Windsor.ai MCP connector for AI agents

Tools your marketing agent reaches for on Windsor.ai, scoped per user.

CALL ANY TOOL

Query marketing and business data across 350+ connectors, manage fields and exports, and run write actions with OAuth-scoped calls.

windsoraimcp_contact_windsor

Contact Windsor.ai

Send feedback, a support request, or a feature request and get a reference id for follow-up.

Parameters

Name

Type

Required

Description

category

string

Required

Type of inquiry: "feedback" for general impressions or complaints, "support" for technical issues or unexpected behavior, "feature_request" for new features or improvements.

message

string

Required

Full message body. Include all relevant context, for example: which tools were called and in what order; connector IDs (e.g. "facebook", "google_ads"); field names and account IDs; date ranges or presets used; what you were trying to achieve; actions that did not work or were missing; what went wrong - error messages, missing or unexpected data, wrong numbers; expected vs actual behavior; steps to reproduce; or, for feature requests, the use case and the desired outcome.

subject

string

Required

Short one-line title describing the inquiry.

windsoraimcp_create_custom_field

Create custom field

windsoraimcp_create_destination_task

Create destination task

windsoraimcp_execute_action

Execute action

windsoraimcp_get_connector_authorization_url

Get connector authorization URL

windsoraimcp_get_connector_connect_info

Get connector connect info

windsoraimcp_get_connectors

List connectors

windsoraimcp_get_current_user

Get current user

windsoraimcp_get_custom_fields

List custom fields

windsoraimcp_get_data

Get data

Retrieve rows from a connector using field ids discovered by get_fields.

Parameters

Name

Type

Required

Description

connector

string

Required

Connector ID, e.g. "facebook" for Meta Ads (Facebook & Instagram Ads), "google_ads" for Google Ads. Call get_connectors if unsure or on error.

fields

string

Required

Field IDs to retrieve, e.g. ["campaign", "date", "spend"]. Must come from get_fields - call it first, do not guess field names.

accounts

string

Optional

Optional account IDs, e.g. ["1234567890"]. Default: all accounts. Pass a subset to keep the response small.

date_filters

string

Optional

Per-table date columns used instead of the defaults, e.g. {"orders": "created_at"}. Required for warehouse connectors (mysql, postgresql, redshift, mongodb, snowflake, big_query) when filtering by date.

date_from

string

Optional

Start date, e.g. "2025-09-01".

date_preset

string

Optional

Predefined range: "last_Xd", "last_Xw", "last_Xm", "last_Xy" (last X days/weeks/months/years), "last_year", "last_2years", "this_month", "this_year". Append "T" to include today - valid only as "last_XdT", "last_yearT", "last_2yearsT", "this_monthT", "this_yearT". E.g. "last_7d", "this_monthT".

date_to

string

Optional

End date, e.g. "2025-09-07".

filters

string

Optional

Filter conditions. A condition is ["", "", ] or the equivalent dict {"field": ..., "operation": ..., "value": ...}; combine with "and"/"or" strings; nesting allowed. Operators: eq, neq, gt, gte, lt, lte, contains, ncontains, null, notnull, in (value is a JSON-array-encoded string of exact alternatives, e.g. ["campaign", "in", "[\"Summer\",\"Autumn\"]"]). E.g. [["spend", "gt", 100]], [{"field": "spend", "operation": "gt", "value": 100}] or [[["campaign", "eq", "Summer"], "or", ["spend", "eq", 10]], "and", ["clicks", "notnull", null]]. May also be passed as a single JSON-encoded string of the same structure - required for clients that cannot send nested arrays (e.g. Microsoft Copilot Studio / PowerFx): "[[\"spend\", \"gt\", 100]]".

options

string

Optional

Connector options, e.g. {"attribution_window": "7d_view,1d_click"}.

windsoraimcp_get_destination_setup_info

Get destination setup info

windsoraimcp_get_destination_tasks

List destination tasks

windsoraimcp_get_destinations

List destinations

windsoraimcp_get_fields

Get fields

windsoraimcp_get_options

Get options

windsoraimcp_get_subscription_url

Get subscription URL

windsoraimcp_get_windsor_login_url

Get Windsor.ai login URL

windsoraimcp_list_actions

List actions

Build your Agent

The same OAuth vault and scoped tool list across LangChain, OpenAI, Anthropic, and Google ADK.

import { ScalekitClient } from "@scalekit-sdk/node";
import { DynamicStructuredTool } from "@langchain/core/tools";
const sk = new ScalekitClient(
  process.env.SCALEKIT_ENVIRONMENT_URL,
  process.env.SCALEKIT_CLIENT_ID,
  process.env.SCALEKIT_CLIENT_SECRET,
);
const page = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["windsoraimcp"], toolNames: [
    "windsoraimcp_get_connectors",
    "windsoraimcp_get_fields",
    "windsoraimcp_get_data",
  ]}, pageSize: 100,
});

