HoopAI MCP + n8n

Use n8n’s native MCP Client node to call any of HoopAI’s 36 MCP tools directly inside your workflows — no custom code or API wrappers needed.

Requires n8n v1.104 or later. The MCP Client node with HTTP Streamable transport was introduced in that release. Update before proceeding.

Prerequisites

n8n v1.104+
A HoopAI Private Integration Token (see API Keys)
Your HoopAI Location ID (see Finding your Location ID)

Step 1: Add the MCP Client node

In your n8n workflow, search for and add the MCP Client node.

n8n MCP Client node — Add the MCP Client node from the node panel

Step 2: Configure the connection

Set the following fields in the MCP Client node:

Field	Value
MCP URL	`https://services.leadconnectorhq.com/mcp/`
Transport	HTTP Streamable
Auth type	Header-based
Header key	`Authorization`
Header value	`Bearer YOUR_API_KEY`

Step 3: Pass your Location ID

You have two options for providing the Location ID: Option A — Add as a second header (recommended for single-account workflows): In the MCP Client node headers, add:

Key: locationId
Value: YOUR_LOCATION_ID

Option B — Set it in the LLM system prompt (best for multi-account workflows): If you’re using an AI Agent node before the MCP Client, add this to the system prompt:

The location id is YOUR_LOCATION_ID

The MCP server will extract it from context automatically.

Step 4: Select a tool and run

Once connected, the MCP Client node will fetch the list of available tools from HoopAI. Select the tool you want to call, provide the required parameters, and execute. Example — Get recent contacts:

{
  "tool": "contacts_get-contacts",
  "parameters": {
    "limit": 10,
    "sortBy": "date_added",
    "sortDirection": "desc"
  }
}

Example workflow

A common pattern: AI Agent + MCP Client for conversational automation.

Webhook trigger
  → AI Agent node (with system prompt + locationId)
    → MCP Client node (HoopAI tools)
      → Set node (format response)
        → Respond to Webhook

The AI Agent node receives a natural language request, decides which MCP tool to call, and passes structured parameters to the MCP Client node. System prompt template for the AI Agent:

You are a HoopAI assistant. You have access to MCP tools for contacts,
conversations, calendars, opportunities, payments, social media, and more.

The location id is {{ $json.locationId }}

Always confirm before creating or updating records.
Return results in a clear, structured format.

Multi-account workflows

For workflows that handle multiple HoopAI accounts, pass the Location ID dynamically rather than hardcoding it:

Accept locationId as a parameter in your webhook or form trigger
Store it in a workflow variable or pass it as {{ $json.locationId }}
Inject it into the AI Agent system prompt dynamically:

The location id is {{ $('Webhook').item.json.body.locationId }}

This lets a single workflow serve multiple accounts without duplication.

Troubleshooting

Error	Cause	Fix
`401 Unauthorized`	Invalid or expired token	Regenerate your Private Integration Token
`403 Forbidden`	Missing scope	Add the required scope in Private Integrations settings
`400 Bad Request`	Missing locationId	Add locationId as a header or in the AI Agent system prompt
Tools not loading	n8n version too old	Update to n8n v1.104+
Empty tool list	Wrong MCP URL	Verify URL is `https://services.leadconnectorhq.com/mcp/`

HoopAI MCP

Docs for AI

Prerequisites

Step 1: Add the MCP Client node

Step 2: Configure the connection

Step 3: Pass your Location ID

Step 4: Select a tool and run

Example workflow

Multi-account workflows

Troubleshooting

Next steps

All 36 MCP tools

Build a multi-agent system

​Prerequisites

​Step 1: Add the MCP Client node

​Step 2: Configure the connection

​Step 3: Pass your Location ID

​Step 4: Select a tool and run

​Example workflow

​Multi-account workflows

​Troubleshooting

​Next steps

All 36 MCP tools

Build a multi-agent system

Prerequisites

Step 1: Add the MCP Client node

Step 2: Configure the connection

Step 3: Pass your Location ID

Step 4: Select a tool and run

Example workflow

Multi-account workflows

Troubleshooting

Next steps