会議メモジェネレーター

中級

これはAI Summarization, Multimodal AI分野の自動化ワークフローで、8個のノードを含みます。主にCode, Slack, GoogleDrive, OpenAi, GoogleDriveTriggerなどのノードを使用。 GPT を使って会議録要を自動生成し、Google Drive から Slack へ共有

前提条件
  • Slack Bot Token または Webhook URL
  • Google Drive API認証情報
  • OpenAI API Key
ワークフロープレビュー
ノード接続関係を可視化、ズームとパンをサポート
ワークフローをエクスポート
以下のJSON設定をn8nにインポートして、このワークフローを使用できます
{
  "meta": {
    "instanceId": "c8a161ccc9e6113553701825d60b54fb91bb0fc28cec75dbb04ffae3dcfe1522"
  },
  "name": "minutes_generator",
  "tags": [],
  "nodes": [
    {
      "id": "af25423f-a461-4b1d-9614-c91438b3196f",
      "name": "Google Drive Trigger",
      "type": "n8n-nodes-base.googleDriveTrigger",
      "position": [
        -704,
        144
      ],
      "parameters": {
        "event": "fileCreated",
        "options": {},
        "pollTimes": {
          "item": [
            {}
          ]
        },
        "triggerOn": "specificFolder",
        "folderToWatch": {
          "__rl": true,
          "mode": "id",
          "value": "__GOOGLE_DRIVE_FOLDER_ID__"
        }
      },
      "typeVersion": 1
    },
    {
      "id": "6049c9c7-3d0b-4eda-b309-98727bc8af8a",
      "name": "ファイルをダウンロード",
      "type": "n8n-nodes-base.googleDrive",
      "position": [
        -496,
        144
      ],
      "parameters": {
        "fileId": {
          "__rl": true,
          "mode": "id",
          "value": "={{$json.id || $json.fileId}}"
        },
        "options": {},
        "operation": "download"
      },
      "typeVersion": 3
    },
    {
      "id": "43d793f5-7bf8-441a-ab4e-afe05f35a9cc",
      "name": "モデルにメッセージ",
      "type": "@n8n/n8n-nodes-langchain.openAi",
      "position": [
        -80,
        144
      ],
      "parameters": {
        "modelId": {
          "__rl": true,
          "mode": "list",
          "value": "gpt-5",
          "cachedResultName": "GPT-5"
        },
        "options": {},
        "messages": {
          "values": [
            {
              "role": "system",
              "content": "# Goal\nSummarize the meeting transcript concisely and in a structured way. Include all key discussion points, decisions, and follow-up tasks without omission.\n\n# Steps\n1. Review the provided meeting transcript text.\n2. Summarize into 3–7 key topics of discussion.\n3. Separate confirmed decisions from pending/undecided items.\n4. List follow-up tasks (Task | Owner | Deadline | Notes) in a table if mentioned.\n5. Mark unclear or missing information as N/A.\n6. Verify the final summary with a checklist to ensure no omissions.\n\n# Output format\n- Structure:\n  1. Meeting overview (date, participants, purpose – only if available)\n  2. Key discussion points (list format)\n  3. Decisions (list format)\n  4. Pending items (list format)\n  5. Follow-up tasks (table: Task | Owner | Deadline | Notes)\n- Use Markdown\n- Use N/A where information is missing\n\n# Notes\n- If there are uncertainties, ask up to 3 clarification questions before proceeding.\n- Base the summary strictly on the original text. Do not guess or invent.\n- Self-check before submission:\n  • Are all key discussions, decisions, and tasks included?  \n  • Are N/A markers used for missing info?  \n  • Is the Markdown structure followed?"
            },
            {
              "content": "={{$json.transcript}}\n"
            }
          ]
        }
      },
      "credentials": {
        "openAiApi": {
          "id": "1Lp07lb1bLg8lVeX",
          "name": "OpenAi account"
        }
      },
      "typeVersion": 1.8
    },
    {
      "id": "97a34395-14af-42f6-9103-0f993bc89e08",
      "name": "議事録を準備",
      "type": "n8n-nodes-base.code",
      "position": [
        -288,
        144
      ],
      "parameters": {
        "jsCode": "const out = [];\nfor (const it of $input.all()) {\n  const b64 = it.binary?.data?.data; // from Drive Download\n  if (!b64) throw new Error('No binary \"data\". Use Download for txt/md');\n  const text = Buffer.from(b64, 'base64').toString('utf8');\n  const title = it.json?.name || it.json?.fileName || '회의록';\n\n  out.push({\n    json: { title, transcript: text },\n    binary: it.binary // keep original file for Slack upload\n  });\n}\nreturn out;\n"
      },
      "typeVersion": 2
    },
    {
      "id": "f96a5216-a8f8-4e73-8ca2-b772f5918a4a",
