SuperMCP(預覽)
SuperMCP by Supermetrics - 一台 MCP(模型情境協定)伺服器,提供來自 Google Analytics、Facebook 廣告、Google 廣告等連接平台的行銷資料存取。
此 API 使大型語言模型能查詢行銷資料、發現資料來源與帳戶、探索可用欄位,並取得績效指標。
此連接器適用於以下產品和區域:
| 服務 | Class | Regions |
|---|---|---|
| 副駕駛工作室 | 進階 | 除下列區域外的所有 Power Automate 區域 : - 美國政府 (GCC) - 美國政府(海灣合作委員會高中) - 由 21Vianet 營運的中國雲 - 美國國防部 (DoD) |
| 邏輯應用程式 | 標準 | 所有 Logic Apps 區域, 但下列區域除外: - Azure Government 區域 - Azure 中國區域 - 美國國防部 (DoD) |
| Power Apps | 進階 | 除下列區域外的所有 Power Apps 區域 : - 美國政府 (GCC) - 美國政府(海灣合作委員會高中) - 由 21Vianet 營運的中國雲 - 美國國防部 (DoD) |
| Power Automate(自動化服務) | 進階 | 除下列區域外的所有 Power Automate 區域 : - 美國政府 (GCC) - 美國政府(海灣合作委員會高中) - 由 21Vianet 營運的中國雲 - 美國國防部 (DoD) |
| 連絡人 | |
|---|---|
| 名稱 | Supermetrics |
| URL | https://supermetrics.com |
| 連接器中繼資料 | |
|---|---|
| Publisher | Supermetrics |
| 網站 | https://supermetrics.com |
| 隱私策略 | https://supermetrics.com/privacy |
| 類別 | Marketing |
SuperMCP(Supermetrics)自訂連接器
此連接器使 Microsoft Power Automate、Power Apps 及 Azure Logic Apps 能透過 SuperMCP (Supermetrics 的 MCP 伺服器)查詢行銷資料。
利用它來發現資料來源和帳戶、探索可用欄位,並執行行銷績效查詢。
Publisher
Supermetrics
連接器類型
自訂連接器(Swagger/OpenAPI 2.0)
服務 / API
SuperMCP API(MCP 伺服器)
-
主:
mcp.supermetrics.com -
基地路線:
/ -
方案:
https
先決條件
您將需要:
- 一個可以建立自訂連接器的 Power Automate 或 Power Apps 環境
- 一個至少連接一個資料來源的 Supermetrics 帳戶(例如 Google Analytics、Google Ads、Facebook Ads 等)。 如果你沒有 Supermetrics 帳號,初次登入時會為你建立一個有 14 天試用期的帳號。
Authentication
此連接器使用 OAuth 2.0。
支援的作業
發現
-
探索可用的 Supermetrics 資料來源 (
POST /mcp/data_source_discovery) -
發現資料來源的帳號 (
POST /mcp/accounts_discovery) -
發現可用欄位 (
POST /mcp/field_discovery)
資料查詢
-
執行非同步資料查詢 (
POST /mcp/data_query)- 返回 a
schedule_id
- 返回 a
-
取得非同步查詢結果 (
POST /mcp/get_async_query_results)- 使用直到
schedule_id結果回傳
- 使用直到
公用事業/資源(通常為進階可視性)
-
今天取得 (UTC) (
POST /mcp/get_today) -
健康檢查 (
GET /mcp/resources/health)
Support
- 網站:https://supermetrics.com
- MCP 伺服器文件:https://mcp.supermetrics.com
- 支援:https://support.supermetrics.com
許可證
專有 — © 超級度量
節流限制
| 名稱 | 呼叫 | 續約期間 |
|---|---|---|
| 每個連線的 API 呼叫 | 100 | 60 秒 |
