HubSpot MCP connector for AI agents

HubSpot MCP tools for AI agents

CALL ANY TOOL

12 tools covering get.

hubspotmcp_get_campaign_analytics

Retrieve engagement analytics (sessions, new contacts, influ

Retrieve engagement analytics (sessions, new contacts, influenced contacts) for one or more hubspot campaigns.

Parameters

Name

Type

Required

Description

requests

array

Optional

Required: List of analytics requests to process. Each entry is a full CampaignAnalyticsRequest. Maximum 20 requests per batch.

hubspotmcp_get_campaign_asset_metrics

Retrieve performance metrics for assets (emails, landing pag

hubspotmcp_get_campaign_contacts_by_type

Retrieve paginated contact ids for a campaign filtered by at

hubspotmcp_get_content_analytics_report

Run a content analytics report across hubspot landing pages,

hubspotmcp_get_crm_objects

Fetch multiple crm objects of the same type in a single requ

hubspotmcp_get_organization_details

Retrieve organization-wide details including teams, job titl

hubspotmcp_get_properties

Fetch property definitions for a crm object type, including

hubspotmcp_get_user_details

Return details for the current user including team membershi

hubspotmcp_manage_crm_objects

Create or update crm objects with properties and association

hubspotmcp_manage_landing_page

Read from or write to hubspot landing pages. specify the act

Read from or write to hubspot landing pages. specify the action field to control the operation (list, get, create, update, delete, etc.).

Parameters

Name

Type

Required

Description

action

string

Required

Action to perform. One per call. READ actions (no side effects): "MODULES": list every module on the page draft. Each entry includes the module's field values, its layout position (sectionId, rowIndex, parentCellName when it lives inside a nested column, and cellWidth/cellX on the 12-column grid), and ownership flags (overrideable, globalOwned, globalPartialPath, themePartialOwned). Positions are populated even on a freshly created template page, before any edit. Capture a module's position here BEFORE you REMOVE or MOVE it so you can restore or reposition it precisely. Check ownership before REMOVE/MOVE: true global partials (header/footer) and global widgets are refused, but modules inside theme dnd_partial sections are removable and movable. "MODULE_TYPES": list available module types for this page's theme (call before INSERT). "MODULE_DEF": full field schema for a module type (requires modulePath). The response also includes the curated guide in moduleGuide when one exists, so a separate MODULE_GUIDE call is usually unnecessary. "MODULE_STYLES": read the layout styles (padding, margin, background, visibility) for a module (requires moduleId). "MODULE_GUIDE": (rarely needed — MODULE_DEF already returns the same guide inline) curated markdown for the modulePath covering required fields, common traps, and a minimal INSERT body. Errors if no guide exists; check MODULE_TYPES.hasGuide first. "REVISIONS": list recent page revisions for undo. "TEMPLATES": list landing page templates available in the portal (contentId not required). "FORMS": list forms available in the portal for embedding via the @hubspot/form module (contentId not required). Returns guid, name, portalId, formType, isPublished, createdAt, updatedAt. "BRAND_KIT": fetch the portal's primary brand kit (colors, fonts, logos) when brand-consistent content would help — e.g. picking a CTA color, a hero font, or a logo to drop into a header (contentId not required). WRITE actions (mutate the same shared draft — issue them ONE AT A TIME, never in parallel): "CLONE": duplicate an existing page (requires contentId of the source page; optional pageName and slug for the clone). Returns a new contentId. "CREATE_FROM_TEMPLATE": create the page from a template (requires pageName and templatePath; optional slug). MUST be preceded by a proposal card (call render_landing_page_ui directly with a "propose" target — there is no separate write step to propose) and an explicit user confirmation IN CHAT — never the first step of creating a page, and choosing a template is NOT confirmation. Returns contentId. "SET_MODULE_FIELDS": verbatim field patch on one module (requires contentId, moduleId, fieldOverridesJson). "SET_MODULE_STYLES": set layout styles (padding, margin, background, visibility) on one module (requires contentId, moduleId, stylesJson). Merge semantics — only supplied fields are updated. Call action=MODULE_STYLES first to see current values. "SET_SECTION_STYLES": set the background, padding, margin, or full-width of the ENTIRE SECTION band a module sits in (requires contentId, moduleId, stylesJson; moduleId is any module inside the band, used as an anchor). This is the correct way to do "make this section's background " — do NOT inject CSS via SET_METADATA headHtml for section backgrounds. Merge semantics. Page-scoped bands — including theme dnd_partial sections — CAN be styled; only true global sections (header/footer/global widgets) are refused. "INSERT": add a new module (requires contentId, modulePath; then EITHER anchorModuleId+anchorPosition for anchor-based placement, OR sectionId+rowStart+columnStart+columnsSpanned for grid-coord placement; optional fieldOverridesJson). "REMOVE": delete one module (requires contentId, moduleId). If it was the last module in its section band, the now-empty band — background and padding included — is collapsed too, so no empty "Drop modules here" ghost band is left behind. "REMOVE_SECTION": delete the ENTIRE section band a module sits in — the whole top-level row, every module in it, AND the band's own background/padding styles (requires contentId, moduleId; moduleId is any module inside the target band, used as an anchor). Page-scoped sections — including theme dnd_partial sections like a logo bar — CAN be removed; only true global sections are refused. "MOVE": reposition module (requires contentId, moduleId; optional sectionId, rowStart, columnStart). "SET_METADATA": update page metadata — pageName, htmlTitle, slug, metaDescription, featuredImageUrl, language, headHtml (requires contentId; all metadata fields optional). "PUBLISH": publish draft (requires contentId; confirm with the user first). "RESTORE_REVISION": roll back to revision (requires contentId, revisionId; confirm first).

