Google Sheets (gsheets_api.py)

Read and surgically edit spreadsheets via https://sheets.googleapis.com/v4/. gdrive_api.py read (see drive.md) still exports a Sheet as CSV; the commands here read structure, target A1 ranges, and write cells. <spreadsheet_id> is the Drive file id of the Sheet.

python .opencode/skills/google-drive/gsheets_api.py <command> [args]

Read structure and values

python gsheets_api.py get <spreadsheet_id> [--fields ...]
python gsheets_api.py values <spreadsheet_id> "Sheet1!A1:C10" [--formulas]

get lists the sheets (tabs) with their sheetId and grid size. values takes A1 notation; a bare sheet title reads the whole tab.

Create a spreadsheet (write)

python gsheets_api.py create --title "Budget" [--sheet "Q1" --sheet "Q2"]

To build a Sheet from a produced CSV instead, use gdrive_api.py upload --convert-to application/vnd.google-apps.spreadsheet.

Write cell values (write)

python gsheets_api.py update-values <spreadsheet_id> "Sheet1!A1" '[["a","b"],[1,2]]'
python gsheets_api.py append-values <spreadsheet_id> "Sheet1" '[["new","row"]]'
python gsheets_api.py clear-values <spreadsheet_id> "Sheet1!A2:B10"

Values are a JSON array of row arrays (--file values.json instead of inline). Input is parsed like UI entry (numbers, dates, =SUM(A1:A5) formulas); pass --raw-input to store strings verbatim. append-values adds rows after the last row of the table that contains the given range.

Structural edits: raw batchUpdate (write)

python gsheets_api.py batch-update <spreadsheet_id> '[<request>, ...]'
python gsheets_api.py batch-update <spreadsheet_id> --file requests.json

The escape hatch for everything else — a JSON array of Sheets API request objects (e.g. addSheet, deleteSheet, repeatCell, mergeCells, updateBorders, autoResizeDimensions). Requests that target a tab need its numeric sheetId from get (not the tab title).

Output

get / create return {"ok": true, "spreadsheet": {...}}. values and the write commands return {"ok": true, "data": {...}}.