Connect your AI (MCP setup)

MCP (Model Context Protocol) is the open standard that lets AI tools work with outside apps. Connecting Banter over MCP means the AI you already use — ChatGPT, Claude, Codex, Gemini, Cursor — can search your notes, read transcripts, check your todos, and file new notes, all from its own chat window, on your own subscription.

What you need

• The server address: https://api.usebanter.ai/mcp
• Your Banter account: sign in with the same Google account you use for Banter when your AI asks you to connect. You do not need to copy a token.
• A test: once connected, ask your AI “List my Banter folders.” If it answers with your folders, you’re done.

Open Settings → Connections in Banter and choose Connect next to your AI. Claude opens a prefilled connection form; ChatGPT and Codex show setup instructions.

Claude

1. In Banter, open Settings → Connections and choose Connect next to Claude, or use Connect Banter to Claude.
2. Review the prefilled name and server address, then add the connector.
3. Click Connect, sign in with the same Google account you use for Banter, and approve access.
4. In a new chat, select the Banter connector and ask: “List my Banter folders.”

The connector is also available in Claude Code when you sign in with the same Claude subscription account. Use /mcp to find it. On Team and Enterprise plans, an organization admin may need to add the connector first.

ChatGPT

1. Open ChatGPT on the web.
2. Open Settings → Security and login and turn on Developer mode.
3. Go to ChatGPT Plugins, select the plus button, and create a plugin named Banter.
4. Enter https://api.usebanter.ai/mcp as the MCP server URL, choose OAuth for authentication, and create the plugin.
5. Open your personal plugins, select Banter, and use the plus button to install it.
6. Return to the ChatGPT homepage and start a new chat.
7. In the prompt box, type @ and select Banter. When ChatGPT asks you to connect, sign in with the same Google account you use for Banter and approve access.
8. Ask: “List my Banter folders.”

If your ChatGPT account or workspace does not show Developer mode or Plugins, the required features may not be available for that account. You can still connect Banter from another MCP client below.

Codex

On the computer where you use Codex, install the Codex CLI if you have not already. Open Terminal and run:

codex mcp add banter --url https://api.usebanter.ai/mcp
codex mcp login banter

A browser window opens for sign-in. Use the same Google account you use for Banter and approve access. Start a new Codex task and ask it to list your Banter folders. This configuration is shared by the Codex desktop app, CLI, and IDE extension on that computer.

If you previously configured Banter with a token, remove that old entry with codex mcp remove banter before running the commands above.

Any other MCP client

In your client’s MCP settings, add a remote HTTP server using https://api.usebanter.ai/mcp. Choose OAuth or the client’s Sign in action and authenticate with your Banter account.

For OAuth-capable clients configured with a JSON file, the server entry usually looks like this; check your client’s documentation for its exact format:

{
  "mcpServers": {
    "banter": {
      "url": "https://api.usebanter.ai/mcp"
    }
  }
}

Your client must support OAuth for remote MCP servers to use this setup.

If it doesn’t connect

• 401 / unauthorized — reconnect and sign in again. In Codex, run codex mcp login banter.
• ChatGPT asks you to reconnect — open the Banter plugin connection and sign in again. You do not need to paste a Banter token into ChatGPT.
• ChatGPT doesn’t show Banter — make sure you created and installed it from your personal plugins, then start a new chat and invoke Banter with @.
• Claude doesn’t show Banter — after adding the connector, you still need to enable it in the chat’s tools menu.
• The AI can’t find something — tell it to list folders first, then search. Like a new colleague, it navigates better once it’s seen the layout.

What now?

See what connected agents can do for the full picture of what your AI can read and write, and MCP recipes for the workflows this unlocks — bulk imports, workspace cleanup, and having your AI write its work into Banter.