Skip to main content
Glama

google-sheets-mcp

An MCP server that gives an LLM cell-level and formatting-level control over Google Sheets via the Sheets API v4. Two deployment modes:

  • local (default) — runs over stdio for one user; authenticates with your own Google account via a desktop OAuth flow and caches the token on disk.

  • group — centrally hosted over HTTP for many users; each connects from Claude via the native "Connect" button and acts as their own Google identity. The server is stateless — it persists no per-user tokens (see Group mode).

Capabilities

Natural interactions (start here)

A high-level layer sits on top of the ~75 low-level API wrappers so common requests map to a single call phrased the way a user thinks. These accept a spreadsheet ID or a full URL, and natural range references (a header name like Revenue, last row, top 10 rows, whole sheet).

  • describe_spreadsheet — one-call snapshot: sheets, dimensions, column headers with inferred types, sample rows, tables/charts. Call this first to ground a request before acting.

  • build_table — write data + style the header + freeze + banding + auto-fit + per-column number formats in one shot (preview=True to see the plan first).

  • apply_style_preset — clean / financial / report-header / input.

  • highlight_where — conditional highlighting from a plain predicate: "> 100", "between 10 and 20", "contains overdue", "blank", "duplicates", "top 10%".

  • format_numbers — currency / percent / date / thousands / plain.

  • add_totals_row, autofit, sort_by (by header name).

Prompts (in the client's slash/prompt menu): format_as_report, clean_sheet, build_dashboard, analyze — natural-language workflows that orchestrate the tools.

The low-level wrappers below remain available for precise control.

Low-level tools

Structure

  • get_spreadsheet_info — list sheets with ids, dimensions, frozen rows/cols

  • create_spreadsheet, add_sheet, rename_sheet, delete_sheet, duplicate_sheet

Cell values

  • read_range, batch_read — read A1 ranges (formatted, unformatted, or formulas)

  • write_range, batch_write, append_rows, clear_range, batch_clear

Data operations

  • insert_rows, delete_rows, insert_columns, delete_columns

  • append_rows_to_sheet, append_columns_to_sheet, move_rows_or_columns

  • insert_range, delete_range, copy_range, cut_paste_range, paste_delimited_data, auto_fill

  • sort_range, find_replace, trim_whitespace, delete_duplicates, text_to_columns, randomize_range

Formatting

  • format_cells — bold/italic/underline/strikethrough, font size & family, text & background color, horizontal/vertical alignment, wrap strategy, number formats

  • set_borders — outer + inner borders, styles, colors

  • merge_cells, unmerge_cells

  • set_dimension_size (column width / row height), auto_resize_dimensions

  • freeze rows/columns

  • add_conditional_format, update_conditional_format, delete_conditional_format

  • add_banding, update_banding, delete_banding

Filters, validation, and ranges

  • set_basic_filter, clear_basic_filter

  • add_filter_view, update_filter_view, delete_filter_view

  • set_data_validation, clear_data_validation

  • add_named_range, update_named_range, delete_named_range

  • add_protected_range, update_protected_range, delete_protected_range

  • add_dimension_group, update_dimension_group, delete_dimension_group

  • add_table, update_table, delete_table

Charts

  • create_chart — create embedded column, bar, line, area, scatter, combo, stepped-area, pie, and donut charts from existing row/column ranges

  • create_chart_from_spec — create any Sheets API chart spec, including histogram, scorecard, bubble, candlestick, waterfall, treemap, and org charts

  • update_chart, delete_chart, move_chart, set_chart_border

  • create_pivot_table

  • add_slicer, update_slicer, delete_slicer

Advanced

  • batch_update_advanced — allowlisted raw Sheets API batchUpdate requests (max 50 per call; data-source lifecycle requests are blocked)

Related MCP server: @node2flow/google-sheets-mcp

Setup (local mode)

1. Get an OAuth client secret

  1. In the Google Cloud Console, create (or pick) a project and enable the Google Sheets API (APIs & Services → Library).

  2. Configure the OAuth consent screen (External is fine for personal use; add your own Google account as a Test user).

  3. APIs & Services → Credentials → Create Credentials → OAuth client ID → application type Desktop app. Download the JSON.