anchorModuleId

string

Optional

Anchor-based placement: moduleId of an existing module that the new INSERT should land relative to. When provided, omit rowStart / columnStart / columnsSpanned — the backend resolves the target row from the anchor and auto-redistributes column widths across siblings. Pair with anchorPosition.

anchorPosition

string

Optional

Anchor placement direction. Used with anchorModuleId on INSERT. Values: "BEFORE" (new module on a new row immediately above the anchor's row), "AFTER" (new row immediately below), "IN_ROW_WITH" (sibling cell appended to the right of all existing cells in the anchor's row — siblings give up width), "IN_ROW_BEFORE" (sibling cell positioned to the LEFT of the anchor within the same row), "IN_ROW_AFTER" (sibling cell positioned to the RIGHT of the anchor within the same row, regardless of any cells beyond the anchor), "START_OF_SECTION" (first row of anchor's section), "END_OF_SECTION" (last row of anchor's section), "STACK_ABOVE" (wrap the anchor in a column and place the new module ABOVE it as a stacked row), "STACK_BELOW" (same, stacked BELOW). STACK_* builds a "unified section" — both modules share ONE wrapper cell as separate rows (the shape the editor's drag-to-stack creates); use it instead of BEFORE/AFTER when the two modules should read as one visual group rather than independent section rows. STACK requires a module anchor inside a section (not a top-level section id).

availableForNewContent

boolean

Optional

When action=TEMPLATES: filter to templates available for new page creation. Omit to return all templates.

columnsSpanned

integer

Optional

Width in columns on the 12-column grid. Required for INSERT when using legacy grid-coord placement. Full width=12, two halves=6 each, three thirds=4 each. Omit when using anchor-based placement (anchorModuleId + anchorPosition) — width is computed automatically by stealing from siblings.

columnStart

integer

Optional

1-based column position on a 12-column grid. Required for INSERT (grid-coord placement). Optional for MOVE. columnStart=1 is leftmost.

hubspotmcp_query_crm_data

Query hubspot crm data using sql with hubspot-specific exten

hubspotmcp_render_landing_page_ui

Display the landing page card (preview image and open-in-edi

Build your Agent

Same auth pattern across every framework.

