GitHub Copilot

How to use Zendesk with GitHub Copilot

You want Zendesk inside GitHub Copilot so it can answer tickets from the knowledge base, triage an overloaded support queue, add updates to waiting customers, and close resolved support tickets — without you copy-pasting between tabs. Here is the whole setup, once.

Create your Zendesk gateway

The gateway is the thing GitHub Copilot connects to, and it is where you decide which Zendesk tools an agent may use. You create it once and every client reuses it.

  1. Create an Arcade account

    Sign up and open the dashboard. The free tier is enough to follow this through.

  2. Create an MCP Gateway and select your Zendesk tools

    In the dashboard, create a gateway and pick the Zendesk tools you want to expose. The gateway is what your client connects to, and it is where tool access is decided.

  3. Copy your gateway URL

    Every client below needs it.

    https://api.arcade.dev/mcp/<YOUR-GATEWAY-SLUG>

Connect the gateway to GitHub Copilot

Heads up

JetBrains IDEs do not support remote MCP servers with Dynamic Client Registration — configure the gateway as “Arcade Headers” in the dashboard first. Setup also differs on Visual Studio, Eclipse and Xcode; check the docs for yours.

  1. In VS Code

    Copilot picks up MCP servers configured in VS Code automatically, so follow the VS Code steps and you are done.

  2. In JetBrains IDEs

    Copilot icon → Open Chat → Agent mode → tools icon → “+ Add More Tools…”, which opens mcp.json.

  3. Add the gateway with Arcade Headers

    {
      "servers": {
        "mcp-arcade": {
          "url": "https://api.arcade.dev/mcp/<YOUR-GATEWAY-SLUG>",
          "requestInit": {
            "headers": {
              "Authorization": "Bearer {arcade_api_key}",
              "Arcade-User-ID": "{arcade_user_id}"
            }
          }
        }
      }
    }

  4. Try it out

    Open Copilot Chat in Agent mode and ask it to use one of your Zendesk tools. The first time it calls a Zendesk tool, Arcade replies with an authorization link. Open it, sign in to Zendesk and approve access, then ask again. You do this once per account.

Get started with these Zendesk prompts

GitHub Copilot can now act on your Zendesk account. Copy a prompt and paste it in.

answer tickets from the knowledge base

Review open Zendesk tickets that have not been updated today. Search the Help Center for an answer to each one, identify which tickets can be answered from existing articles, and return the ticket IDs, suggested replies, and matching article titles. Do not comment yet.

triage an overloaded support queue

List the open Zendesk tickets and summarize the ones that need attention first. For each, return the ticket ID, current conversation summary, customer issue, and recommended next step. Do not change any tickets.

add updates to waiting customers

Find the Zendesk tickets in this queue that need a customer update, read their full conversations, and add a concise comment with the latest status I provide. Return each ticket ID and the exact comment added, and ask for approval before posting if the wording is ambiguous.

close resolved support tickets

Review the Zendesk tickets in this list and identify the ones that are clearly resolved from their conversations. Show me the ticket IDs and a proposed final comment first; after I approve, add the comment and mark only those tickets as solved.

Use tool permissions to let an agent read tickets and search articles without giving it permission to change ticket records. Pair it with Contextual Access policies so comments and ticket resolution require approval.

Ready to actually ship your agent?

Free to start. Ready for enterprise when you are.