  4. Save it as credentials.json under the config directory:

    ~/.config/google-sheets-mcp/credentials.json

    (Or set $GOOGLE_SHEETS_CREDENTIALS to its path.)

2. Install

uv sync

3. Authenticate once

uv run google-sheets-mcp auth

This opens a browser, you log in, and the token is cached at ~/.config/google-sheets-mcp/token.json. After this the server runs without prompting (the token auto-refreshes).

Running

The server speaks MCP over stdio — point your MCP client at it.

Claude Code

claude mcp add google-sheets -- uv run --directory /path/to/google-sheets-mcp google-sheets-mcp

Claude Desktop (claude_desktop_config.json)

{
  "mcpServers": {
    "google-sheets": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/google-sheets-mcp", "google-sheets-mcp"]
    }
  }
}

Group (hosted) mode

One server, many users. Each user clicks Connect in Claude, consents with their own Google account, and from then on every tool call runs as that user — so Google's own sharing permissions and audit trail apply per person.

How auth works (and why nothing is stored)

The server is both an OAuth Authorization Server to Claude and an OAuth client to Google (a "bridge"). When a user connects, the flow is:

Claude  --register/authorize-->  this server  --redirect-->  Google consent
Google  --code-->  /oauth/google/callback  (exchange for Google tokens)
this server  --issues-->  MCP access + refresh tokens  -->  Claude stores them
Claude  --Bearer token-->  tool calls  -->  per-request Google client

Every MCP token the server issues is a self-contained, encrypted blob that wraps the user's Google token. The user's Google refresh token lives inside the MCP refresh token that Claude stores client-side — the server keeps no per-user token database. The only durable secrets are app-level: the Google client secret and a symmetric wrap key. This is the secure-but-good-UX middle ground: the same one-click Connect experience as a first-party connector, without a central store of user credentials.

Setup

  1. Google Cloud: enable the Sheets API; configure the OAuth consent screen; create an OAuth Web application client. Add this authorized redirect URI:

    https://YOUR_HOST/oauth/google/callback

    Download the client JSON (or note the client id/secret).

  2. Generate a wrap key:

    uv run google-sheets-mcp genkey
  3. Configure and run (put secrets in a secret manager / env, not in files):

    export GSHEETS_MCP_MODE=group
    export GSHEETS_PUBLIC_URL=https://YOUR_HOST          # public https base URL
    export GSHEETS_GOOGLE_CLIENT_SECRETS=/path/web_client.json   # or *_CLIENT_ID/_SECRET
    export GSHEETS_FERNET_KEY=<key from genkey>          # or GSHEETS_FERNET_KEYS=new,old
    export GSHEETS_HOST=0.0.0.0 GSHEETS_PORT=8000
    uv run google-sheets-mcp serve --mode group

    Terminate TLS in front of it (nginx/caddy/cloud LB) so GSHEETS_PUBLIC_URL is https://. Then add https://YOUR_HOST/mcp as a custom/remote connector in Claude and click Connect.

Security notes

  • TLS is mandatory — bearer tokens travel in request headers.

  • The wrap key is the crown jewel. Keep it in a KMS/secret manager. To rotate with zero downtime, set GSHEETS_FERNET_KEYS=<new>,<old> (encrypt with the new key, still accept the old) until old tokens expire, then drop the old key.

  • Least privilege: the server requests only the spreadsheets scope (plus openid/email for per-user identity in logs).

  • Revocation: /revoke revokes the user's Google grant. Because issued access tokens are self-contained, they can't be individually revoked before expiry without a denylist — mitigated by short (≈1h) access-token lifetimes.

