Adrails

ai.adrailsv1.0.1更新於 Oct 3, 2026

Specialist AI agents for Meta Ads and Google Ads. Every change waits for your approval.

已驗證Streamable HTTP可網頁執行Business & Commerce

概覽

AI 產生的概覽

將助理連接到 Meta Ads 和 Google Ads 帳戶、商店與分析資料，廣告變更僅以需你核准的提案形式回傳。

功能
這是 Adrails 付費媒體工作空間的託管遠端 MCP 伺服器，提供 63 個工具，依你授予的權限和工作空間角色篩選。功能涵蓋 Meta 與 Google Ads 的期間與每日成效、指標比較、診斷和組合檢視，並並列顯示商店營收。還可進行廣告活動結構、草稿、範本、素材、受眾和關鍵字研究，預算、狀態或投放變更會產生附前後對照連結的提案。此外可存取 Shopify、Klaviyo、Stripe、Triple Whale 和 PostHog 的業務資料，以及工作空間的自動化、Knowledge、頻道和工作狀態。
適用情境
適合讓助理結合商店營收讀取與分析 Meta 和 Google Ads 的付費媒體成效，或準備廣告活動與預算變更供人工審核。已在 Adrails 工作空間作業的團隊，若希望在 MCP 用戶端內完成分析與提案草擬，可使用此伺服器。
執行需求
遠端 streamable HTTP 端點 adrails.ai，無需安裝或執行任何項目。需要 Adrails 帳戶，透過 OAuth 2.1（PKCE 與動態用戶端註冊）登入，並在同意頁面選擇工作空間和權限。用戶端需支援帶 OAuth 的遠端 MCP。
安裝前請注意
此伺服器可存取真實的 Meta Ads 和 Google Ads 帳戶，以及 Shopify、Stripe、Klaviyo 等已連接的業務系統，因此能讀取廣告和營收資料。廣告變更不會直接生效：它們以提案形式回傳，需你登入 Adrails 明確核准，且提案會過期。可隨時在 Adrails 的 MCP 頁面撤銷連接。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Adrails，將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "adrails": {
      "type": "http",
      "url": "https://adrails.ai/api/mcp"
    }
  }
}

README

[Adrails]

Adrails MCP server

Remote MCP server for Adrails, the AI workspace for paid media. Connect Claude, Claude Code, ChatGPT, Codex or any remote MCP client to your Meta Ads and Google Ads accounts, your store and your analytics. Reads run at once; every advertising change comes back as a proposal you approve in the browser before it reaches Meta or Google.

This repository holds the listing metadata. The server itself is hosted by Adrails: there is nothing to install or run.

Server URLhttps://adrails.ai/api/mcp
TransportStreamable HTTP
AuthenticationOAuth 2.1 with PKCE and dynamic client registration
Official MCP Registryai.adrails/adrails
Documentationadrails.ai/docs/mcp

Connect

You need an Adrails account. Every client uses the same URL; the client registers itself and you sign in to Adrails, then choose the workspace and the permissions on the consent screen.

Claude (claude.ai and desktop): Settings, Connectors, Add custom connector. Name it Adrails, paste https://adrails.ai/api/mcp, leave the OAuth client ID and secret empty, then connect.

Claude Code

sh
claude mcp add --transport http adrails https://adrails.ai/api/mcp

Then run /mcp, select adrails and authenticate.

ChatGPT: turn on Developer mode in Settings (it depends on your plan and workspace policy), open Plugins and add a custom MCP server named Adrails with the URL above and OAuth authentication.

Codex

sh
codex mcp add adrails --url https://adrails.ai/api/mcpcodex mcp login adrails

Full steps, including other clients: Connect an AI client.

What it can do

63 tools, filtered by the permissions you grant and your role in the workspace:

  • Performance and analysis: period and daily performance, metric comparisons, diagnosis, portfolio and business reviews across Meta and Google Ads, with store revenue next to ad metrics.
  • Campaigns: campaign structure, drafts, templates, media, audience and keyword research (including Google Keyword Planner volumes).
  • Changes under approval: budget, status and campaign launches are prepared as proposals with a link to review the exact before and after.
  • Business data: Shopify, Klaviyo, Stripe, Triple Whale and PostHog connections, orders, customers and revenue.
  • Workspace: automations, Knowledge, channels and job status.

Every tool, its inputs and its access level: MCP tools reference.

Try

  • Which Meta and Google Ads campaigns wasted the most spend last week, and why?
  • Compare blended ROAS this month with last month, using Shopify revenue.
  • Prepare a 20% budget increase on my best campaign. Nothing changes until you approve the link it returns.

Security

A tool call runs no Adrails AI model, so it uses no agent credit (except an automation with an agent step), and it never writes to an ad account by itself: a change to Meta or Google Ads needs your explicit approval, signed in to Adrails, before the proposal expires. Connections can be revoked at any time from the MCP page in Adrails. Details: MCP security.

Support

[email protected] · adrails.ai · Built by TBSS Labs, LLC. Not affiliated with Meta Platforms, Inc. or Google LLC.

來源：README.md，提交 b565b8c

工具

0
工具後設資料尚未被收錄。

版本歷史

1
  1. v1.0.1最新Oct 3, 2026