Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Workbook Ops: Spreadsheet Build & Audit

    1

    Build, edit, and audit .xlsx, .csv, and .tsv files with Python while keeping live formulas, references, charts, and formatting intact.

    Free

    7 usesSecurity scanned
    Workbook Ops: Spreadsheet Build & Audit

    Workbook Ops: Spreadsheet Build & Audit

    Example session with this skill installed

    I have model/roi_forecast.xlsx with an Inputs sheet and a Summary sheet. The Summary totals were pasted as hardcoded values last quarter, and I think some cross-sheet formulas shifted after we inserted two rows into Inputs. Please rebuild the totals as live SUM formulas, fix any broken references, apply #,##0 number formats to the new cells, keep the existing header styles, and save the result as model/roi_forecast_v2.xlsx without touching the original. Then run a formula error scan and give me a short report of what changed and what still needs a manual look in Excel.

    • Read your context and instructions
    • Compiled the workbook ops

    Summary

    Audited and repaired roi_forecast.xlsx and delivered model/roi_forecast_v2.xlsx — 26 hardcoded totals converted to live SUM formulas, 2 shifted cross-sheet references fixed, and number formats applied, with every original input value and header style untouched.

    MetricBeforeAfter
    Hardcoded values in calc cells260 (live formulas)
    Formula errors found (#REF!, #DIV/0!)40
    Formula cells scannednever audited318
    Refs broken by inserted rows2 stalere-pointed to Inputs!B6:B27

    Next steps

    • Open roi_forecast_v2.xlsx once in Excel or LibreOffice so cached values recalculate (openpyxl writes formulas, it does not evaluate them).
    • Confirm Summary!E12 (ROI %) now divides by the repaired Inputs!C9 growth-rate cell.
    • Diff the rendered layout in tmp/workbooks/roi_forecast_v2.pdf against the original before circulating.
    • Keep the original as the system of record; future edits save under new version names.

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

    About this skill

    The problem

    Standard data tools often strip away Excel formulas, broken references, and custom branding when saving files. Developers struggle to programmatically edit workbooks without destroying the complex logic and formatting that business users rely on.

    What it does

    • Edits existing .xlsx files while preserving live formulas, named ranges, and charts.
    • Builds complex workbooks from scratch with custom styles, number formats, and cross-sheet references.
    • Performs automated audits to detect formula errors like #REF! or #DIV/0! across all sheets.
    • Integrates pandas for heavy data manipulation and openpyxl for high-fidelity spreadsheet styling.

    Frameworks & tools

    Python, openpyxl, pandas, and optional rendering via LibreOffice or Poppler.

    Why this beats prompting it yourself

    General LLMs often suggest overwriting cells with static values, which breaks dynamic models. This skill enforces a "formula-first" workflow and includes specific logic for handling merged cells and caching, preventing the accidental corruption of enterprise-grade spreadsheets.

    Use cases

    • Automate the generation of monthly financial reports with professional IB-style formatting.
    • Audit legacy spreadsheets for broken logic and inconsistent cell references.
    • Inject fresh data into existing templates without damaging pre-built pivot tables or charts.
    • Convert raw CSV exports into polished, multi-sheet workbooks with data validation.

    Known limitations

    openpyxl does not execute formulas. Live calculation results are only visible once the file is opened in a spreadsheet application like Excel or LibreOffice.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.

    openpyxl does not evaluate formulas, so cached values in an agent-edited workbook stay stale until recalculated in Excel or LibreOffice. Visual rendering checks require LibreOffice and poppler installed locally.

    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 buying a skill.

    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

    7 uses

    Times used to date

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Software
    Python

    Permissions

    Read Files
    Write Files

    File Scopes

    *.xlsx
    *.csv
    *.tsv
    output/workbooks/**
    tmp/workbooks/**

    Needs shell access to install openpyxl/pandas and read/write spreadsheet files, always saving outputs under new names so originals stay untouched.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    Unlike standard tools that convert everything to static values, this skill uses specific Python logic to preserve live formulas, named ranges, pivot tables, and custom styling during the editing process.

    The skill is designed for any AI agent or development environment running a standard Python 3.x stack, leveraging industry-standard libraries like pandas and openpyxl.

    The package includes the core logic for spreadsheet manipulation, a set of pre-configured functions for auditing and styling, and documentation on how to integrate the skill into your existing workflow.

    Yes, the auditing feature scans every sheet in a workbook to identify and flag #REF!, #DIV/0!, and other common spreadsheet errors programmatically.

    While the skill can write and modify formulas, the actual calculation results will only populate once the file is opened in a spreadsheet application like Microsoft Excel.

    Yes, the skill is specifically optimized to take raw data from .csv or .tsv files and transform it into fully styled, multi-sheet .xlsx workbooks.

    Free