  • Scaling caveat: dynamic client registrations are held in memory (public redirect URIs only, never tokens). For multi-instance deployments, run behind a sticky-session LB or add a shared client store; tokens themselves need no shared state.

Notes on ranges

  • A1 notation (Sheet1!A1:C10, A:C, 2:5, B2) is used by all value/formatting tools. If you omit the Sheet! prefix the first sheet is used.

  • Dimension tools (set_dimension_size, auto_resize_dimensions) use zero-based, half-open indices: start_index=0, end_index=3 = first three.

  • Colors accept #RRGGBB, short #RGB, named colors (red, lightgray, …), or a {"red":..,"green":..,"blue":..} dict (0.0–1.0).

Config / environment variables

Variable

Default

Purpose

GOOGLE_SHEETS_MCP_HOME

~/.config/google-sheets-mcp

Config directory (local)

GOOGLE_SHEETS_CREDENTIALS

<home>/credentials.json

OAuth client secret (local)

GOOGLE_SHEETS_TOKEN

<home>/token.json

Cached user token (local)

GSHEETS_MCP_MODE

local

local or group

GSHEETS_PUBLIC_URL

—

Public https base URL (group, required)

GSHEETS_GOOGLE_CLIENT_SECRETS

—

Path to Google Web client JSON (group)

GSHEETS_GOOGLE_CLIENT_ID / _SECRET

—

Google client creds if not using the JSON (group)

GSHEETS_FERNET_KEY / GSHEETS_FERNET_KEYS

—

Token wrap key(s); first is primary (group, required)

GSHEETS_HOST / GSHEETS_PORT

0.0.0.0 / 8000

Bind address (group)

Development

uv run pytest

Secrets (credentials.json, token.json) are git-ignored — never commit them.

Available Tools

83 tools
add_bandingC

Add alternating-color banding to an A1 range.

Inputs: range (required), spreadsheet_id (required), row_properties, column_properties.

add_conditional_formatC

Add a boolean conditional-format rule to an A1 range.

condition_type is a Sheets ConditionType such as NUMBER_GREATER, NUMBER_LESS, NUMBER_BETWEEN, TEXT_CONTAINS, TEXT_EQ, CUSTOM_FORMULA, BLANK, NOT_BLANK. values supplies the comparison operands (e.g. ["100"] or ["10","20"]; for CUSTOM_FORMULA a single ["=$A1>5"]). The format applies when the condition is true.

Inputs: range (required), condition_type (required), spreadsheet_id (required), bold, +3 more.

add_dimension_groupC

Group rows or columns over a zero-based, half-open range.

Inputs: sheet (required), dimension (required), end_index (required), start_index (required), +1 more.

add_filter_viewC

Add a filter view over an A1 range.

Inputs: range (required), spreadsheet_id (required), title, sort_specs, +1 more.

add_named_rangeC

Create a named range over an A1 range.

Inputs: name (required), range (required), spreadsheet_id (required).

add_protected_rangeB

Protect an A1 range; defaults to warning-only protection.

Inputs: range (required), spreadsheet_id (required), editors, description, +1 more.

add_sheetC

Add a new sheet/tab to an existing spreadsheet.

Inputs: title (required), spreadsheet_id (required), row_count, column_count.

add_slicerC

Add a slicer for a data range.

Inputs: sheet (required), data_range (required), column_index (required), spreadsheet_id (required), +5 more.

add_tableC

Add a Sheets table over an A1 range.

Inputs: range (required), spreadsheet_id (required), name, rows_properties, +1 more.

add_totals_rowA

Append a bold totals row beneath a table.

By default sums every column except the first (which gets a "Total" label). Override per column with functions keyed by header name, e.g. {"Margin": "AVERAGE", "Count": "COUNT"}.

Inputs: spreadsheet (required), table_range (required), functions.

append_columns_to_sheetC

Append empty columns to the end of a sheet.

Inputs: sheet (required), length (required), spreadsheet_id (required).

append_rowsC

Append rows after the last row of data within/near the given A1 range.

