More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Messy Excel Fixer: LATAM & EU Formats

    1

    Cleans messy Excel and CSV exports into a tidy, formatted workbook, with LATAM and EU number and date formats built in, and checks that the file's own total still adds up.

    $9

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Messy Excel Fixer: LATAM & EU Formats

    Messy Excel Fixer: LATAM & EU Formats

    Example session with this skill installed

    Here's ventas_octubre.xlsx, our October sales export from the ERP. The amounts come as text with dots and commas, the accents are broken and there's a TOTAL row at the bottom. Clean it and give me the total per salesperson (Vendedor).

    • Read your context and instructions
    • Compiled the messy excel fixer
    • Generated the document

    Cleaned ventas_octubre.xlsx: 1,210 rows read, 1,201 kept. The file's own grand total matches the sum of the clean rows. (Demo data is fictitious.)

    CheckResult
    HeaderFound on row 5, under 3 title rows and a blank one
    Removed4 repeated headers, 3 exact duplicates, 1 TOTAL row
    Accents272 cells repaired ("Peña" became "Peña")
    To review1 cell: row 650, Total says "pendiente"
    TotalDecimal comma detected. File total 1.501.471.019,93, sum of rows 1,501,471,019.93: matches
    DatesThree formats mixed (05/10/2026, 5-oct-2026, 05.10.26), all read day-first
    NITKept as text, leading zeros intact

    Total per salesperson

    VendedorInvoicesTotal (COP)
    Carlos Peña315412,525,864.93
    Lucía Martínez316392,949,247.25
    Andrea Gómez292374,487,862.69
    Jorge Ibáñez278321,508,045.06
    TOTAL1,2011,501,471,019.93

    Saved as ventas_octubre_clean.xlsx with the sheets Clean data, Summary, Report, Needs review and Removed rows. The original wasn't touched.

    Two things you may want to do next: fill in the "pendiente" amount on invoice FV-10640x, and if this report comes every month, I can set up the double-click kit so your team runs it without me.

    ventas_octubre.xlsx

    PDF · document

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Every ERP, bank or POS export comes out a little broken. Title rows on top of the header, amounts like 1.234.567 stored as text, dates in three formats, "Droguería" instead of "Droguería", a TOTAL row in the middle and the header repeated on every page.

    This skill gives your agent two Python scripts that clean all of that and write down every decision they make. It was built for Latin American and European files, where the decimal comma and dd/mm dates are the norm, but it handles US formats too.

    What you get back is a regular Excel file: typed columns, proper number and date formats, filters, and yellow on the few cells it couldn't read. Next to it there's a summary of what changed, a list of every row it removed and why, and a check of the file's own total against the clean rows. You can also ask for totals per customer, salesperson or month.

    If a client needs to clean the same report every month, the agent can leave them a small folder with a .bat they double-click. They don't need to know any Python, they just need it installed once.

    It runs on your machine. No network calls, no API keys. Your agent reads the summary the scripts print, not your whole file.

    Tested on a 50,000-row file, which takes about 20 seconds on an ordinary laptop.

    What it doesn't do: it expects one table per sheet, so side-by-side or cross-tab layouts need reshaping first. Old .xls files have to be saved as .xlsx. Formulas come in as their last saved values. Past 300,000 rows or so, a database is the better tool.

    One table per sheet. Old .xls files must be saved as .xlsx first. Formulas come in as their last saved values. Past roughly 300,000 rows, a database is a better fit.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 6 days ago

    Requires

    Python
    openpyxl (pip install openpyxl)

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed6 days ago

    What's inside

    Frequently Asked Questions

    Pasting a sheet into a chat cuts big files short and can invent values. Here plain code does the work on your machine, row by row, and tells you what it changed.

    Any agent that can run Python: Claude Code, Codex, Cursor, Gemini CLI and the like. It needs Python 3.9+ and openpyxl.

    That's what it was built for: decimal commas, dd/mm dates, month names in Spanish and Portuguese, and accents broken by a bad export.

    Two scripts (one cleans, one builds the workbook and the optional report), the format rules, and a double-click kit for clients with a one-page guide in English and Spanish.

    It's been tested with 50,000 rows. Past about 300,000, use a database. Old .xls files need to be saved as .xlsx first.

    No. The scripts run locally and make no network calls. Your agent only sees the summary they print.

    New in Finance & Accounting