Connect AI Assistants to ACCELQ MCP

Applies to: ACCELQ Autopilot
Supported assistants: ChatGPT, Claude, Claude Code, VS Code (GitHub Copilot), Cursor, Gemini

Overview

ACCELQ works with AI assistants through the Model Context Protocol (MCP). Once connected, you can ask the assistant in plain language to plan and create tests, run them, and tell you how your runs went.

How access works: The assistant works in ACCELQ as you, with your own permissions. It can only do what you can already do in ACCELQ, and everything it does is recorded like your own actions.

Before you start

Step 1: Copy your MCP server URL

1. In ACCELQ, click your profile picture (top right).
2. Open API Key & Integrations.
3. Under MCP server URL, click the copy icon.

Note: The URL is specific to your company's ACCELQ account. Use it exactly as copied.

Step 2: Add ACCELQ to your assistant

Follow the steps for your assistant below.

ChatGPT

Custom MCP connectors in ChatGPT are available on paid plans through Developer mode.

1. Open Settings → Apps & Connectors → Advanced settings and turn on Developer mode.
2. Back in Apps & Connectors, click Create.
3. Enter a name (for example, ACCELQ), paste the MCP server URL, and choose OAuth as the authentication.
4. Click Create. Sign in to ACCELQ when asked and approve access (see Step 3).
5. In a chat, open the + menu, choose Developer mode, and turn on the ACCELQ connector.

Note: On Business, Enterprise and Edu workspaces, an admin may need to enable Developer mode or add the connector for the workspace first.

Claude (claude.ai and Claude Desktop)

1. Open Settings → Connectors.
2. Click Add custom connector.
3. Enter a name (for example, ACCELQ) and paste the MCP server URL.
4. Click Add, then Connect.
5. Sign in to ACCELQ when asked and approve access (see Step 3).

Note: On Team and Enterprise plans, an organization owner may need to add the connector first.

Claude Code

1. Run:

   claude mcp add --transport http accelq <your MCP server URL>

2. Start Claude Code and run /mcp to sign in.

VS Code (GitHub Copilot, agent mode)

1. Create or open .vscode/mcp.json in your workspace, or run MCP: Add Server from the Command Palette.

2. Add:

   {
     "servers": {
       "accelq": {
         "type": "http",
         "url": "<your MCP server URL>"
       }
     }
   }

3. Click Start above the server entry and sign in when asked.

Cursor

1. Open Settings → MCP → Add new MCP server, or edit ~/.cursor/mcp.json.

2. Add:

   {
     "mcpServers": {
       "accelq": {
         "url": "<your MCP server URL>"
       }
     }
   }

3. Save. Cursor opens a browser window for you to sign in.

Gemini

1. At gemini.google.com, open Connected Apps in the left menu.
2. Under Custom apps, paste the MCP server URL and click Next.
3. Under Additional settings, leave Client ID and Client secret empty. ACCELQ registers Gemini automatically. Click Next.
4. Sign in to ACCELQ when asked and approve access (see Step 3).

Note: Custom connected apps may depend on your Gemini plan or workspace settings.

Other assistants

Any assistant that supports remote MCP servers with sign-in (OAuth) works the same way: add the URL, then sign in.

Step 3: Sign in and approve access

When you connect, your browser opens the ACCELQ sign-in page.

1. Sign in with your usual ACCELQ account, including your company's single sign-on if you use it.

2. Review the access the assistant is requesting:

   Permission	Allows the assistant to
   Read	See projects, tests, runs and results
   Write	Create and update tests
   Execute	Start and stop test runs

3. Click Approve.

Important: If the approval screen lists only Read, remove the connector and add it again. The assistant needs all three permissions to create and run tests.

After an ACCELQ update

Some assistants remember the list of ACCELQ tools from when you first connected. If the assistant says it can't do something this guide says it can, remove the ACCELQ connector and add it again.