The range selects the table to append to (e.g. Sheet1!A1 or Sheet1!A:D); new rows are added below the existing data.

value_input_option: RAW (verbatim strings; default) or USER_ENTERED (evaluate formulas/dates/numbers — opt in explicitly).

Inputs: range (required), values (required), spreadsheet_id (required), value_input_option.

append_rows_to_sheetC

Append empty rows to the end of a sheet.

Inputs: sheet (required), length (required), spreadsheet_id (required).

apply_style_presetA

Apply a named style preset to a range.

Presets: clean (bold shaded header + underline), financial (currency, right-aligned body + bold header), report-header (merged centered title), input (highlighted bordered input cells). range accepts A1 or natural refs like "whole sheet".

Inputs: range (required), preset (required), spreadsheet (required).

auto_fillC

Autofill within an A1 range by detecting existing source data.

Inputs: range (required), spreadsheet_id (required), use_alternate_series.

autofitA

Auto-fit column widths to their contents.

target may be a sheet name (fits all columns) or an A1 range like Sheet1!A:D (fits just those columns).

Inputs: target (required), spreadsheet (required).

auto_resize_dimensionsC

Auto-fit columns or rows to their contents (zero-based, half-open range).

Inputs: sheet (required), dimension (required), end_index (required), start_index (required), +1 more.

batch_clearC

Clear values from multiple A1 ranges in one call.

Inputs: ranges (required), spreadsheet_id (required).

batch_readC

Read several A1 ranges at once. Returns {"valueRanges": [...]}.

Inputs: ranges (required), spreadsheet_id (required), value_render_option.

batch_update_advancedA

Run allowlisted raw Sheets API batchUpdate requests.

Each request must contain exactly one allowed request kind. Data-source lifecycle requests are blocked, and a call may contain at most 50 requests. Prefer the typed tools when one exists; use this for advanced API coverage.

Inputs: requests (required), spreadsheet_id (required), response_ranges, include_spreadsheet_in_response.

batch_writeA

Write values to multiple ranges in one values batchUpdate call.

data items use the Sheets API ValueRange shape, e.g. {"range": "Sheet1!A1:B2", "values": [[1, 2], [3, 4]]}.

Inputs: data (required), spreadsheet_id (required), value_input_option.

build_tableC

Write a titled table and format it in one shot.

anchor is where the top-left header goes: Sheet1!A1, a bare cell (first sheet), or a sheet name (its A1). Applies a style preset (clean/financial/report-header/input) to the header, optionally freezes the header row, adds banding, auto-fits columns, and applies per-column number_formats ({"Revenue": "currency"}). Set preview=True to get the planned ranges/requests without changing anything.

Inputs: rows (required), anchor (required), headers (required), spreadsheet (required), +6 more.

clear_basic_filterC

Clear the basic filter on a sheet.

Inputs: sheet (required), spreadsheet_id (required).

clear_data_validationB

Clear data validation from an A1 range.

Inputs: range (required), spreadsheet_id (required).

clear_rangeA

Clear the values in an A1 range (formatting is preserved).

Inputs: range (required), spreadsheet_id (required).

copy_rangeC

Copy a range to another range.

Inputs: source_range (required), spreadsheet_id (required), destination_range (required), paste_type, +1 more.

create_chartA

Create an embedded chart from existing sheet data.

chart_type: COLUMN, BAR, LINE, AREA, SCATTER, COMBO, STEPPED_AREA, PIE, or DONUT.

domain_range is the category/label range; series_ranges is one or more value ranges. Each source must be a single row or single column, e.g. Sheet1!A2:A10 plus Sheet1!B2:B10. Ranges without a sheet prefix use sheet. anchor_cell is where the chart is overlaid on that sheet.

Inputs: sheet (required), chart_type (required), domain_range (required), series_ranges (required), +14 more.

create_chart_from_specA

Create an embedded chart from a full Sheets API ChartSpec object.

