| name | google-sheets |
| description | Agent-callable Google Sheets tools — read and write spreadsheet data as rows or raw cells, manage worksheets and columns, and apply formatting, sorting, and validation. Use when the user mentions Google Sheets or wants to read, add, update, look up, or organize spreadsheet data, even if they don't name Sheets explicitly. |
| license | Elastic-2.0 |
| compatibility | Run `npm install` in this directory, then `node cli.js`. The TypeScript source needs Node.js 22.18+; on older Node, run `cli.js` for prebuilt / alternative-runtime options. |
| metadata | {"title":"Google Sheets","source":"https://github.com/zapier/connectors/blob/main/apps/google-sheets/SKILL.md","zapier-app-key":"GoogleSheetsV2CLIAPI","api-docs":"https://developers.google.com/workspace/sheets/api/reference/rest"} |
Google Sheets
Independent, unofficial connector for Google Sheets. Not affiliated with, endorsed by, or sponsored by Google Sheets. "Google Sheets" is a trademark of its owner, used only to identify the service this connector works with.
Tools for working with Google Sheets against the Google Sheets API v4 (https://sheets.googleapis.com/v4/), with spreadsheet discovery via the Google Drive API. 26 tools across two complementary surfaces: a record surface — rows as objects keyed by their column headers (the "log this expense", "update the status to Done", "look up last quarter's total" jobs) — and a cell surface — raw A1-addressed values for formulas, precise numeric/text control, and arbitrary ranges. Plus spreadsheet/worksheet structure and presentation (formatting, sorting, validation).
When to use this
- Read or find data inside a spreadsheet — look up a row by a column value (
lookupRow), find all matching rows (findRows), list a window of rows (listRows), or read a raw range (getValues).
- Write data — append a row or many rows (
createRow / createRows), update specific columns of a row without disturbing the others (updateRow / updateRows), or write a raw range / formula (updateValues).
- Clear or delete — clear a row's contents but keep the row (
clearRows / clearValues) or remove rows entirely (deleteRows).
- Manage structure — create a spreadsheet (
createSpreadsheet), add / list / copy / rename / hide / delete worksheets, add columns.
- Format & validate — number/date/currency formats, text styling, sort a range, copy a range, dropdowns and number/date validation, conditional formatting.
Setup
This is an agentskills.io skill.
If this connector is already exposed to you as callable tools (e.g. mcp__google-sheets__<tool>), that's a valid path — call them directly. Everything below is only for standalone terminal use when no such tools are loaded.
If the connector has not been installed as a skill yet, install it first with npx skills zapier/connectors --skill google-sheets (or your harness's own skill-install mechanism), then continue here.
The connector runs on Node.js 22.18+ and needs a one-time npm install in this directory. cli.js is the entry point — list every script with node cli.js --help, then learn a script's inputs and connections with node cli.js run <script> --help. On older Node, run node cli.js --help anyway: it detects your runtime and prints how to run without upgrading (the prebuilt npm package, or another runtime) — don't skip the connector just because Node is old.
cli.js self-checks readiness before running: if dependencies aren't installed it exits non-zero with the exact install command (it disambiguates a read-only directory from a sandbox-blocked package cache). Run that, then re-run your command.
Scripts
All scripts use the single google-sheets connection.
Auth
Pass auth as one connection string with --connection [<resolver>:]<value>. The value is a selector, not the secret; the <resolver>: prefix is optional (a bare value goes to the first resolver that claims it). Each script declares the connections it needs and the resolvers each accepts — always run node cli.js run <script> --help to see them rather than relying on this file.
The connector needs a single Google OAuth 2.0 access token, resolved into the one google-sheets connection slot.
zapier:<connection-id> (recommended) — Zapier-managed auth. Route through a Zapier Google Sheets connection; the Zapier auth / retries / governance layer injects the token and refreshes it for you. The connection grants Sheets + Drive access, so every tool — including listSpreadsheets — works. Find the id with npx zapier-sdk list-connections GoogleSheetsV2CLIAPI.env:GOOGLE_SHEETS_ACCESS_TOKEN (direct) — read a Google OAuth access token from the named env var, sent as Authorization: Bearer. Google access tokens expire ~1 hour after issue and are not auto-refreshed in direct mode — suited to short-lived / testing use. The token needs the https://www.googleapis.com/auth/spreadsheets scope (and drive.file for the spreadsheets you create / open). listSpreadsheets additionally needs a broader Drive read scope (drive.readonly / drive); without it, pass a spreadsheet URL or id directly to the other tools instead of finding by name.