      "name": "Make Minutes",
      "type": "n8n-nodes-base.code",
      "position": [
        272,
        144
      ],
      "parameters": {
        "jsCode": "const it = $input.first();\n\n// Pick the model output from any of the shapes n8n returns\nconst md =\n  it.json?.message?.content ??                    // Message a model (simplified)\n  it.json?.output_text ??                         // some builds expose this\n  it.json?.data?.choices?.[0]?.message?.content ??// raw OpenAI response\n  it.json?.choices?.[0]?.message?.content ??      // another raw variant\n  '';\n\nconst title = it.json?.title || '회의록';\nconst filename = `${title}.md`;\nconst data = Buffer.from(md, 'utf8').toString('base64');\n\nreturn [{\n  json: { filename, title },\n  binary: { minutes: { data, fileName: filename, mimeType: 'text/markdown' } }\n}];"
      },
      "typeVersion": 2
    },
    {
      "id": "f96c7ca7-48ea-4fdd-92b8-29f15a22eff6",
      "name": "メッセージを送信",
      "type": "n8n-nodes-base.slack",
      "position": [
        480,
        144
      ],
      "webhookId": "__WEBHOOK_INTERNAL_ID__",
      "parameters": {
        "text": "={{\"*Auto-generated minutes:* \" + ($json.title || $json.name || \"Minutes\")}}",
        "select": "channel",
        "channelId": {
          "__rl": true,
          "mode": "id",
          "value": "___SLACK_CHANNEL_ID___"
        },
        "otherOptions": {}
      },
      "typeVersion": 2.3
    },
    {
      "id": "6b557b78-cc8c-4b86-ad48-d662fc1accbd",
      "name": "ファイルをアップロード",
      "type": "n8n-nodes-base.slack",
      "position": [
        480,
        288
      ],
      "webhookId": "__WEBHOOK_INTERNAL_ID__",
      "parameters": {
        "options": {
          "fileName": "={{$json.filename}}",
          "channelId": "___SLACK_CHANNEL_ID___"
        },
        "resource": "file",
        "binaryPropertyName": "minutes"
      },
      "typeVersion": 2.3
    },
    {
      "id": "e3f63e52-eba0-4b81-8fbf-ba835410c392",
      "name": "付箋",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -736,
        -192
      ],
      "parameters": {
        "width": 736,
        "height": 288,
        "content": "## Minutes generator (overview)\n**What it does**: Watches a Drive folder, downloads new transcript, summarizes via GPT, posts message + uploads Markdown minutes to Slack.\n\n**Setup**\n1. Google Drive: set folder ID in *Google Drive Trigger*.\n2. OpenAI: add credentials and pick a model in *Message a model*.\n3. Slack: connect account, set channel ID in both Slack nodes.\n\n**Notes**\n- Files must be text (.txt/.md) unless you add a converter.\n- Bot must be in the Slack channel to avoid `not_in_channel`."
      },
      "typeVersion": 1
    }
  ],
  "active": false,
  "pinData": {},
  "settings": {
    "executionOrder": "v1"
  },
  "versionId": "",
  "connections": {
    "f96a5216-a8f8-4e73-8ca2-b772f5918a4a": {
      "main": [
        [
          {
            "node": "f96c7ca7-48ea-4fdd-92b8-29f15a22eff6",
            "type": "main",
            "index": 0
          },
          {
            "node": "6b557b78-cc8c-4b86-ad48-d662fc1accbd",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "6049c9c7-3d0b-4eda-b309-98727bc8af8a": {
      "main": [
        [
          {
            "node": "97a34395-14af-42f6-9103-0f993bc89e08",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "43d793f5-7bf8-441a-ab4e-afe05f35a9cc": {
      "main": [
        [
          {
            "node": "f96a5216-a8f8-4e73-8ca2-b772f5918a4a",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "97a34395-14af-42f6-9103-0f993bc89e08": {
      "main": [
        [
          {
            "node": "43d793f5-7bf8-441a-ab4e-afe05f35a9cc",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "af25423f-a461-4b1d-9614-c91438b3196f": {
      "main": [
        [
          {
            "node": "6049c9c7-3d0b-4eda-b309-98727bc8af8a",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  }
}
よくある質問

このワークフローの使い方は?

上記のJSON設定コードをコピーし、n8nインスタンスで新しいワークフローを作成して「JSONからインポート」を選択、設定を貼り付けて認証情報を必要に応じて変更してください。

このワークフローはどんな場面に適していますか?

中級 - AI要約, マルチモーダルAI

有料ですか?

このワークフローは完全無料です。ただし、ワークフローで使用するサードパーティサービス(OpenAI APIなど)は別途料金が発生する場合があります。

ワークフロー情報
難易度
中級
ノード数8
カテゴリー2
ノードタイプ6
難易度説明

経験者向け、6-15ノードの中程度の複雑さのワークフロー

外部リンク
n8n.ioで表示

このワークフローを共有

カテゴリー

カテゴリー: 34