Giriş Yap
Poyais One'a hoş geldiniz.

Hesabınız yok mu? Kayıt olun.

Giriş yaparak Kullanım Şartları'nı kabul edersiniz.

Developers

Poyais One MCP Connector

A read-only Model Context Protocol server for the Poyais One advertising & analytics reporting platform. Connect an AI assistant to your Google Ads, Meta Ads, GA4, Shopify and Pixa (e-commerce/ERP) data — no write or spend actions are exposed.

Server URL
https://app.poyaisone.com/mcp/

Authentication is OAuth 2.0 (authorization code flow with PKCE, plus Dynamic Client Registration) — no manual API key needed. Access tokens expire after 1 hour and refresh tokens after 30 days; you can revoke access at any time from your Poyais One account settings.

Connecting

1
Add the connector
In your MCP client (e.g. Claude), add a custom connector with the server URL below. The client discovers OAuth support automatically.
2
Sign in
You're redirected to a Poyais One sign-in and consent screen. Log in with your existing Poyais One account — the same one you use for the dashboard.
3
Approve read-only access
Review the requested scope and approve. No password or credential is ever shared with the MCP client itself; only a short-lived, revocable access token is issued.
4
Query your data
Call list_clients first to see which accounts you can access, then pass a client_id into any other tool.

Available tools (read-only)

list_clientsLists the agency client accounts your account can access, with flags for which ad/analytics platforms are connected.
ga4_overviewGoogle Analytics 4 summary metrics for a client: sessions, users, new users, bounce rate, average session duration, page views.
google_ads_overviewGoogle Ads account summary: spend, clicks, impressions, conversions for a client and date range.
google_ads_campaignsGoogle Ads campaign-level performance breakdown for a client and date range.
meta_ads_overviewMeta Ads account summary: spend, results, ROAS for a client and date range.
meta_ads_campaignsMeta Ads campaign-level performance breakdown for a client and date range.
shopify_overviewShopify store order and revenue summary for a client and date range.
pixa_overviewPixa e-commerce/ERP summary across all sales channels: total revenue, order count and average order value, with % change vs. the previous period.
pixa_ordersPixa orders for a client, filterable by status and date range, with human-readable status, shipping and payment labels.
pixa_productsPixa product catalog (name, SKU, barcode, price, cost, stock), with free-text search on name/SKU/barcode.
pixa_low_stockProducts whose stock has fallen below their critical threshold.
pixa_customersPixa customers with order count and total spend, sortable by spend, order count or most recent.
pixa_pending_questionsUnanswered customer questions from marketplaces. Read-only: it cannot post answers.
pixa_listingsMarketplace listing status per product (price, listing stock, matched/closed flags), filterable by marketplace and status.
pixa_shipment_problemsUndelivered or problematic shipments (order number, carrier, tracking number, shipment status).
pixa_anomaliesOpen issues Poyais One found in the Pixa data: failed marketplace transfers, unmatched listings, below-cost prices, critical stock, unanswered questions, shipment problems — grouped and ranked by severity.
list_tasksLists tasks (Görevler) tracked for a client, optionally filtered by status.
list_notesLists notes (Notlar) saved for a client.
list_alertsLists active anomaly/alert rules (Uyarılar) configured for a client.

Scope & data handling

  • Strictly read-only. No tool can publish a campaign, spend budget, create/delete a task, note or alert, change a price or stock level, or answer a marketplace question.
  • Every call is scoped to the client accounts the authenticated user already has access to in the Poyais One dashboard — the same membership rules as the web app.
  • Calls are rate-limited per token and logged for audit purposes.
  • No data accessed through this connector is used to train any AI or machine-learning model, sold, or used for ad/audience targeting.

See our Privacy Policy for full details on how account and integration data is processed, and our Contact page for support.