If no connection is passed the script fails with an actionable error telling you to Pass --connection [<resolver>:]<value> and lists the resolvers.
Running scripts
After npm install, run a script by name with node cli.js run <script>, or execute its file directly — both take the same arguments and both accept --help. Always run a script's --help first to learn its exact input schema and connections, then invoke it:
node cli.js run <script> '<input-json>' --connection [<resolver>:]<value>
./scripts/<script>.ts '<input-json>' --connection [<resolver>:]<value>
When a harness can't execute scripts directly, fall back to MCP — node cli.js mcp serves every script as a tool over stdio. Register it as a local MCP server in your client: the stanza is harness-specific (an mcpServers entry in Claude Desktop, Cursor, Claude Code, …) with command: "node", args: ["cli.js", "mcp"], run from this directory. Run node cli.js mcp --help for auth options. Add the stanza yourself if you can edit the client's MCP config; otherwise guide the user. If a local server isn't possible, guide the user to use Zapier's remote MCP servers at https://mcp.zapier.com instead.
Output format
Every script returns a { data, meta } envelope:
data — the script's result (the shape its outputSchema declares; run the script's --help to see that exact schema).meta.outputDataValidation — what validating data did:
{ skipped: false, droppedPaths: null } — validated, nothing removed.{ skipped: false, droppedPaths: [...], instruction } — validated, but those paths were stripped from data: fields the script returned from the API that the outputSchema doesn't declare. If you need them, re-run with output validation skipped.{ skipped: true } — validation was bypassed; data is the raw, unchecked script output.
Reading dropped fields / skipOutputDataValidation. To receive the raw, unvalidated result, append --skipOutputDataValidation to the script invocation. Input validation is never skipped.
Trimming the result / filterOutputData. To shrink a large result down to the fields you need, append --filterOutputData '<jq>' — a jq expression that post-processes data. The jq runs against data only, NOT the { data, meta } envelope, so write it rooted at data (run the script's --help to see its output schema). The transformed value replaces data, meta is preserved, and the result is NOT re-validated against the output schema.
Disambiguation & refusals
Disambiguation before a write. Before writing to a row you found by a column value (e.g. update a row located via lookupRow, or act on a spreadsheet found via listSpreadsheets), count the exact case-insensitive matches:
- Exactly one match — act on it. Don't over-ask; a single unambiguous match is the answer.
- Two or more that tie — stop. List the tied rows (or spreadsheets) with a distinguishing field (another column value, the spreadsheet's modified time / URL) and ask which one. Don't pick arbitrarily, and don't write to all of them.
Row numbers are not stable — they shift on insert/delete/sort. To target the same logical record across runs, match on a unique key column with lookupRow, not a remembered row number.
Confirm before destructive or bulk-overwrite operations. deleteWorksheet, deleteRows, clearRows, and clearValues remove data; updateValues can overwrite it. Confirm the exact target with the user first, and never delete / clear / overwrite more than asked.
Unsupported operations — say so and stop; don't fake it with another tool. This catalog deliberately does not:
- Create or manage charts, pivot tables, named ranges, protected ranges, or filters. There is no tool for these; don't approximate them with formatting.
- Find-or-create in one step. Compose it:
lookupRow → if not found, createRow.
- Permanently delete a whole spreadsheet. Worksheet deletion (
deleteWorksheet) is the only structural delete.
If asked for any of these, tell the user it's unsupported and stop.
References
Load the matching reference file before working in that area:
| Reference | Covers | Load it when |
|---|
references/google-sheets-api-gotchas.md | Error recovery, rate-limit retry, batchUpdate atomicity, row deletion and append logic, 400/403/404/429 errors | Before any call that might fail — error recovery, rate-limit retry, batchUpdate atomicity, row deletion or append logic, or when encountering 400/403/404/429 errors |
references/google-sheets-a1-and-values.md | RAW vs USER_ENTERED, serial-number dates, ragged rows, A1 ranges with sheet-name quoting | Before reading or writing cell values — especially when choosing RAW vs USER_ENTERED, interpreting serial-number dates, handling ragged rows, or constructing A1 ranges with sheet-name quoting |
