Quickstart

Prerequisites

• A Google account with at least one spreadsheet in Drive
• Claude Desktop or Cursor IDE installed

Steps

1. Create your sheets-mcp account. Go to sheets-mcp-xi.vercel.app and sign in. You can use Google (recommended) or email + password. If you use email, connect Google from the dashboard afterward.

2. Connect your Google account. After login, the dashboard shows an amber banner if Google is not connected. Click Connect Google Account. This grants sheets-mcp access to your Google Spreadsheets (read + write) and Google Drive metadata (list + convert files).

Note

sheets-mcp requests drive.readonly + drive.file — not full Drive access. It cannot read arbitrary Drive files outside the spreadsheets scope.

3. Add to Claude Desktop. Open ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows) and add:

   {
     "mcpServers": {
       "sheets-mcp": {
         "url": "https://sheets-mcp-xi.vercel.app/api/mcp"
       }
     }
   }

   Restart Claude Desktop. Claude will open a browser window to sheets-mcp-xi.vercel.app/authorize to complete the OIDC consent flow.

   See the full Claude Desktop guide for a step-by-step walkthrough.

4. Add to Cursor (alternative). In Cursor → Settings → Features → MCP Servers, add:

   {
     "mcpServers": {
       "sheets-mcp": {
         "url": "https://sheets-mcp-xi.vercel.app/api/mcp"
       }
     }
   }

   See the full Cursor guide for details including API key auth.

5. Run your first command. In Claude Desktop or Cursor, try:

   "List all my Google Sheets spreadsheets"

   Claude calls list_spreadsheets and returns your Drive file list. Then try:

   "Read the first 10 rows of Sheet1 in [paste a spreadsheetId]"

   You now have a fully wired AI ↔ Google Sheets pipeline.

---

What's next?

• Authentication — OIDC vs API key — understand both paths.
• Tools Overview — Browse all 14 tools with AX selection guide.
• SQL in Sheets — Aggregate sheet data with Postgres SQL.