IntermediateAgent Skill

xlsx

Use AI Excel spreadsheet creation to turn operating data into a workbook your team can keep using. Give a file-capable assistant the source tables and required calculations; this Skill guides workbook creation, editing, formatting and formula checks, making it useful for weekly store reports where changing an input should update the derived figures.

Views
0
Copies
0
Favorites
0
Editorial Score
88/100
01

What this Skill does

The xlsx Skill focuses on spreadsheet files rather than a text-only table in a chat response. It provides guidance for preserving formulas and formatting, recalculating the workbook and checking for errors before handing over the file.

The worked example on this page builds a store report with editable inputs and recalculating formulas using openpyxl and LibreOffice. Those are local spreadsheet tools, not a required SaaS subscription. The assistant needs file execution and the relevant dependencies to produce and verify the workbook.

02

What makes it different

01

Approach and advantage

Separate entered values from calculated figures so the workbook remains useful after the first report. Keep calculations as formulas where appropriate, then format the sheet so an operator can distinguish inputs, totals and the figures used in the review.

02

What you can use it for

Recalculation and error checks are part of the handoff, not an optional visual polish. Verify the saved values and formula results before sharing the workbook, so recipients can inspect both the business numbers and the logic that produced them.

03

Before you use it

Inputs to prepare

A supported Claude environment with file tools, source workbook or tabular data, and a clear output specification.

Agent and runtime

Use Anthropic’s supported spreadsheet environment and current terms.

Scope to confirm

This is proprietary source under Anthropic’s terms, not an MIT Skill. This page does not redistribute its body or offer a local installation command; use the official supported environment.

04

Install and get started

Use Anthropic’s supported document tools. The local installation below is for Claude Code; Claude web uses its own available file tools.

  1. Use Claude Code with plugin installation enabled. Run these slash commands inside Claude Code, not in your operating-system terminal.
  2. Install Anthropic’s document-skills collection, which includes XLSX and other document Skills; this follows the current marketplace version.
  3. Start a new session if needed and ask Claude to use xlsx on a small test workbook. Follow its runtime requirements and inspect formulas and the saved file.
Help me set up Anthropic’s XLSX Skill in Claude Code using the official document-skills plugin. Official source: https://github.com/anthropics/skills. Check whether it is already available. If not, guide me through these Claude Code commands: /plugin marketplace add anthropics/skills /plugin install document-skills@anthropic-agent-skills This installs the document collection, not just XLSX. Inspect the installed XLSX instructions and required runtime, then report its actual path and what is ready. Ask me for a sample workbook before doing any spreadsheet task. If I am using Claude on the web, guide me to its supported spreadsheet/file tools instead; do not claim that a local plugin was installed.
05

Get started

Starter prompts for the main use cases—copy and use them directly.

01

Clean an uploaded product-catalog spreadsheet

Clean an uploaded product-catalog spreadsheet Use for: Clean an uploaded product-catalog spreadsheet. Prepare: A supported Claude environment with file tools, source workbook or tabular data, and a clear output specification.

Show prompt and variablesHide prompt and variables
Use the spreadsheet tools in the supported Claude environment. Task: Clean an uploaded product-catalog spreadsheet. Inputs: [INPUT]. Constraints: [CONSTRAINTS]. Check missing information first. Produce a reviewable result with evidence, assumptions, and next steps. This is proprietary source under Anthropic’s terms, not an MIT Skill. This page does not redistribute its body or offer a local installation command; use the official supported environment.

Replace these variables

[INPUT]
A supported Claude environment with file tools, source workbook or tabular data, and a clear output specification.
[CONSTRAINTS]
Scope, allowed actions, output format, and review requirements.
02

Build a formula-based margin workbook from supplied data

Build a formula-based margin workbook from supplied data Use for: Build a formula-based margin workbook from supplied data. Prepare: A supported Claude environment with file tools, source workbook or tabular data, and a clear output specification.

Show prompt and variablesHide prompt and variables
Use the spreadsheet tools in the supported Claude environment. Task: Build a formula-based margin workbook from supplied data. Inputs: [INPUT]. Constraints: [CONSTRAINTS]. Check missing information first. Produce a reviewable result with evidence, assumptions, and next steps. This is proprietary source under Anthropic’s terms, not an MIT Skill. This page does not redistribute its body or offer a local installation command; use the official supported environment.

Replace these variables

[INPUT]
A supported Claude environment with file tools, source workbook or tabular data, and a clear output specification.
[CONSTRAINTS]
Scope, allowed actions, output format, and review requirements.
03

Review formulas and formatting before delivering an XLSX file

Review formulas and formatting before delivering an XLSX file Use for: Review formulas and formatting before delivering an XLSX file. Prepare: A supported Claude environment with file tools, source workbook or tabular data, and a clear output specification.

Show prompt and variablesHide prompt and variables
Use the spreadsheet tools in the supported Claude environment. Task: Review formulas and formatting before delivering an XLSX file. Inputs: [INPUT]. Constraints: [CONSTRAINTS]. Check missing information first. Produce a reviewable result with evidence, assumptions, and next steps. This is proprietary source under Anthropic’s terms, not an MIT Skill. This page does not redistribute its body or offer a local installation command; use the official supported environment.

Replace these variables

[INPUT]
A supported Claude environment with file tools, source workbook or tabular data, and a clear output specification.
[CONSTRAINTS]
Scope, allowed actions, output format, and review requirements.
06

Worked example: Build a weekly Excel report with recalculating formulas

Two synthetic weeks contain 100 orders/$2,400 gross/4,000 sessions and 120 orders/$2,880 gross/4,800 sessions, plus refunds and ad spend, for an editable weekly workbook.