動作
| 使用 schedule_id 來檢索資料查詢的結果 |
使用 schedule_id 來檢索資料查詢結果 data_query 的 schedule_id 可取得實際查詢結果。 |
| 健康檢查 |
取得伺服器健康狀態和設定資訊。 |
| 從 Supermetrics 檢索經過認證的使用者資訊 |
從 Supermetrics 檢索經過認證的使用者資訊 |
| 探索可用的 Supermetrics 資料來源 |
發現可用的 Supermetrics 資料來源 這通常是第一步——先打電話查詢,先看看有哪些資料來源可用,再查詢資料。 |
| 查詢來自 Supermetrics 資料來源的行銷資料 |
查詢 Supermetrics 資料來源的行銷資料 這會回傳一個schedule_id - 你必須打電話給 get_async_query_results 才能取得實際資料。 |
| 獲取最新的UTC日期與時間資訊 |
取得最新的 UTC 日期與時間資訊 當使用者提到像是「上個月」或「昨天」等相對日期時,請先撥打這個來查詢當前日期。 |
| 發現包含結構資訊的資料來源可用欄位 |
在查詢前,發現可用資料來源的欄位,包含結構資訊。呼叫此資料以查看資料來源可用的指標與維度,然後再進行查詢。 |
| 發現特定資料來源的可用帳戶 |
發現特定資料來源的可用帳號 data_source_discovery後呼叫此帳號以取得資料查詢所需的帳號 ID。 |
使用 schedule_id 來檢索資料查詢的結果
使用 schedule_id 來檢索資料查詢結果 data_query 的 schedule_id 可取得實際查詢結果。
參數
| 名稱 | 機碼 | 必要 | 類型 | Description |
|---|---|---|---|---|
|
排程 ID
|
schedule_id | True | string |
排程ID是data_query請求回來的。 Schedule ID 必須與data_query回覆中回覆的完全相同。 把身分證當作不透明。 逐字複製。 請勿縮寫、格式化或意譯。 |
|
壓縮反應
|
compress | boolean |
啟用卡通壓縮以回應 |
傳回
健康檢查
從 Supermetrics 檢索經過認證的使用者資訊
探索可用的 Supermetrics 資料來源
發現可用的 Supermetrics 資料來源 這通常是第一步——先打電話查詢,先看看有哪些資料來源可用,再查詢資料。
參數
| 名稱 | 機碼 | 必要 | 類型 | Description |
|---|---|---|---|---|
|
壓縮反應
|
compress | boolean |
啟用卡通壓縮以回應 |
傳回
查詢來自 Supermetrics 資料來源的行銷資料
查詢 Supermetrics 資料來源的行銷資料 這會回傳一個schedule_id - 你必須打電話給 get_async_query_results 才能取得實際資料。
參數
| 名稱 | 機碼 | 必要 | 類型 | Description |
|---|---|---|---|---|
|
資料來源識別碼
|
ds_id | True | string |
可透過data_source_discovery工具檢索的資料來源ID。 範例:Facebook 廣告用「FA」,Google 廣告用「AW」,Google Analytics 用「GAWA」。 |
|
ds_accounts
|
ds_accounts | True | array of string | |
|
fields
|
fields | True | array of string | |
|
日期範圍類型
|
date_range_type | string |
請選擇兩種方法之一:
重要提示:對於使用者提及的任何相關日期,請先撥打get_today工具以掌握當前年份和日期!
如果時間段有任何歧義(例如「今天」、「本週」、「去年八月」等詞語的意義),請呼叫 |
|
|
開始日期
|
start_date | string |
若 date_range_type=「自訂」則為必要。 接受:
重要提示:對於使用者提及的任何相關日期,請先撥打get_today工具以掌握當前年份和日期!
如果時間參考有歧義(例如「今天」、「本週」、「月底」等相對詞語), |
|
|
結束日期
|
end_date | string |
若 date_range_type=「自訂」則為必要。 接受:
重要提示:對於使用者提及的任何相關日期,請先撥打get_today工具以掌握當前年份和日期!
