GSC MCP Server

Google Search Console MCP Server — a TypeScript Model Context Protocol server that provides tools for interacting with the Google Search Console API through AI assistants like Claude Desktop.

This is a TypeScript port of the mcp-gsc Python server, fully compatible with the MCP specification.

Features

• 21 MCP Tools covering all aspects of Google Search Console

• Dual Authentication — OAuth 2.0 (browser login) and Service Account

• Cross-platform — Windows, macOS, Linux

• STDIO and SSE transport support

• Zero-config fallbacks — sensible defaults for all settings

• Safety-first — destructive operations disabled by default

Related MCP server: gsc-mcp

Tools Overview

Prerequisites

• Node.js 18+ (LTS recommended)

• A Google Cloud project with the Search Console API enabled

• At least one authentication method set up (see below)

Installation

# Clone the repository
git clone <repo-url> gsc-mcp
cd gsc-mcp

# Install dependencies
npm install

# Build
npm run build

Authentication Setup

You must configure one of two authentication methods:

Option 1: OAuth 2.0 (Recommended)

Best for individual users. Opens a browser window for Google login.

1. Create a Google Cloud project at console.cloud.google.com

2. Enable the Google Search Console API

3. Create OAuth 2.0 Client ID credentials (Desktop application type)

4. Download the client secrets JSON file

Set the environment variable:

# Windows (PowerShell)
$env:GSC_OAUTH_CLIENT_SECRETS_FILE = "C:\path\to\client_secrets.json"

# macOS / Linux
export GSC_OAUTH_CLIENT_SECRETS_FILE="/path/to/client_secrets.json"

Or place the file as client_secrets.json in the project source directory.

On first use, the server will automatically open your browser for authentication. No manual token management needed.

Option 2: Service Account

Best for server/automation use cases.

1. Create a service account in your Google Cloud project

2. Download the JSON key file

3. Add the service account email as a user in GSC for each property

# Windows (PowerShell)
$env:GSC_CREDENTIALS_PATH = "C:\path\to\service_account_credentials.json"

# macOS / Linux
export GSC_CREDENTIALS_PATH="/path/to/service_account_credentials.json"

Claude Desktop Configuration

Edit claude_desktop_config.json:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

• Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "gsc": {
      "command": "node",
      "args": [
        "C:\\path\\to\\gsc-mcp\\build\\index.js"
      ],
      "env": {
        "GSC_OAUTH_CLIENT_SECRETS_FILE": "C:\\path\\to\\client_secrets.json"
      }
    }
  }
}

For macOS/Linux, use forward slashes and the export-style environment configuration.

Restart Claude Desktop after saving. You should see the hammer icon in the input box.

Environment Variables

Running with SSE Transport

# Windows (PowerShell)
$env:MCP_TRANSPORT = "sse"
$env:MCP_HOST = "0.0.0.0"
$env:MCP_PORT = "3001"
npm run start

# macOS / Linux
MCP_TRANSPORT=sse MCP_HOST=0.0.0.0 MCP_PORT=3001 npm run start

Development

# Install dependencies
npm install

# Build TypeScript
npm run build

# Run (stdio)
npm run start

Project Structure

gsc-mcp/
├── src/
│   ├── index.ts    # MCP server entry point + all 21 tool registrations
│   └── auth.ts     # Google authentication (OAuth 2.0 + Service Account)
├── build/          # Compiled JavaScript output
├── package.json
└── tsconfig.json

Credits

Original Python MCP-GSC server created by Amin Foroutan.

License

ISC