Use this for histogram, scorecard, bubble, candlestick, waterfall, treemap, org, or other chart specs not covered by create_chart.

Inputs: sheet (required), chart_spec (required), spreadsheet_id (required), anchor_cell, +4 more.

create_pivot_tableC

Create a pivot table at a destination cell.

rows, columns, values, and filter_specs use the Sheets API PivotGroup/PivotValue/PivotFilterSpec shapes.

Inputs: source_range (required), spreadsheet_id (required), destination_cell (required), rows, +4 more.

create_spreadsheetA

Create a new spreadsheet and return its id and URL.

Optionally provide sheet_titles to seed multiple named sheets/tabs.

Inputs: title (required), sheet_titles.

cut_paste_rangeA

Cut a range and paste it starting at a destination cell.

Inputs: source_range (required), spreadsheet_id (required), destination_cell (required), paste_type.

delete_bandingB

Delete banding by ID.

Inputs: spreadsheet_id (required), banded_range_id (required).

delete_chartC

Delete a chart or other embedded object by ID.

Inputs: chart_id (required), spreadsheet_id (required).

delete_columnsC

Delete columns; indices are zero-based and half-open.

Inputs: sheet (required), end_index (required), start_index (required), spreadsheet_id (required).

delete_conditional_formatB

Delete a conditional-format rule by sheet and index.

Inputs: index (required), sheet (required), spreadsheet_id (required).

delete_dimension_groupC

Delete a row/column group.

Inputs: sheet (required), dimension (required), end_index (required), start_index (required), +1 more.

delete_duplicatesB

Remove duplicate rows from a range, optionally comparing only column indexes.

Inputs: range (required), spreadsheet_id (required), comparison_columns.

delete_filter_viewC

Delete a filter view by ID.

Inputs: filter_view_id (required), spreadsheet_id (required).

delete_named_rangeC

Delete a named range by ID.

Inputs: named_range_id (required), spreadsheet_id (required).

delete_protected_rangeC

Delete a protected range by ID.

Inputs: spreadsheet_id (required), protected_range_id (required).

delete_rangeA

Delete cells in an A1 range, shifting remaining cells by ROWS or COLUMNS.

Inputs: range (required), spreadsheet_id (required), shift_dimension.

delete_rowsB

Delete rows; indices are zero-based and half-open.

Inputs: sheet (required), end_index (required), start_index (required), spreadsheet_id (required).

delete_sheetB

Delete a sheet/tab by name or numeric sheetId.

Inputs: sheet (required), spreadsheet_id (required).

delete_slicerC

Delete a slicer by embedded object ID.

Inputs: slicer_id (required), spreadsheet_id (required).

delete_tableC

Delete a table by ID.

Inputs: table_id (required), spreadsheet_id (required).

describe_spreadsheetA

Compact, model-friendly snapshot of a spreadsheet in one call.

Accepts a spreadsheet ID or a full Google Sheets URL. Returns, per sheet: dimensions, frozen rows/cols, column headers with inferred types (number/date-ish/boolean/text/formula), a few sample rows, and any tables/charts. Use this first to ground natural requests like "the Revenue column" before calling other tools.

Inputs: spreadsheet (required), sheets, sample_rows.

duplicate_sheetB

Duplicate a sheet/tab, optionally setting the new title and index.

Inputs: sheet (required), spreadsheet_id (required), new_title, insert_index.

find_replaceB

Find and replace text in one range or all sheets.

Inputs: find (required), replacement (required), spreadsheet_id (required), range, +5 more.

format_cellsA

Apply cell formatting to an A1 range. Only the arguments you pass change.

  • text_color / background_color: "#RRGGBB", a named color (red, blue, lightgray, ...), or omit to leave unchanged.

  • horizontal_alignment: LEFT | CENTER | RIGHT.

  • vertical_alignment: TOP | MIDDLE | BOTTOM.

  • wrap_strategy: OVERFLOW_CELL | CLIP | WRAP.

