Provides tools for interacting with Google Sheets, allowing AI agents to read and filter spreadsheet data, list sheets, write cell values, append rows, and update rows matching a query.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
MCP server for Google Sheets with read AND write.
Works with any Google spreadsheet — trackers, CRMs, registries, or any Excel-style sheet living on Google.
Read without credentials. Any sheet shared by link is readable (gviz export) — no Google Cloud setup needed to start.
Server-side filtering. query filters rows on the server, so you never dump the whole 500-row spreadsheet into the model's context.
Write by human names. Address columns by header name and rows by content query — not by A1 notation.
update_by_query — the killer feature: "set Status = Rejected in every row containing Acme".
Handles a title row above the header row automatically.
Lightweight: Bun + a single dependency.
Related MCP server: Google Sheets MCP Server
Tool | What it does |
| Read a sheet. Optional |
| List sheets of the spreadsheet (via API when a key is configured). |
| Set a cell: |
| Append a row to the end from a |
| Set |
git clone https://github.com/Ap1-Pie/mcp-google-sheets-rw
cd mcp-google-sheets-rw
bun installReading works out of the box for link-shared sheets. Writing needs a service account:
Open console.cloud.google.com and create a project.
APIs & Services → Library → enable Google Sheets API.
APIs & Services → Credentials → Create credentials → Service account (no role needed).
Open the account → Keys → Add key → JSON → download. Put the file next to the server as service-account.json.
Copy the service account email (looks like xxx@yyy.iam.gserviceaccount.com).
Share your spreadsheet with that email as Editor.
{
"mcpServers": {
"gsheets": {
"command": "bun",
"args": ["/absolute/path/to/mcp-google-sheets-rw/src/gsheets-rw.ts"],
"env": {
"GSHEETS_SPREADSHEET_ID": "1AbC...your-spreadsheet-id",
"GSHEETS_SHEETS": "Sheet1",
"GOOGLE_SERVICE_ACCOUNT_FILE": "/absolute/path/to/service-account.json"
}
}
}
}Env var | Meaning |
| Spreadsheet ID from its URL (required) |
| Comma-separated sheet names; the first is the default |
| Path to the service account JSON (optional; enables writes) |
Never commit service-account.json — .gitignore already excludes it.
Grant the service account access only to the specific spreadsheet it serves.
Reads go through the public gviz export; only share by link what you consider public-readable.
mcp, google sheets, google excel, spreadsheet, read, write, model context protocol, agent, llm
MIT
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Query your Google Sheets as structured JSON: list sheets and tabs, read schemas, filter rows.
Google Sheets via Apify: write Apify datasets or JSON rows to a sheet, or read a sheet as JSON.
Read and write Google Sheets rows via SheetDB APIs; per-API agent permissions, approved deletes.
Write rows to Google Sheets (append, replace, update by key) or read a tab. No OAuth needed.
