Connect an AI assistant

Open Claude Desktop on your computer.

Click the settings icon and go to Settings → Connectors.

Click “Add custom connector”.

On Team or Enterprise, only an owner can add it, from Organization settings → Connectors: “Add” → “Custom” → “Web”. Everyone else joins at step 5.

Name it “Teramot”, paste the server URL and finish with “Add”.

Leave the authentication options as Claude detects them — the server publishes its own OAuth details. Only if your version asks for OAuth credentials (under “Advanced settings”, or “Use your own OAuth client”), paste the Client ID below and leave the Client Secret blank.

Click “Connect” on the Teramot connector and sign in with your Teramot account in the browser that opens.

Go back to Claude Desktop and ask it about your Teramot data.

Open your terminal.

Run this command:

claude mcp add --transport http --client-id <client-id> --callback-port 8090 teramot https://mcp.teramot.com/mcp

The browser opens automatically. Complete OAuth authentication and close the window.

Go back to the terminal. You can now use claude to ask about your data.

claude "how many active users did I have this month?"

Open chatgpt.com in your browser and sign in.

Enable developer mode: Settings → Apps → Developer mode, then confirm.

In a chat, open the apps menu (＋) → ChatGPT Plugins and click “Create”.

Paste the server URL.

Click “Connect” and complete OAuth authentication in the window that opens.

Ask ChatGPT to query your Teramot data in any conversation.

Open VS Code (1.116 or later) and sign in with the GitHub account that has Copilot.

Open a chat and switch it to Agent mode.

Create .vscode/mcp.json in your workspace and paste this configuration.

{
  "servers": {
    "teramot": {
      "type": "http",
      "url": "https://mcp.teramot.com/mcp",
      "oauth": {
        "clientId": "<client-id>"
      }
    }
  }
}

Reload the window when VS Code asks you to.

Under Agent mode tools, sign in to “teramot” and complete OAuth in the browser.

Ask Copilot about your Teramot data.

Open gemini.google.com in your browser and sign in.

Go to Settings → Connectors and click “Add MCP Server”.

Give it a name and paste the server URL.

On the “Auth” tab, complete the OAuth fields with these values.

The form forces a Client Secret, but the server uses a public OAuth client — any value works; it is discarded before reaching the identity provider.

Click “Add” and sign in with your Teramot account in the window that opens.

Ask Gemini to query your Teramot data in any conversation.

Open the Teams Developer Portal (dev.teams.microsoft.com) and go to Tools → OAuth client registration → “New OAuth client registration”.

Name it “Teramot” and complete the fields with these values.

The portal requires a Client Secret, but the server uses a public OAuth client — any value works; it is discarded before reaching the identity provider.

Save the registration and copy the “OAuth client registration ID” the portal shows.

In Microsoft 365 Copilot, add an MCP connector named “Teramot” and complete its fields with these values.

Use “Teramot” as the developer name and OAuth 2.0 as the authentication type. In “Reference ID”, paste the registration ID from step 3.

Click “Authorize” and sign in with your Teramot account in the window that opens.

Ask Copilot about your Teramot data in any conversation.

Generate an access key for this client.

On the Connect AI page, enter a name (Name — e.g. cursor-work) and click Generate key. Save this key. It won’t be shown again.

Open your client’s MCP configuration.

In Cursor: Settings → MCP. In v0: Settings → Integrations. In Antigravity: main menu → MCP Servers.

Add this configuration to the JSON file.

{
  "mcpServers": {
    "teramot": {
      "url": "https://mcp.teramot.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_KEY_HERE"
      }
    }
  }
}

Save the file and restart the client to apply the changes.

The client can now access your Teramot data.