Automate Google Sheets Data Operations

Google Sheets is a lightweight integration using standalone OAuth authentication - no MCP server required - that provides full read/write access to Google Sheets through local Python scripts (scripts/auth.py and scripts/sheets.py).

Use this when a task needs to read, search, update, or restructure Google Sheets data through direct scriptable access on a Workspace account, without introducing an MCP server dependency. It requires a Google Workspace account - personal Gmail accounts are not supported.

First-time setup authenticates via a browser flow with auth.py login, with status checked via auth.py status and access revoked via auth.py logout. Read commands run through scripts/sheets.py, which auto-authenticates on first use: get-text returns full spreadsheet content as human-readable text by default, or as CSV/JSON with --format csv/--format json; get-range fetches values from a specific A1-notation range, e.g.:

python scripts/sheets.py get-range SPREADSHEET_ID "Sheet1!A1:D10"

find searches for spreadsheets by title (optionally --limit), and get-metadata returns sheet names, dimensions, and similar structural info. Write commands include update-range (values passed as a JSON 2D array, parsed as USER_ENTERED by default so numbers, dates, and formulas are interpreted, or stored literally with the --raw flag so =SUM(...) is kept as text instead of evaluated), append-rows (adds after the last data row), clear-range (removes values but keeps formatting), and batch-update for advanced operations like cell formatting or merging via the raw Sheets batch-update request format. Either the spreadsheet ID or its full edit URL can be passed - the script extracts the ID automatically from a URL. A1-notation examples include Sheet1!A1:B10 for a bounded range, Sheet1!A:A for a whole column, Sheet1!1:1 for a whole row, and A1:C5 for a range on the first sheet without naming it. The default text output is pipe-separated and human-readable, printing the spreadsheet title and sheet name above the rows; CSV output is standard comma-separated values suitable for further processing; and JSON output nests row arrays under each sheet name.

Uses standalone Google OAuth with no MCP server, storing tokens in the OS keyring - macOS Keychain, Windows Credential Locker, or Linux Secret Service API (GNOME Keyring, KDE Wallet, and similar) - under the service name google-sheets-skill-oauth, with automatic refresh of expired tokens via Google's cloud function.

Automations or agents on a Google Workspace account that need scriptable spreadsheet access - extracting data as text, CSV, or JSON, updating or clearing specific ranges, appending rows, or applying batch formatting - without standing up an MCP server.