Documents and files

Can Cursor read Excel files?

Not directly. Cursor reads text files, and .xlsx is a binary format. Its agent can write a script to parse one, or a Spreadsheet server reads it in one call.

Cursor can't read .xlsx with its file reader. Its agent can write a parsing script. The Spreadsheet server reads, filters and summarizes Excel files by path in one call.

Connect free in 1 minute See all Spreadsheet tools

What you get

orders_q3.xlsx, net sales by region

Computed from all 3,412 rows, largest first

Table

Region	PLN
Mazowieckie	1,284,300
3 more regions	...
Opolskie	96,420
Rows read	3,412

Example from "Sales by region from a raw export". Invented names, real math from the tools.

Free to start. Spreadsheet Pro is $19 once. Free vs Pro

Without a tool, and with the tool

Without a tool

Cursor's file reader can't parse an .xlsx. The agent can write and run a Python or Node script in the terminal, which depends on the libraries you have installed.

With Spreadsheet

The Spreadsheet server reads .xlsx and .csv by path, shows sheets and columns, filters rows, searches every cell and summarizes columns. It runs over stdio in Cursor and writes changes to a new file.

An example, from prompt to finished file

Example scenarios built on the real tools of Spreadsheet. Names and amounts are invented.

Example · Developer importing client data

Can Cursor read Excel files

You

Show the sheets and columns in data/customers.xlsx, then list rows where country is DE.

sheet_info Lists sheets, columns and row counts
sheet_query Filters rows where country is DE

Result

The workbook layout and the matching rows, without writing a parsing script.

Example · Ewa Dabrowska, sales coordinator at a furniture maker in Bydgoszcz

Sales by region from a raw export

You

Here is orders_q3.xlsx (uploaded). Give me total net sales in PLN by region, largest first.

sheet_info Finds sheet Orders, 3,412 rows, columns region, net_pln, date
sheet_query Groups by region, sums net_pln, sorts descending

Result

A 5-row table: Mazowieckie 1,284,300 PLN first, Opolskie 96,420 PLN last, computed from all 3,412 rows.

About 15 min saved (estimate)

Example · Graham Lewis, bookkeeper in Bristol

Find one invoice in a year of exports

You

Search ledger_2026.csv for INV-1187 and show me that row.

sheet_find Finds INV-1187 at cell B2214 with a preview of the row

Result

The row with the 1,560.00 GBP amount and its posting date, without opening Excel.

About 5 min saved (estimate)

Every example here works on the free tier. One $39 key removes the limits on all of them. Pricing

Common mistakes with this job

Committing a one-off parsing script to the repo.
Reading only the first sheet of a workbook.
Dates coming back as serial numbers.

Connect in 1 minute

Paste one URL into Claude.ai, Claude Desktop, Cursor or VS Code. No install, no account.

1. Get a free token (30 days, 600 calls an hour)

curl -s -X POST https://mcp.zovo.one/mcp/token

2. Paste this URL as a custom connector

https://mcp.zovo.one/mcp/spreadsheet/t/<token>

Or run it locally (Node.js 18+)

npx -y @theluckystrike/mcp-spreadsheet

Free vs Pro for Spreadsheet

Free ($0, no key)	Pro ($19 once)
Read files up to 5 MB and 5,000 rows (larger files return the first 5,000 rows with a note)	No read limit up to the 50 MB file ceiling
Write up to 500 rows per file	No write row limit
All sheets, formats and expressions
Hosted, up to 2 MB of loaded sheets per token

Where these limits come from

servers/spreadsheet/src/sheet.ts:14-16 (FREE_MAX_ROWS=5000, FREE_MAX_BYTES=5 MB, FREE_WRITE_ROWS=500); servers/spreadsheet/src/index.ts:102,122; https://mcp.zovo.one/s/spreadsheet