MCP Gsheets

What freema/mcp-gsheets does

freema/mcp-gsheets exposes Google Sheets operations to MCP-compatible applications. It lets an agent read spreadsheet content, write updates, manage sheets, apply formatting, work with charts, and configure conditional formatting. Batch operations are also supported for workflows that need to make multiple spreadsheet changes together.

The server is suited to automations where spreadsheet data is both an input and an output. For example, an MCP client can use it to inspect an existing workbook, update cells or ranges, and perform document-management tasks without requiring a separate application-specific connector.

How it works

The package runs as a Node.js MCP server, typically through npx and the package's @latest tag. An MCP client starts the process with a command and environment variables, then exposes its Google Sheets capabilities to the connected model.

Google Cloud supplies the credentials. The Google Sheets API must be enabled in the selected project, and the target spreadsheets must be shared with the service account. Editor access is used in the setup instructions so the server can perform write and management operations.

Setup and configuration

Install Node.js 22 or newer and npm. In Google Cloud, create or select a project, enable the Google Sheets API, create a service account, and generate a JSON key. Share each target spreadsheet with the service account email.

A typical MCP configuration uses:

{
  "mcpServers": {
    "mcp-gsheets": {
      "command": "npx",
      "args": ["-y", "mcp-gsheets@latest"],
      "env": {
        "GOOGLE_PROJECT_ID": "your-project-id",
        "GOOGLE_APPLICATION_CREDENTIALS": "/absolute/path/to/service-account-key.json"
      }
    }
  }
}

Credentials can instead be supplied through GOOGLE_SERVICE_ACCOUNT_KEY as a single-line, escaped JSON value. Another option uses GOOGLE_PRIVATE_KEY and GOOGLE_CLIENT_EMAIL; the private key must retain its PEM markers, with newline characters represented as \\n. GOOGLE_PROJECT_ID is optional when the selected credential value contains the project ID or when using the private-key method.

For local development, clone the repository, run npm install, and build with npm run build. The project also provides an interactive setup command and a local configuration that starts the compiled dist/index.js file directly.

Tools and capabilities

The server supports the following categories of Google Sheets work:

• Read spreadsheet data.
• Write spreadsheet data.
• Manage spreadsheets and sheets.
• Perform batch operations.
• Apply spreadsheet formatting.
• Create or manage charts.
• Configure conditional formatting.

The exact operations are available to the connected MCP client after the server starts successfully and authenticates with Google Cloud.

Limitations and notes

Access depends on Google Cloud configuration and spreadsheet sharing. Enabling the Sheets API alone is not sufficient; each workbook must also grant the service account access. File-based credentials require a readable absolute path, while inline credentials must be correctly escaped for the MCP client's JSON configuration.

The server requires Node.js 22 or newer. Its standard examples use an MCP client configuration that launches a local stdio process with npx; clients using another transport or deployment model need an equivalent way to start and configure the process. The project is licensed under MIT.