Steps and prompt

  1. Prepare the attached business materials, preserving amounts, units, dates and metric definitions.
  2. Create an Excel trading workbook from the two weekly rows supplied. Include an editable input sheet, formula-based summary and a short guide. Calculate net revenue, conversion rate, average order value, ad-spend ratio and week-on-week changes. Recalculate and check the saved values, then create a second what-if workbook with current orders changed to 150 while the other inputs remain unchanged.
  3. Submit the exact prompt below. Read the primary deliverable on the right and download all deliverables for the team handoff.

Case input files: business-inputs.zip

Use the xlsx Skill.

Create an Excel trading workbook from the two weekly rows supplied. Include an editable input sheet, formula-based summary and a short guide. Calculate net revenue, conversion rate, average order value, ad-spend ratio and week-on-week changes. Recalculate and check the saved values, then create a second what-if workbook with current orders changed to 150 while the other inputs remain unchanged.

The attached business materials are:
- inputs/weekly.csv

Deliver a complete result.md report that the operations team can read, plus any finished files needed for this task. Keep supporting calculations and evidence with the deliverables. Write the report in English; use the requested target language for customer-facing copy. The business datasets are fictional examples unless the brief explicitly names a live public website.

What the run produced

The primary file produced by this prompt is shown below in its original language.

Weekly trading workbook — saved cell values

The tables below are extracted from the recalculated baseline workbook. The complete workbook and the 150-order scenario are included in the download bundle.

Inputs

RowABCDEF
1WeekOrdersGross revenue (USD)Refunds (USD)Ad spend (USD)Sessions
2Prior10024001203004000
3Current12028801443604800
4
5Edit yellow cells B2:F3. Prior and Current are the two supplied example weeks.
6Source: weekly.csv; fictional merchant data.
7All money values are USD. Orders and sessions are nonnegative whole counts.

Summary

RowABCDE
1Weekly trading summary
2MetricPriorCurrentChangeWoW change
3Orders1001202020.00%
4Gross revenue (USD)2400288048020.00%
5Refunds (USD)1201442420.00%
6Ad spend (USD)3003606020.00%
7Sessions4000480080020.00%
8Net revenue (USD)2280273645620.00%
9Conversion rate2.50%2.50%00
10Average order value (USD)22.822.800
11Ad-spend ratio13.16%13.16%00
12
13Rate changes in column D are percentage-point differences; column E is relative growth.
14N/A means the denominator is zero or a comparison is undefined.

Guide

RowA
1Workbook guide
21. Edit Inputs!B2:F3 (yellow). Rows 2 and 3 are the supplied Prior and Current examples.
32. Open Summary to compare weeks. Excel automatic calculation updates formulas after edits.
43. Net revenue = gross revenue minus refunds; no other costs or taxes are deducted.
54. Conversion rate = orders / sessions.
65. Average order value = net revenue / orders (net AOV).
76. Ad-spend ratio = ad spend / net revenue. This is a spend share, not ROAS.
87. Change = Current minus Prior; WoW change = Current / Prior minus 1.
98. Zero denominators show N/A. A zero Prior value makes relative WoW undefined.
109. USD is the supplied currency. No dates, order-level data or profit costs were supplied.
1110. The 150-order scenario changes only Inputs!B3; revenue, refunds, spend and sessions stay fixed.
1211. Scenario AOV falls because the same net revenue is divided across more orders.
13Data note: supplied fictional merchant examples; scenario is a sensitivity calculation.

Merchant business data is synthetic. Website observations, when present, come from the public URLs listed in the report.

07

Risks and operating notes

Known limits

This is proprietary source under Anthropic’s terms, not an MIT Skill. This page does not redistribute its body or offer a local installation command; use the official supported environment.

Evidence and review

Source and package files reviewed: GitHub dbd4588f9e10. This is a static review, not an installation or business-outcome test.

Before acting on the result

Use authorized inputs and keep secrets out of prompts. Review customer-facing claims, calculations, code changes, and any proposed external action before applying them.

Frequently asked questions

What does it do, and what is the advantage?

Create, edit, format, and check spreadsheet files within Anthropic’s supported environment. The source requires formula recalculation and error checks, with spreadsheet-specific handling for formulas and formatting.

How do I install it and start?

Open the official Anthropic source and usage documentation linked below to confirm the supported environment and applicable terms. In a Claude environment with spreadsheet/file tools enabled, upload a workbook or table and ask for an XLSX deliverable. Describe formulas, sheets, and formatting; download the result and check recalculated values before use.

Which situations is it useful for?

Clean an uploaded product-catalog spreadsheet; Build a formula-based margin workbook from supplied data; Review formulas and formatting before delivering an XLSX file

What limitations should I know?

This is proprietary source under Anthropic’s terms, not an MIT Skill. This page does not redistribute its body or offer a local installation command; use the official supported environment.

What was checked for this page?

Source and package files reviewed: GitHub dbd4588f9e10. This is a static review, not an installation or business-outcome test.

No reviews yet.

Leave a review

Rate every dimension, then share your real experience.

Task effectiveness
Setup and ease of use
Reliability and guardrails
Documentation clarity
Time to first useful result
Author / maintainer

Anthropic

Source and package files reviewed: GitHub dbd4588f9e10. This is a static review, not an installation or business-outcome test.