from langchain_mcp_adapters.client import MultiServerMCPClient
from scalekit import ScalekitClient

client = ScalekitClient(env_url=ENV_URL, client_id=CLIENT_ID, client_secret=SECRET)
token = client.agent.get_token(user_id="user_id", connector="hubspotmcp")

mcp = MultiServerMCPClient({
    "hubspotmcp": {
        "url": "https://mcp.scalekit.com/hubspotmcp",
        "headers": {"Authorization": "Bearer " + token}
    }
})
tools = await mcp.get_tools()

import OpenAI from "openai";
import { ScalekitClient } from "@scalekit-sdk/node";

const client = new ScalekitClient({ envUrl, clientId, clientSecret });
const token = await client.agent.getToken({ userId: "user_id", connector: "hubspotmcp" });

const openai = new OpenAI();
// Connect to MCP at https://mcp.scalekit.com/hubspotmcp
// Pass: Authorization: Bearer + token

import Anthropic from "@anthropic-ai/sdk";
import { ScalekitClient } from "@scalekit-sdk/node";

const client = new ScalekitClient({ envUrl, clientId, clientSecret });
const token = await client.agent.getToken({ userId: "user_id", connector: "hubspotmcp" });

const anthropic = new Anthropic();
// Connect to MCP at https://mcp.scalekit.com/hubspotmcp
// Pass: Authorization: Bearer + token

from google.adk.agents import LlmAgent
from scalekit import ScalekitClient

client = ScalekitClient(env_url=ENV_URL, client_id=CLIENT_ID, client_secret=SECRET)
token = client.agent.get_token(user_id="user_id", connector="hubspotmcp")
# Connect to MCP at https://mcp.scalekit.com/hubspotmcp
# Pass: Authorization: Bearer + token

Try these prompts

Paste any prompt into your agent to get started.

Get started

Copy the prompt

Copied

Retrieve usage instructions and guidance for one or more HubSpot MCP tools?

Copy the prompt

Copied

Submit user feedback about the HubSpot MCP connector to HubSpot?

Advanced

Copy the prompt

Copied

Find the most relevant CRM property definitions using keyword-based search?

Copy the prompt

Copied

Display the landing page card (preview image and open-in-editor link) in the chat UI?

SEE HOW AUTH WORKS

User authorises once. Every agent call after uses their token with scope enforcement.

Authorize

Your user connects

HubSpot 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

HubSpot 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

HubSpot 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

HubSpot 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

See the same per-user auth pattern across other connectors.

SALES

Outbound prospecting agent

Build targeted prospect lists with Apollo, enrich with firmographic data, and draft personalised outreach. Runs on a schedule.

Apollo

View agent

GTM

HubSpot to Slack updates agent

Watch HubSpot deal stage changes and post structured updates to the right Slack channel. Reps stop checking the CRM all day.

HubSpot

Slack

View agent

Why Scalekit

Secure your agent's access. Connectors ship in minutes

Other connector libraries treat auth as a demo afterthought.

01.

Shared tokens break per-user analytics

A shared token looks fine in a demo. In production every call looks like a service account. Scalekit resolves the real user credential.

// shared token
 audit → bot_service_account

 // scalekit
 audit → user_abc ✓

02.

Authentication is not authorization

03.

Multi-tenancy is architectural

04.

One connector today. Ten 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 HubSpot as the user or as a shared key?
As the user. Each workspace member authorizes once and Scalekit resolves their credential at request time. Audit logs attribute every action to that user, not a shared service account.

Where is the HubSpot OAuth token stored?
In Scalekit's managed AES-256 token vault, namespaced per tenant. Refresh is automatic. Revocation is a single dashboard action. Tokens never appear in prompts, logs, or LLM context.

Can I limit what the agent is allowed to do in HubSpot?
Yes. Pass a tool name filter to listScopedTools so the marketing agent only sees the subset you authorize. Pre-API-call scope checks block out-of-policy actions before the request reaches HubSpot.

What happens when a user revokes HubSpot access?
The connection is invalidated on the next tool call. Subsequent requests for that user fail closed with a clear error. Other users in the tenant remain unaffected. The event is logged for audit.

Can the agent work across multiple HubSpot portals?
Yes. Each portal is its own connected account with per-user, per-portal vault namespacing. Campaign analytics and CRM reads inherit the user's HubSpot seat permissions.

On this page

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