Skip to content
Remote MCP · OAuth · 13 tools
Google Sheets logo

Google Sheets MCP Server: Secure AI Agent Access to Google Sheets

Connect Claude, ChatGPT, and Cursor to Google Sheets through PortEden's remote MCP server. 13 tools, OAuth, and a data firewall that controls every read and write.

Free tier · No credit card required

What is the Google Sheets MCP server?

The Google Sheets MCP server lets AI agents read and write your spreadsheets through the Model Context Protocol. Instead of pasting data into a chat or wiring up a brittle script, you connect a client like Claude, ChatGPT, or Cursor to one remote endpoint at https://mcp.porteden.com/google-sheets and the agent works with Sheets directly.

PortEden's server exposes all 13 Google Sheets tools, from range reads and searches to writes, appends, and new spreadsheet creation. What makes it different from a generic Google Sheets MCP is the data firewall in front of those tools: PortEden inspects every tool-call request, applies your access policy, and redacts sensitive cell values in the response before the agent ever sees them.

The result is secure Google Sheets access for AI agents. You decide whether a client gets read-only or read-write access, scope it to a single spreadsheet or range, mask sensitive columns, and review every call in an audit log. OAuth handles authentication, so there is no static key to leak.

All 13 Google Sheets MCP tools

Every tool can be allowed or denied per client. Access levels are enforced at the firewall on each call.

ToolAccessWhat it does
google_sheets_readreadRead a cell range from a spreadsheet, for example Sheet1!A1:D50.
google_sheets_read_tabreadRead every cell in a single named tab.
google_sheets_read_allreadRead all tabs in a spreadsheet in one call.
google_sheets_get_metadatareadReturn spreadsheet metadata: tab names, IDs, dimensions, and properties.
google_sheets_searchreadFind spreadsheets in the connected account by name.
google_sheets_writewriteWrite values into a specific range, overwriting existing cells.
google_sheets_appendwriteAppend new rows to the end of a tab without overwriting existing data.
google_sheets_renamewriteRename a spreadsheet or a tab within it.
google_sheets_add_tabwriteAdd a new tab (worksheet) to an existing spreadsheet.
google_sheets_createcreateCreate a new spreadsheet in the connected Google account.
google_sheets_copycreateDuplicate an existing spreadsheet, or copy a tab into another file.
google_sheets_delete_tabdeleteDelete a tab from a spreadsheet.
google_sheets_deletedeleteDelete an entire spreadsheet from Drive.

Connect Google Sheets to Claude, ChatGPT, Cursor, and more

Point any MCP-compatible client at the remote URL and sign in to PortEden once with OAuth.

# Claude Web or Desktop, then Settings, Connectors, Add custom connector
https://mcp.porteden.com/google-sheets
# Authenticate to PortEden once with OAuth. Claude can now call the
# Google Sheets tools under the access policy you set.

Available on Claude Pro, Team, and Enterprise. The same flow works for Claude Cowork.

The security-first Google Sheets MCP

PortEden is the data firewall for AI.

Granular access control

Decide exactly what each AI client can do with Sheets. Allow reads but deny writes, or scope a connection to a single spreadsheet or even a named range, so an agent that asks for everything only ever receives what policy allows.

  • Read-only by default: deny google_sheets_write, append, and delete for a research agent.
  • Per-spreadsheet allow and deny: pin a client to one workbook and block the rest of the account.
  • Per-range scoping: expose Sheet1!A:F to the agent and keep the salary columns out of reach.
Learn more about granular access control

Data redaction

PortEden runs a redaction pass on every tool response before it leaves the firewall. Sensitive cell values are replaced with stable placeholders, so the agent can reason over the data without the raw values entering the model's context.

  • Strip PII, PAN (card numbers), and email patterns from cell values before the agent sees them.
  • Mask a whole column (for example SSN or salary) while leaving the rest of the row readable.
  • Round-trip writes are de-redacted server-side, so an update aimed at a placeholder lands on the right cell.
Learn more about data redaction

Audit trail

Every Google Sheets tool call is recorded: which client, which user, which tool, the arguments passed, the policy decision, and the redacted response. Export it or stream it to your SIEM for review.

