# Zendesk: how to read it into Maplith

*Helpdesk · also known as Zendesk Support, Zendesk Suite · last checked 2026-10-08 · recipe `maplith://systems/zendesk`*

Support tickets, the people who raise them and their organisations, which show what customers need help with and how fast the team answers.

## Best path

**Recommended: export files.** Zendesk has no official MCP server for reading an account yet, and a small team can have thousands of tickets a month, too many to pass through a chat. The JSON export from Admin Center gives tickets, users and organisations in one go.

Try the paths in this order (ADR-0031):

1. **Your agent fetches it**, with the access your human already gave you. Connectors a vendor or client page confirms: Zendesk plugin in ChatGPT (beta). The customer's agent checks the tools it already has; if it can reach Zendesk, it looks at a few recent tickets, users and organisations, read-only, and reports which fields and ids are in use.
2. **Maplith reads an MCP server your human already has.** No official MCP server confirmed. Zendesk has announced an MCP server, but its own documentation does not yet describe one a customer can connect to; community servers exist but are not Zendesk's. Maplith reads only a server that serves MCP resources and takes a token a person types under Data sources › Systems in the Maplith app, such as the customer-run template for a database (ADR-0028).
3. **Export files**, the fallback that always works: see *Export steps* below.

## Datasets that make a good map

| Dataset | One row is | Key columns | Joins to |
|---|---|---|---|
| Tickets | One support request, with its status, timestamps, group, assignee and custom fields. | ticket id, requester id, organization id, external id, group id, assignee id, status, created at, solved at, custom field values | Users by requester id; Organizations by organization id; shop orders by order number when it is kept in a custom ticket field. |
| Users | One person in Zendesk: an end user who asks for help, or an agent or admin. | user id, email (Maplith tries to mask it), external id, organization id, role | CRM contacts, shop customers and billing customers by email or by external id when the company sets it. |
| Organizations | One customer company that users and tickets belong to. | organization id, external id, domain names, created at | CRM companies and accounting customers by external id or email domain; Tickets and Users by organization id. |
| Ticket comments | One public reply or internal note on a ticket (in the JSON export, not the CSV). | ticket id, author id, public or internal, created at | Tickets by ticket id; Users by author id. |

Columns are named, never their values. Maplith tries to mask personal data such as email addresses, but that is best effort: your human decides what goes in.

## Read-only access

**Who usually has the rights:** An admin. Prefer OAuth with the read scope: Zendesk calls API tokens deprecated, they give broad access to the whole account, and they stop working by 30 April 2027.

**For the key your own agent's client uses, choose only these read permissions:** `read`, `tickets:read`

**Where to click:**

1. Admin Center
2. Apps and integrations
3. APIs
4. OAuth clients
5. Add OAuth client, fill in the fields and click Save
6. When the access token is issued, ask for the read scope only

A key or token made here stays in the tool that reads the system for your human, never in a chat. Maplith asks for one only under Data sources › Systems in the app, and only for an MCP server it can read.

## Export steps

**Format:** JSON (CSV for tickets only) · **Date range:** The last three months

1. One time: the account owner asks Zendesk Customer Support to turn on data exports for the account.
2. In Admin Center, click Account, then Tools, then Reports, and open the Export tab.
3. For JSON, pick a date range, choose tickets, users or organizations, and click Export; repeat for each.
4. For CSV, pick a date range and click Export (tickets only, without comments).
5. Zendesk emails a link to a zip file when the export is ready, which can take minutes to a day.
6. These exports are available on some plans only; the API works on all plans.

## Volume hint

**About 200–3,000 tickets a month** (an estimate). Estimate for a small company, not a measured figure: one support team handling a handful to about a hundred requests a day.

Every row an agent reads passes through its context, about one token per four characters. Above a few thousand rows a dataset, or for a daily refresh, prefer an MCP server or an export over fetching through the chat.

## Always

- Read-only, and never a credential sent to Maplith through the chat.
- The same recent period, a few months, from every system.
- Your human decides what goes in.

## Sources

Written from the vendor's public documentation, last checked 2026-10-08:

- https://support.zendesk.com/hc/en-us/articles/4408886165402-Exporting-ticket-user-or-organization-data-from-your-account
- https://support.zendesk.com/hc/en-us/articles/4408889192858-Managing-access-to-the-Zendesk-API
- https://support.zendesk.com/hc/en-us/articles/4408845965210-Using-OAuth-authentication-with-your-application
- https://developer.zendesk.com/api-reference/introduction/security-and-auth/
- https://www.zendesk.com/newsroom/articles/relate-2026/
- https://help.openai.com/en/articles/20001512-using-zendesk-in-chatgpt-and-codex