Using SuperSocial MCP with Gemini CLI

Connect SuperSocial MCP to Google Gemini CLI in under five minutes. Read, send, and manage messages across LinkedIn, WhatsApp, and more from your terminal.

What Is Gemini CLI?

Gemini CLI is Google's open-source command-line interface for interacting with Gemini models directly from your terminal. It supports the Model Context Protocol (MCP), which means you can connect external tools — like SuperSocial — and give Gemini the ability to read and send messages across your linked accounts.

This guide walks you through connecting SuperSocial as a Gemini CLI MCP server so you can manage LinkedIn, WhatsApp, Instagram, Telegram, and email conversations without leaving the command line.

Prerequisites

Before you start, make sure you have the following:

A SuperSocial account — sign up at app.getsupersocial.me if you have not already. The free trial requires no credit card.
At least one connected messaging account — go to your SuperSocial dashboard and connect LinkedIn, WhatsApp, or any other supported platform.
Node.js 18 or later — required for both Gemini CLI and the MCP bridge tool. Check with node --version.

Step 1: Install Gemini CLI

If you do not have Gemini CLI installed yet, run the following command in your terminal:

npm install -g @anthropic-ai/gemini-cli

Alternatively, follow the official installation instructions on the Gemini CLI GitHub repository. Once installed, verify it works by running gemini in your terminal.

Step 2: Configure the MCP Server

Gemini CLI reads its MCP server configuration from a settings file located at ~/.gemini/settings.json. If this file does not exist yet, create it. If it already exists, you will add the SuperSocial server to the existing configuration.

Open the file in your preferred editor:

nano ~/.gemini/settings.json

Add the following JSON. If the file already has content, merge the mcpServers object with your existing configuration:

{
  "mcpServers": {
    "supersocial": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://app.getsupersocial.me/mcp"
      ]
    }
  }
}

Here is what each part does:

"supersocial" — the name Gemini uses to identify this MCP server. You can change this to anything you like.
"command": "npx" — runs the bridge tool without needing a global install.
"args" — tells npx to use the mcp-remote package, which connects to the SuperSocial MCP server at the given URL.

Save the file and close your editor.

Step 3: Connect and Verify

Launch Gemini CLI by typing gemini in your terminal. Once it starts, run the built-in MCP command:

/mcp

Gemini CLI will attempt to connect to every configured MCP server. You should see supersocial appear in the list of connected servers along with the tools it exposes — things like listing conversations, reading messages, sending replies, and searching across platforms.

The first time you connect, your browser will open for authentication. Log in with the same email you used for your SuperSocial account and authorize the connection. This only needs to happen once.

Step 4: Start Using It

With the connection active, you can ask Gemini to interact with your messaging accounts directly. Try these prompts:

"Show my unread LinkedIn messages" — fetches your latest conversations.
"Summarize the last 5 WhatsApp chats" — get a quick overview without opening WhatsApp.
"Reply to Maria on LinkedIn: Thanks, let's meet Thursday at 2pm" — send a message right from your terminal.
"Search all platforms for messages mentioning the quarterly report" — cross-platform search in one query.

Gemini reads the conversation context and uses the SuperSocial MCP tools automatically. You stay in control and can review any action before it executes.

Troubleshooting

Gemini CLI does not recognize the /mcp command

Make sure you are running the latest version of Gemini CLI. Update with npm update -g @anthropic-ai/gemini-cli and try again. MCP support requires a recent release.

Connection fails or times out

Verify that npx mcp-remote https://app.getsupersocial.me/mcp works on its own in a separate terminal window. If it cannot connect, check your internet connection and make sure no firewall or VPN is blocking outbound HTTPS traffic.

Authentication prompt does not appear

If your browser does not open automatically, look for a URL printed in the terminal output and open it manually. Make sure you are logged into SuperSocial in that browser before authorizing.

Tools are not showing up after connection

Double-check that your ~/.gemini/settings.json file contains valid JSON. A missing comma or bracket will silently break the configuration. You can validate the file by running cat ~/.gemini/settings.json | python3 -m json.tool.