }
    },
    {
      "parameters": {
        "jsCode": "const email    = $('Filter & Parse Replies').item.json;\nconst rawReply = $input.item.json.text || $input.item.json.response || '{}';\n\nlet classification = {};\ntry {\n  const jsonStart = rawReply.indexOf('{');\n  const jsonEnd   = rawReply.lastIndexOf('}') + 1;\n  if (jsonStart >= 0) classification = JSON.parse(rawReply.slice(jsonStart, jsonEnd));\n} catch(e) {\n  classification = { intent: 'other', sentiment: 'neutral', suggested_action: 'no_action' };\n}\n\nreturn [{\n  json: {\n    email_id:         email.uid || email.id || \"unknown\",\n    from:             email.from,\n    subject:          email.subject,\n    intent:           classification.intent,\n    sentiment:        classification.sentiment,\n    suggested_action: classification.suggested_action,\n    urgency:          classification.urgency,\n    summary:          classification.summary,\n    timestamp:        new Date().toISOString(),\n  }\n}];"
      },
      "id": "parse-classification",
      "name": "Parse Classification",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        1120,
        300
      ]
    },
    {
      "parameters": {
        "conditions": {
          "options": {
            "caseSensitive": false
          },
          "conditions": [
            {
              "leftValue": "={{ $json.intent }}",
              "rightValue": "interested",
              "operator": {
                "type": "string",
                "operation": "equals"
              }
            }
          ]
        }
      },
      "id": "route-intent",
      "name": "Is Interested?",
      "type": "n8n-nodes-base.if",
      "typeVersion": 2,
      "position": [
        1340,
        300
      ]
    },
    {
      "parameters": {
        "method": "POST",
        "url": "http://sales-outreach:8080/webhook/hot-lead",
        "sendBody": true,
        "bodyParameters": {
          "parameters": [
            {
              "name": "from",
              "value": "={{ $json.from }}"
            },
            {
              "name": "subject",
              "value": "={{ $json.subject }}"
            },
            {
              "name": "intent",
              "value": "={{ $json.intent }}"
            },
            {
              "name": "urgency",
              "value": "={{ $json.urgency }}"
            },
            {
              "name": "suggested_action",
              "value": "={{ $json.suggested_action }}"
            },
            {
              "name": "summary",
              "value": "={{ $json.summary }}"
            }
          ]
        }
      },
      "id": "alert-hot-lead",
      "name": "Alert: Hot Lead!",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        1560,
        220
      ],
      "notes": "Replace with Telegram Bot or Slack for urgent notification push"
    },
    {
      "parameters": {
        "method": "POST",
        "url": "http://lead-miner:8000/leads/update-status",
        "sendBody": true,
        "bodyParameters": {
          "parameters": [
            {
              "name": "email",
              "value": "={{ $('Parse Classification').item.json.from }}"
            },
            {
              "name": "status",
              "value": "={{ $('Parse Classification').item.json.intent }}"
            }
          ]
        }
      },
      "id": "update-outreach-log",
      "name": "Update outreach_log",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        1560,
        560
      ],
      "continueOnFail": true,
      "notes": "Update outreach_log.status field to record reply status"
    }
  ],
  "connections": {
    "Check Every 2 Hours": {
      "main": [
        [
          {
            "node": "Fetch Unread Email (IMAP)",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Filter & Parse Replies": {
      "main": [
        [
          {
            "node": "Classify Intent (Gemini)",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Classify Intent (Gemini)": {
      "main": [
        [
          {
            "node": "Parse Classification",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Parse Classification": {
      "main": [
        [
          {
            "node": "Is Interested?",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Is Interested?": {
      "main": [
        [
          {
            "node": "Alert: Hot Lead!",
            "type": "main",
            "index": 0
          }
        ],
        [
          {
            "node": "Update outreach_log",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Alert: Hot Lead!": {
      "main": [
        [
          {
            "node": "Update outreach_log",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Fetch Unread Email (IMAP)": {
      "main": [
        [
          {
            "node": "Filter & Parse Replies",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  },
  "active": false,
  "settings": {
    "executionOrder": "v1",
    "saveExecutionProgress": true
  },
  "tags": [],
  "meta": {
    "description": "Check unread emails every 2 hours \u2014 use Gemini to classify reply intent, push alerts for hot leads immediately"
  },
  "id": "357b1ca7-4294-4f04-81ad-63200eb64b7f"
}

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.

• googlePalmApi
• imap

About this workflow

04 - Reply Detection & Follow-up. Uses emailReadImap, lmChatGoogleGemini, httpRequest. Scheduled trigger; 8 nodes.

Source: https://github.com/kaneliu120/lead-mining-system/blob/9197dfc0289d444b23dc2d461544adc8380bffb8/n8n-workflows/04-reply-detection.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.