import { ScalekitClient } from "@scalekit-sdk/node";
import OpenAI from "openai";
const sk = new ScalekitClient(
  process.env.SCALEKIT_ENVIRONMENT_URL,
  process.env.SCALEKIT_CLIENT_ID,
  process.env.SCALEKIT_CLIENT_SECRET,
);
const page = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["windsoraimcp"], toolNames: [
    "windsoraimcp_get_connectors",
    "windsoraimcp_get_fields",
    "windsoraimcp_get_data",
  ]}, pageSize: 100,
});
const openai = new OpenAI();

import { ScalekitClient } from "@scalekit-sdk/node";
import Anthropic from "@anthropic-ai/sdk";
const sk = new ScalekitClient(
  process.env.SCALEKIT_ENVIRONMENT_URL,
  process.env.SCALEKIT_CLIENT_ID,
  process.env.SCALEKIT_CLIENT_SECRET,
);
const page = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["windsoraimcp"], toolNames: [
    "windsoraimcp_get_connectors",
    "windsoraimcp_get_fields",
    "windsoraimcp_get_data",
  ]}, pageSize: 100,
});
const anthropic = new Anthropic();

import { ScalekitClient } from "@scalekit-sdk/node";
const sk = new ScalekitClient(
  process.env.SCALEKIT_ENVIRONMENT_URL,
  process.env.SCALEKIT_CLIENT_ID,
  process.env.SCALEKIT_CLIENT_SECRET,
);
const page = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["windsoraimcp"], toolNames: [
    "windsoraimcp_get_connectors",
    "windsoraimcp_get_fields",
    "windsoraimcp_get_data",
  ]}, pageSize: 100,
});

Try these prompts

Copy a prompt into your agent. Each one maps to Windsor.ai marketing and business data for the signed-in user.

Cross-channel reporting

Copy the prompt

Copied

How did paid spend trend across Meta and Google Ads last month?

Copy the prompt

Copied

Which campaigns moved CPA the most in the last 14 days?

Copy the prompt

Copied

Compare blended ROAS this month against the same period last month.

Fields and custom metrics

Copy the prompt

Copied

Which fields can I segment by for the facebook connector?

Copy the prompt

Copied

Create a custom field for cost per qualified lead on Google Ads.

Copy the prompt

Copied

List my custom fields and the connectors they belong to.

Exports and account admin

Copy the prompt

Copied

Schedule a weekly export of Meta Ads data to BigQuery.

Copy the prompt

Copied

Which destination tasks run on my account right now?

Copy the prompt

Copied

What plan am I on and where do I upgrade?

SEE HOW AUTH WORKS

Your users connect once. Their Windsor.ai OAuth credentials stay vaulted, every call is checked, and every action is logged.

Authorize

Your user connects

Windsor.ai MCP

once. We tie it to their identity and the meetings they approved — no shared bot account, no org-wide access

Who:

user ‘A’

when:

Once per user

access:

Limited to user

Store

Their

Windsor.ai MCP

token lives in a vault scoped to them. User A's meetings are never reachable by an agent acting for user B, even on the same connection

vault:

encrypted

scope:

per-user

tokens:

auto-refreshed

Resolve

When your agent calls a

Windsor.ai MCP

tool, we fetch the right token server-side. It never touches your agent, never appears in the LLM context, never shows up in your logs

speed:

~40ms

check:

before every call

seen by:

nobody

Audit

Every

Windsor.ai MCP

tool call is logged — who triggered it, which meeting was fetched, what came back. 90 days of history, tied to the user who authorized it

history:

90 days

export:

SIEM-ready

logged:

every call

Test other agents

Reuse the same OAuth vault and audit chain on other marketing and analytics connectors.

No items found.

Why Scalekit

Secure your agent's access. Connectors ship in minutes

01.

Shared tokens break per-user analytics

A shared Windsor.ai credential makes every query look like a bot. Scalekit resolves the credential of the actual user who triggered the agent, never a shared service account.

// shared key
audit → bot_acct

// scalekit
audit → user_abc ✓

02.

Authentication is not authorization

03.

Multi-tenancy is architectural

04.

Windsor.ai today. Others tomorrow.

“Our agents act across Salesforce, Gong, Google Drive, and more, on behalf of every customer. Scalekit behind the scenes meant we can keep adding tools without ever rebuilding how credentials or tool calling work.”

Venu Madhav Kattagoni

Head of Engineering / Von

FAQs

Frequently Asked Questions

Does the agent access Windsor.ai as the user or a shared key?

Per-user resolution. Each tool call uses the signed-in user's Windsor.ai OAuth grant, so every query attributes to that user in the audit log.

Where is the Windsor.ai token stored?

In Scalekit's AES-256 vault, namespaced per tenant. Tokens refresh automatically and revoke from one dashboard without rotating shared secrets.

Can I limit what the agent does in Windsor.ai?

Yes. Filter tool names in listScopedTools (for example windsoraimcp_get_data only) and Scalekit checks that scope before the API call.

What happens when a user revokes access?

The next Windsor.ai tool call fails closed for that user. Other users and connectors are unaffected. The revoke is logged.

What about write actions like exports or campaign changes?

Writes stay governed. Actions such as create_destination_task or execute_action run only when the user's own grant allows them, and each one lands in the same audit trail as reads.

On this page

Tool and actions Framework Compatibility Starter Prompt Auth Lifecycle Try Other Agents Why Scalekit