  • See exactly which ranges an agent read and which rows it wrote, with timestamps.
  • Reconstruct any tool call: the request, the rule that fired, and the response returned.
  • Stream to Splunk, Datadog, or S3 for retention and quarterly review.
Learn more about audit trail

RBAC and policy groups

Bind each MCP connection to a user or role with a scoped, revocable token. Group policies by team so a new hire inherits the right Sheets access on day one, and revocation is instant and server-side.

  • Issue per-user tokens scoped to the verbs and spreadsheets that role needs.
  • Apply one policy group across a team instead of editing rules client by client.
  • Revoke a token server-side the moment a contract ends, with no Google account round-trip.
Learn more about rbac and policy groups

Set up the secure Google Sheets MCP in minutes

1

Add the connector

In your AI client, add a custom connector or HTTP MCP server pointing at https://mcp.porteden.com/google-sheets.

2

Authenticate with OAuth

Sign in to PortEden once and connect your Google account. The client never holds your Google refresh token.

3

Set your Sheets policy

Choose read-only or read-write, scope to specific spreadsheets or ranges, and turn on redaction for sensitive columns.

4

Verify in the audit log

Run a prompt, then watch the tool calls land in your PortEden audit log with the rule that fired on each one.

Works with every MCP client
Claude DesktopClaude WebChatGPTCursorWindsurfClineZed

Google Sheets MCP FAQ

What is the Google Sheets MCP server?
It is a remote MCP (Model Context Protocol) server that lets MCP-compatible AI clients like Claude, ChatGPT, and Cursor read and write Google Sheets through a single hosted endpoint at https://mcp.porteden.com/google-sheets. PortEden's server exposes 13 Google Sheets tools and puts a data firewall in front of every one, so you control what the AI can see and do.
How do I connect Claude to Google Sheets using MCP?
In Claude Web or Desktop, open Settings, then Connectors, then Add custom connector, and paste https://mcp.porteden.com/google-sheets. Authenticate to PortEden once with OAuth and Claude can call the Sheets tools. The same URL works in ChatGPT Connectors and in Cursor's MCP settings.
Is the Google Sheets MCP server secure?
That is the whole point of PortEden. The connection uses OAuth, so there is no static key to leak. Every tool call passes through a policy decision, sensitive cell values are redacted before the agent sees them, and every call is written to an audit log. You can scope a client to read-only, to a single spreadsheet, or to specific ranges.
Can I give an AI agent read-only access to Google Sheets?
Yes. Deny the write, append, create, and delete tools in your policy and the agent keeps the five read tools (read, read_tab, read_all, get_metadata, and search). You can also allow writes only on specific spreadsheets or ranges.
Does redaction stop the agent from working with my data?
No. PortEden replaces sensitive values with stable placeholders rather than removing them, so the agent can sort, deduplicate, and summarize by handle. Round-trip writes are de-redacted server-side, so an update aimed at a placeholder still lands on the correct cell. The raw values never enter the model's context.
What Google Sheets tools does the MCP server expose?
Thirteen tools: read, read_tab, read_all, get_metadata, and search for reading; write, append, rename, and add_tab for editing; create and copy for new files; and delete_tab and delete for removal. Each tool can be allowed or denied per client.
Which AI clients work with the Google Sheets MCP server?
Any MCP-compatible client. That includes Claude Desktop, Claude Web, ChatGPT (via Connectors), Cursor, Windsurf, Cline, and Zed. Point the client at the remote URL and authenticate once.
How is this different from connecting Google Sheets directly?
A direct connector or community MCP server usually proxies the Google Sheets API one to one, handing raw cell values to the model with a broad OAuth grant. PortEden sits in between as a data firewall: scoped tokens, per-tool and per-range policy, redaction on every response, and a full audit log. You get the same tools with control over each call.
Does it cost anything to use the Google Sheets MCP server?
There is a free tier that covers 1,000 tool calls per month. Paid plans raise the limit. A tool call is one MCP request, so reading 200 rows is one call, not 200. See the pricing page for current limits.

Connect Google Sheets to AI, without leaking the underlying data.

Five-minute setup over OAuth. The free tier covers 1,000 tool calls per month.

See pricing