如果時間參考有歧義(例如「今天」、「本週」、「月底」等相對詞語), |
|
|
時區
|
timezone | string |
可選擇設定時區以便計算日期(例如「美國/New_York」、「歐洲/倫敦」、「亞洲/東京」)。 使用 IANA 時區名稱。 若未特別說明,則使用 UTC。 |
|
|
篩選
|
filters | string |
過濾表達式字串。 每個條件的形式為:[field_name] [運算子] [值] 條件可與 AND / OR 結合。 支援的運算子:
範例:
|
|
|
設定
|
settings | object |
額外的查詢設定。 使用 JSON 物件,例如:{“show_all_time_values”: true}。 針對特定資料來源設定的進階功能。 除非你有特定要求,否則可以跳過。 |
|
|
比較時期類型
|
compare_type | string |
選擇如何與較早期期間比較結果:
|
|
|
比較價值類型
|
compare_show | string |
如何展示比較:
|
|
|
比較期開始日期
|
compare_start_date | string |
若 compare_type=「自訂」則為必要。 自訂比較期間的開始日期,格式為 YYYY-MM-DD(例如「2024-01-01」)。 |
|
|
比較期結束日期
|
compare_end_date | string |
若 compare_type=「自訂」則為必要。 自訂比較期間的結束日期,以 YYYY-MM-DD 格式表示(例如,「2024-01-31」)。 |
傳回
獲取最新的UTC日期與時間資訊
取得最新的 UTC 日期與時間資訊 當使用者提到像是「上個月」或「昨天」等相對日期時,請先撥打這個來查詢當前日期。
參數
| 名稱 | 機碼 | 必要 | 類型 | Description |
|---|---|---|---|---|
|
|
object |
傳回
發現包含結構資訊的資料來源可用欄位
在查詢前,發現可用資料來源的欄位,包含結構資訊。呼叫此資料以查看資料來源可用的指標與維度,然後再進行查詢。
參數
| 名稱 | 機碼 | 必要 | 類型 | Description |
|---|---|---|---|---|
|
資料來源識別碼
|
ds_id | True | string |
可透過data_source_discovery工具檢索的資料來源ID。 範例:Facebook 廣告用「FA」,Google 廣告用「AW」,Google Analytics 用「GAWA」。 |
|
欄位類型
|
field_type | string |
依欄位類型(維度或度量)篩選 |
|
|
壓縮反應
|
compress | boolean |
啟用卡通壓縮以回應 |
傳回
發現特定資料來源的可用帳戶
發現特定資料來源的可用帳號 data_source_discovery後呼叫此帳號以取得資料查詢所需的帳號 ID。
參數
| 名稱 | 機碼 | 必要 | 類型 | Description |
|---|---|---|---|---|
|
資料來源識別碼
|
ds_id | True | string |
可透過data_source_discovery工具檢索的資料來源ID。 範例:Facebook 廣告用「FA」,Google 廣告用「AW」,Google Analytics 用「GAWA」。 |
|
壓縮反應
|
compress | boolean |
啟用卡通壓縮以回應 |
傳回
定義
帳戶調查回應
| 名稱 | 路徑 | 類型 | Description |
|---|---|---|---|
|
勝
|
success | boolean | |
|
accounts
|
data.accounts | array of object | |
|
account_id
|
data.accounts.account_id | string | |
|
account_name
|
data.accounts.account_name | string |
DataqueryResponse
| 名稱 | 路徑 | 類型 | Description |
|---|---|---|---|
|
勝
|
success | boolean | |
|
schedule_id
|
data.schedule_id | string |
資料來源發現回應
| 名稱 | 路徑 | 類型 | Description |
|---|---|---|---|
|
勝
|
success | boolean | |
|
data_sources
|
data.data_sources | array of object | |
|
識別碼
|
data.data_sources.id | string | |
|
名稱
|
data.data_sources.name | string | |
|
分類
|
data.data_sources.category | string |
現場發現回應
| 名稱 | 路徑 | 類型 | Description |
|---|---|---|---|
|
勝
|
success | boolean | |
|
資料來源
|
data.data_source | string | |
|
總計
|
data.total | integer | |
|
計量
|
data.metrics | array of object | |
|
識別碼
|
data.metrics.id | string | |
|
名稱
|
data.metrics.name | string | |
|
description
|
data.metrics.description | string | |
|
尺寸
|
data.dimensions | array of object | |
|
識別碼
|
data.dimensions.id | string | |
|
名稱
|
data.dimensions.name | string | |
|
description
|
data.dimensions.description | string |
GetasyncqueryresultsResponse
| 名稱 | 路徑 | 類型 | Description |
|---|---|---|---|
|
勝
|
success | boolean | |
|
資料
|
data.data | array of object | |
|
items
|
data.data | object | |
|
狀態
|
data.status | string |
Gettoday回應
| 名稱 | 路徑 | 類型 | Description |
|---|---|---|---|
|
勝
|
success | boolean | |
|
date
|
data.date | date | |
|
iso_string
|
data.iso_string | string | |
|
時間戳記
|
data.timestamp | integer |
使用者資訊回應
| 名稱 | 路徑 | 類型 | Description |
|---|---|---|---|
|
勝
|
success | boolean | |
|
用戶識別碼
|
data.user_id | string | |
|
電子郵件
|
data.email | string | |
|
名稱
|
data.name | string |
字串
這是基本資料類型「string」。