Windsor.ai MCP Server — Connect AI Agents to Marketing Data

{
  "mcpServers": {
    "windsor": {
      "url": "https://mcp.windsor.ai/"
    }
  }
}

Overview

Windsor.ai MCP Server is a remote MCP server that exposes Windsor.ai's 325+ data connectors to any MCP-compatible AI client. Use it to pull live marketing performance data, query your data warehouse, build dashboards, calculate ROAS, and combine multi-channel data — all from natural-language prompts in your AI assistant.

Compatible AI Clients

Windsor.ai MCP works with any MCP-compatible client. Per-client setup guides live at windsor.ai/documentation/windsor-mcp/.

• Claude (Desktop, Web, Code) — listed in the Claude connector directory. One-click install · setup guide.
• ChatGPT — install as a ChatGPT app · setup guide.
• Microsoft Copilot — uses the Windsor.ai Power Platform connector. Setup guide.
• Perplexity — setup guide.
• Claude Code — install with claude plugin install windsor-ai. Source; ships slash commands and an analyst agent.
• Cursor — dedicated plugin or paste the standard MCP config below · setup guide.
• Manus — add https://mcp.windsor.ai/ as an MCP server. Setup guide.
• Google Gemini — standard MCP server config (url: https://mcp.windsor.ai/) · setup guide.
• Windsurf, Cline, GitHub Copilot — standard MCP server config (url: https://mcp.windsor.ai/) · generic setup guide.
• n8n, mcp-proxy, and any application built on the MCP open standard.

Available Tools

• get_current_user — Return the authenticated user's profile (username, email).
• get_connectors — List Windsor.ai connectors, their connected accounts, and the write actions each connector supports. Pass include_not_yet_connected=True to also see connectors the user can set up.
• get_connector_authorization_url — Get a browser link to connect or authorize a connector (OAuth or manual).
• get_options — List available fields, date filter columns, and connector-specific options.
• get_fields — Get descriptions, types and tables for specific fields.
• get_data — Run a query with date ranges, accounts, options, and nested filters.
• list_actions — Discover the write actions a connector exposes (pause / enable campaign, set campaign or ad set budget) and their JSON schemas. Currently available on Meta Ads and Google Ads.
• execute_action — Run a write action against a specific account. Reach for this whenever you would otherwise recommend a platform change; confirm intent with the user first.

Supported Data Sources

325+ marketing, advertising, analytics, e-commerce, CRM, and database sources. The full list is available at /datasources or by calling get_connectors(include_not_yet_connected=True). Highlights:

Advertising

Analytics & Search

Social Organic

CRM & Marketing Automation

E-commerce & Payments

Databases, Warehouses & Files

Use Case Examples

• "Pull last month's Meta Ads spend by campaign and compare to the previous period."
• "Show me Google Ads ROAS by campaign for the last 30 days."
• "Combine Meta, Google and TikTok Ads spend into a unified weekly report."
• "Pull Shopify orders and Klaviyo email engagement so I can attribute revenue."
• "Query my BigQuery warehouse for last quarter's pipeline data."
• "Build a Looker Studio-ready dataset combining GA4 sessions and ad spend."

Setup Instructions

Prerequisites

• An MCP-compatible client (Claude Desktop, Cursor, ChatGPT with developer mode, Manus, Windsurf, etc.).
• A Windsor.ai account — sign up at onboard.windsor.ai.

Server Endpoints

Authentication

OAuth 2.0 (recommended)

The MCP server implements the full MCP OAuth flow. Clients that support MCP OAuth — including Claude, ChatGPT, Cursor and Manus — discover the endpoints automatically and walk the user through a one-time login at Windsor.ai. Dynamic Client Registration is supported.

API key

Pass a Windsor.ai API key as a Bearer token:

Authorization: Bearer <your_windsor_api_key>

Get an API key from your Windsor.ai account at onboard.windsor.ai.

Tool Reference

get_connectors

List Windsor.ai connectors and their connected accounts. By default returns only connectors with at least one connected account. Pass include_not_yet_connected=True to also include connectors the user can set up.

get_connector_authorization_url

Returns a URL the user can open in their browser to set up a connector. Always call get_connectors(include_not_yet_connected=True) first to get the correct connector ID. Works for both OAuth-based and manual-credential connectors.

get_options

For a connector and a set of accounts, return the available fields, the columns that can be used as date filters per table, and connector-specific options (e.g. attribution windows for Meta Ads).

get_fields

Get descriptions, types and tables for specific fields. Field types NUMERIC and PERCENT are metrics (aggregated); other types are dimensions.

get_data

Run a query against a connector. Supports date ranges (date_from/date_to or date_preset), account filtering, connector-specific options, and nested filter conditions.

Filter operators: eq, neq, gt, gte, lt, lte, contains, ncontains, null, notnull, in. Conditions can be nested with "and" and "or".

Warehouse connectors (mysql, postgresql, redshift, mongodb, snowflake, big_query) require date_filters when filtering by date.

Troubleshooting

Authentication failures (401)

• Confirm the Authorization header is Bearer <token> with no quotes.
• For API keys, get one from onboard.windsor.ai.
• For OAuth, clear cached tokens and re-authorize.

"Requested data exceeds size limit"

Narrow the date range, reduce fields, filter by accounts, or add filter conditions to get_data.

"Connector requires specifying date filters"

Warehouse connectors need an explicit date_filters argument when filtering by date.