  • number_format_type: TEXT | NUMBER | PERCENT | CURRENCY | DATE | TIME | DATE_TIME | SCIENTIFIC; pair with an optional number_format_pattern (e.g. "$#,##0.00", "0.0%").

Inputs: range (required), spreadsheet_id (required), bold, italic, +11 more.

format_numbersC

Apply a number format to a range.

kind: currency | percent | date | datetime | thousands | number | plain. range accepts A1 or natural refs (e.g. a header name resolves to its whole column).

Inputs: kind (required), range (required), spreadsheet (required).

freezeA

Freeze the first rows rows and/or columns columns of a sheet.

Pass 0 to unfreeze. Leaving an argument as None keeps it unchanged.

Inputs: sheet (required), spreadsheet_id (required), rows, columns.

get_spreadsheet_infoA

Get a spreadsheet's title, URL, and the list of its sheets.

Returns each sheet's title, numeric sheetId, index, dimensions, and any frozen rows/columns. Use this first to discover sheet names and sizes.

Inputs: spreadsheet_id (required).

highlight_whereA

Conditionally highlight cells from a natural predicate.

condition examples: "> 100", "between 10 and 20", "contains overdue", "blank", "duplicates", "top 10%". Applies the given background color (default light amber), and/or text_color/bold. range accepts A1 or natural refs.

Inputs: range (required), condition (required), spreadsheet (required), bold, +2 more.

insert_columnsC

Insert columns before start_index; indices are zero-based and half-open.

Inputs: sheet (required), end_index (required), start_index (required), spreadsheet_id (required), +1 more.

insert_rangeC

Insert cells in an A1 range, shifting existing cells by ROWS or COLUMNS.

Inputs: range (required), spreadsheet_id (required), shift_dimension.

insert_rowsC

Insert rows before start_index; indices are zero-based and half-open.

Inputs: sheet (required), end_index (required), start_index (required), spreadsheet_id (required), +1 more.

merge_cellsC

Merge cells in an A1 range.

merge_type: MERGE_ALL (one cell), MERGE_COLUMNS (merge each column vertically), or MERGE_ROWS (merge each row horizontally).

Inputs: range (required), spreadsheet_id (required), merge_type.

move_chartC

Move and/or resize a chart using overlay positioning.

Inputs: sheet (required), chart_id (required), spreadsheet_id (required), anchor_cell, +4 more.

move_rows_or_columnsC

Move rows or columns to destination_index; indices are zero-based.

Inputs: sheet (required), dimension (required), end_index (required), start_index (required), +2 more.

paste_delimited_dataC

Paste delimited text/HTML starting at a single cell.

Inputs: data (required), start_cell (required), spreadsheet_id (required), html, +2 more.

randomize_rangeC

Randomize row order inside an A1 range.

Inputs: range (required), spreadsheet_id (required).

read_rangeA

Read cell values from an A1 range (e.g. Sheet1!A1:C10).

value_render_option: FORMATTED_VALUE (as shown, default), UNFORMATTED_VALUE (typed values), or FORMULA (underlying formulas). Returns {"range", "values"} where values is a list of rows.

Inputs: range (required), spreadsheet_id (required), value_render_option.

rename_sheetC

Rename a sheet/tab.

Inputs: sheet (required), new_title (required), spreadsheet_id (required).

set_basic_filterC

Set the sheet's basic filter over an A1 range.

Inputs: range (required), spreadsheet_id (required), sort_specs, filter_specs.

set_bordersA

Set borders on an A1 range.

style: SOLID | SOLID_MEDIUM | SOLID_THICK | DASHED | DOTTED | DOUBLE | NONE (use NONE to clear). top/bottom/ left/right are the outer edges; inner_* set borders between cells.

Inputs: range (required), spreadsheet_id (required), top, left, +6 more.

set_chart_borderC

Set an embedded object's border color.

Inputs: chart_id (required), spreadsheet_id (required), color.

set_data_validationC

Set a data validation rule on an A1 range.

