Available now

Dispatch Tickets for n8n

Give your n8n AI Agent workflows a safe, structured line into your support queue — read, triage, create, update, and reply to tickets, over the Model Context Protocol.

What it is

n8n can talk to Dispatch Tickets over the Model Context Protocol (MCP) — an open standard for connecting AI assistants and automation platforms to external tools. Our hosted MCP server exposes your ticketing workspace as a set of safe, well-described tools that n8n's AI agents can call.

There's nothing to install and nothing to publish. You point n8n's MCP Client Tool node at our server URL, authenticate with a Dispatch Tickets API key, and the ticketing tools become available to your agent. Because the key is multi-tenant, one connection can reach every brand it has access to.

Server URL

https://mcp.dispatchtickets.com/mcp

Transport: Streamable HTTP · Auth: Authorization: Bearer sk_live_…

What it can do

The server ships with ten tools, split between reading your tickets and taking action on them.

Read your tickets

list_brands

List the brands (workspaces) your key can access.

list_tickets

List tickets, filtered by status, priority, assignee, or date.

search_tickets

Full-text search across a brand's tickets.

get_ticket

Fetch a single ticket and its details.

list_comments

Read a ticket's comments — public replies and internal notes.

get_ticket_stats

Get ticket counts per status for a brand.

Take action

create_ticket

Open a new ticket in a brand.

update_ticket

Change status, priority, assignee, category, tags, or fields.

add_internal_note

Add a private note for your team. Sends nothing to the customer.

reply_to_customer

Post a public reply that emails the customer.

One tool sends email. reply_to_customer posts a public reply that is emailed to the customer. Everything else is either read-only or internal to your team — add_internal_note sends nothing. In automated flows, gate reply_to_customer behind a human-approval step if you don't want the agent emailing customers on its own.

What to build

A few things teams wire up with n8n and Dispatch Tickets:

Auto-triage: when a ticket comes in, have the agent read it, set a priority, and route it to the right category.
Draft-and-review: generate a suggested reply for a new ticket and post it as an internal note for an agent to approve.
Daily digest: on a schedule, pull ticket stats per status and send your team a Slack summary.
Cross-system sync: create a Dispatch ticket whenever a row is added in Airtable, a form is submitted, or an alert fires.

How to connect

You'll need a Dispatch Tickets account and an API key. If you don't have one yet, it's free to start — create an account, add a brand, then grab a key from API Keys in your dashboard.

In your workflow, add an AI Agent node (or open an existing one).
Add an MCP Client Tool node and connect it to the agent's tools input.
Set the endpoint to the server URL below and the transport to HTTP Streamable.
Under Authentication, add a Header credential named Authorization with the value Bearer sk_live_… (your Dispatch Tickets API key).
Choose which tools the agent may use, then run the workflow — the ticketing tools are now available to your agent.

Server URL

https://mcp.dispatchtickets.com/mcp

Security & permissions

Scoped to your key. The connection can only touch the brands and tickets your API key already has access to — nothing more.
Nothing sends silently. The only customer-facing action is reply_to_customer. In automated flows, put a human-approval step in front of it if you want the final say.
Per-request auth. One hosted server serves many tenants; your key travels with each request and is never shared across workspaces.
Revoke anytime. Remove the credential in n8n, or revoke the API key in your dashboard, and access stops immediately.

Other hosts

The same MCP server works everywhere. Point any MCP-capable host at the server URL:

Claude — Desktop, Claude.ai, and Claude Code — our featured connector.
ChatGPT / OpenAI — Pass the server URL and header as an MCP tool in the Responses API.
Make — MCP module inside an AI-agent scenario.