SuperAPI AI Directory

google-sheets-mcp

Google Sheets automation tools.

Open website

About this tool

Google Sheets MCP Server is a Python package designed to automate Google Sheets through the Model Context Protocol (MCP). It provides the ability to seamlessly integrate Google Sheets with any MCP-compatible client, enabling users to create, read, update, and delete spreadsheets through a simple secure API layer.

Key Capabilities

  • Full CRUD support for Google Sheets and spreadsheets.
  • Works with Continue.dev, Claude Desktop, Perplexity, and other MCP clients.
  • Secure authentication via a Google service account.
  • Provides a comprehensive set of Google Sheets automation tools.
  • Supports automatic installation via uvx.

Usage

  1. Requirements: Requires the uv package manager (for the uvx command), a Google Cloud project with a service account enabled, and an MCP-compatible client (such as Continue.dev).
  2. Set up a Google service account:
    • Create a Google Cloud project: In the Google Cloud Console, click "Select a project" → "New Project", and enter a project name (for example "my-sheets-automation").
    • Enable the required APIs: In the project, go to "APIs & Services" → "Library", search for "Google Sheets API" and enable it, then search for "Google Drive API" and enable it.
    • Create a service account: Go to "IAM & Admin" → "Service Accounts", click "Create Service Account", enter a service account name (for example "sheets-mcp-service"), click "Create and Continue", skip role assignment, and click "Continue".
    • Generate a JSON key: Click the email of the newly created service account, go to the "Keys" tab → "Add Key" → "Create new key", select the "JSON" format, and click "Create". The JSON file will be downloaded automatically.
    • Extract the required values: Open the downloaded JSON file and record the following values: project_id, private_key_id, private_key, client_email, client_id, client_x509_cert_url.
  3. Configure the MCP client: You can copy the values directly from the Google service account JSON file, and the field names are exactly the same as in the JSON file, with no changes needed. The server also supports the legacy GOOGLE_ prefixed variable names (such as GOOGLE_PROJECT_ID) for backward compatibility.
  4. Share the Google Sheet with the service account: Open the target Google Sheet in a browser, click the "Share" button, enter the service account email (for example your-service@your-project.iam.gserviceaccount.com), assign "Editor" permission, and click "Send" to grant edit permission. After completing the above steps, the MCP client will automatically install and run the package when needed.

Use Cases Suitable for scenarios that require automating Google Sheets operations through an MCP client, such as creating, reading, updating, and deleting spreadsheets in tools like Continue.dev, Claude Desktop, and Perplexity.

Limitations

  • Users need to prepare a Google Cloud project and service account themselves, and complete API enablement and permission configuration.
  • The target Google Sheet needs to be shared with the service account email and granted edit permission.
  • The uv package manager needs to be installed to use the uvx command.
  • This project uses the MIT license; see the LICENSE file for specific details.

Support and Contributions If you have issues, you can report them in GitHub Issues, contact the author with questions via LinkedIn, and pull requests are welcome.