Inputs: range (required), condition_type (required), spreadsheet_id (required), strict, +4 more.

set_dimension_sizeB

Set a fixed pixel size for a band of columns or rows.

dimension: COLUMNS or ROWS. Indices are zero-based and half-open (start_index=0, end_index=3 covers the first three columns/rows).

Inputs: sheet (required), dimension (required), end_index (required), pixel_size (required), +2 more.

sort_byB

Sort a range by a column identified by its header name (or letter).

order is asc or desc. Assumes the first row of range is a header (its cells are not sorted into the body).

Inputs: range (required), column (required), spreadsheet (required), order.

sort_rangeC

Sort a range using Sheets API SortSpec objects.

Inputs: range (required), sort_specs (required), spreadsheet_id (required).

text_to_columnsC

Split a single-column range into multiple columns.

Inputs: source_range (required), spreadsheet_id (required), delimiter, delimiter_type.

trim_whitespaceC

Trim leading/trailing/repeated whitespace in each cell in an A1 range.

Inputs: range (required), spreadsheet_id (required).

unmerge_cellsB

Unmerge any merged cells overlapping an A1 range.

Inputs: range (required), spreadsheet_id (required).

update_bandingC

Update banding using a Sheets API BandedRange object.

Inputs: banded_range (required), spreadsheet_id (required), fields.

update_chartB

Replace a chart's spec. Use move_chart to move or resize it.

Inputs: chart_id (required), chart_spec (required), spreadsheet_id (required).

update_conditional_formatC

Replace a conditional-format rule at index.

Inputs: index (required), range (required), sheet (required), condition_type (required), +6 more.

update_dimension_groupC

Collapse or expand a row/column group.

Inputs: sheet (required), collapsed (required), dimension (required), end_index (required), +2 more.

update_filter_viewC

Update a filter view using a Sheets API FilterView object.

Inputs: filter_view (required), spreadsheet_id (required), fields.

update_named_rangeC

Update a named range's name and/or range.

Inputs: named_range_id (required), spreadsheet_id (required), name, range.

update_protected_rangeC

Update a protected range using a Sheets API ProtectedRange object.

Inputs: spreadsheet_id (required), protected_range (required), fields.

update_slicerC

Update a slicer spec by ID.

Inputs: spec (required), slicer_id (required), spreadsheet_id (required), fields.

update_tableD

Update a table using a Sheets API Table object.

Inputs: table (required), spreadsheet_id (required), fields.

write_rangeB

Write a 2D array of values to an A1 range.

value_input_option: RAW (store verbatim as strings; default) or USER_ENTERED (parse like a user typing — formulas, dates, and numbers are evaluated). Opt in to USER_ENTERED explicitly; it evaluates formulas including those that can make external network requests.

Inputs: range (required), values (required), spreadsheet_id (required), value_input_option.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    C2.8/5.0

    Scored across 83 tools

    Disambiguation4/5

    Despite the large number of tools, each has a clearly defined purpose. Overlaps exist (e.g., multiple formatting tools) but descriptions effectively differentiate them. The high volume may cause selection difficulty, but names and descriptions are distinct enough.

    Naming Consistency4/5

    Most tools follow a clear verb_noun pattern (e.g., clear_range, delete_sheet). A few deviations like 'batch_update_advanced' and 'autofit' break the pattern slightly, but overall the naming is predictable and readable.

    Tool Count2/5

    83 tools is excessive for most use cases. While it aims to cover the full Sheets API, the sheer number overwhelms an agent's selection space. Many tools could be consolidated (e.g., multiple chart creation tools). A trimmed set of 20-30 would be more appropriate.

    Completeness5/5

    The tool set comprehensively covers the Google Sheets API surface: CRUD on sheets, ranges, cells, formatting, charts, pivot tables, data validation, filtering, and more. The inclusion of batch_update_advanced ensures even rare operations are accessible. No obvious gaps for typical spreadsheet tasks.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers