# XLSX

> Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.

- Skill: `foxl-ai/xlsx` (Agent Skill, multi-file: 54 files)
- Install (CLI): `npx skillmds@latest add foxl-ai/xlsx`
- Raw SKILL.md: https://api.skillmd.com/api/skills/foxl-ai/xlsx/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Web & Frontend
- License: Proprietary. LICENSE.txt has complete terms
- Author: foxl-ai (https://skillmd.com/u/foxl-ai)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/foxl-ai/xlsx

---


# Requirements for Outputs

## All Excel files

### Professional Font
- Use a consistent, professional font (e.g., Arial, Times New Roman) for all deliverables unless otherwise instructed by the user

### Zero Formula Errors
- Every Excel model MUST be delivered with ZERO formula errors (#REF!, #DIV/0!, #VALUE!, #N/A, #NAME?)

### Preserve Existing Templates (when updating templates)
- Study and EXACTLY match existing format, style, and conventions when modifying files
- Never impose standardized formatting on files with established patterns
- Existing template conventions ALWAYS override these guidelines

## Financial models

### Color Coding Standards
Unless otherwise stated by the user or existing template

#### Industry-Standard Color Conventions
- **Blue text (RGB: 0,0,255)**: Hardcoded inputs, and numbers users will change for scenarios
- **Black text (RGB: 0,0,0)**: ALL formulas and calculations
- **Green text (RGB: 0,128,0)**: Links pulling from other worksheets within same workbook
- **Red text (RGB: 255,0,0)**: External links to other files
- **Yellow background (RGB: 255,255,0)**: Key assumptions needing attention or cells that need to be updated

### Number Formatting Standards

#### Required Format Rules
- **Years**: Format as text strings (e.g., "2024" not "2,024")
- **Currency**: Use $#,##0 format; ALWAYS specify units in headers ("Revenue ($mm)")
- **Zeros**: Use number formatting to make all zeros "-", including percentages (e.g., "$#,##0;($#,##0);-")
- **Percentages**: Default to 0.0% format (one decimal)
- **Multiples**: Format as 0.0x for valuation multiples (EV/EBITDA, P/E)
- **Negative numbers**: Use parentheses (123) not minus -123

### Formula Construction Rules

#### Assumptions Placement
- Place ALL assumptions (growth rates, margins, multiples, etc.) in separate assumption cells
- Use cell references instead of hardcoded values in formulas
- Example: Use =B5*(1+$B$6) instead of =B5*1.05

#### Formula Error Prevention
- Verify all cell references are correct
- Check for off-by-one errors in ranges
- Ensure consistent formulas across all projection periods
- Test with edge cases (zero values, negative numbers)
- Verify no unintended circular references

#### Documentation Requirements for Hardcodes
- Comment or in cells beside (if end of table). Format: "Source: [System/Document], [Date], [Specific Reference], [URL if applicable]"
- Examples:
  - "Source: Company 10-K, FY2024, Page 45, Revenue Note, [SEC EDGAR URL]"
  - "Source: Company 10-Q, Q2 2025, Exhibit 99.1, [SEC EDGAR URL]"
  - "Source: Bloomberg Terminal, 8/15/2025, AAPL US Equity"
  - "Source: FactSet, 8/20/2025, Consensus Estimates Screen"

# XLSX creation, editing, and analysis

## Overview

A user may ask you to create, edit, or analyze the contents of an .xlsx file. You have different tools and workflows available for different tasks.

## Important Requirements

**Use `officecli` for formula values, and for looking at the sheet.** It is a
single self-contained binary that evaluates Excel formulas itself (350+
functions) and renders the workbook - no LibreOffice, no Excel, no Office
install. Verify it is present before relying on it:

```bash
officecli --version
```

If missing, install it (open a new shell if the binary still is not found):

```bash
# macOS / Linux
curl -fsSL https://d.officecli.ai/install.sh | bash
# Windows (PowerShell)
irm https://d.officecli.ai/install.ps1 | iex
```

Runs on macOS, Linux and Windows (x64 and arm64 on each, plus musl/Alpine
Linux) - the installer picks the right binary. **Write output next to the
workbook or into a directory you created, not `/tmp`**: the examples below use
`/tmp` for brevity, but it does not exist on Windows - use `$env:TEMP` there, or
a relative `out\` folder.

Supported formats are exactly `.docx`, `.xlsx`, `.pptx`. Legacy `.xls` is NOT
supported - ask for an `.xlsx`. `.csv`/`.tsv` are handled with pandas, or
imported with `officecli import`.

**Do NOT assume LibreOffice exists.** It is not installed on most machines this
skill runs on, and nothing here installs it.

## Reading and analyzing data

### Reading with officecli

Fastest way to see what is actually in a workbook, including COMPUTED formula
values:

```bash
# Structure: sheets, dimensions, formula counts
officecli view file.xlsx outline

# Cell values as text (A1=value, tab-separated; empty cells omitted)
officecli view file.xlsx text
officecli view file.xlsx text --cols A,B,C --max-lines 50
officecli view file.xlsx text --range "Sheet1!A1:C10"

# One cell, with formula AND cached/computed value
officecli get file.xlsx /Sheet1/B10 --json

# Find cells: CSS-like selectors, incl. row-by-column-name
officecli query file.xlsx 'cell:has(formula)'
officecli query file.xlsx 'Sheet1!row[Salary>5000]'

# Formula errors and other problems, mechanically
officecli view file.xlsx issues --json

# SEE the sheet (PNG needs a headless browser: Playwright/Chrome/Edge/Firefox)
officecli view file.xlsx screenshot -o /tmp/sheet.png
officecli view file.xlsx screenshot --range "Sheet1!A1:H40" -o /tmp/region.png
officecli view file.xlsx html -o /tmp/sheet.html
officecli watch file.xlsx        # live preview at http://localhost:26315
```

### Data analysis with pandas
For data analysis, visualization, and basic operations, use **pandas** which provides powerful data manipulation capabilities:

```python
import pandas as pd

# Read Excel
df = pd.read_excel('file.xlsx')  # Default: first sheet
all_sheets = pd.read_excel('file.xlsx', sheet_name=None)  # All sheets as dict

# Analyze
df.head()      # Preview data
df.info()      # Column info
df.describe()  # Statistics

# Write Excel
df.to_excel('output.xlsx', index=False)
```

## Excel File Workflows

## CRITICAL: Use Formulas, Not Hardcoded Values

**Always use Excel formulas instead of calculating values in Python and hardcoding them.** This ensures the spreadsheet remains dynamic and updateable.

### ❌ WRONG - Hardcoding Calculated Values
```python
# Bad: Calculating in Python and hardcoding result
total = df['Sales'].sum()
sheet['B10'] = total  # Hardcodes 5000

# Bad: Computing growth rate in Python
growth = (df.iloc[-1]['Revenue'] - df.iloc[0]['Revenue']) / df.iloc[0]['Revenue']
sheet['C5'] = growth  # Hardcodes 0.15

# Bad: Python calculation for average
avg = sum(values) / len(values)
sheet['D20'] = avg  # Hardcodes 42.5
```

### ✅ CORRECT - Using Excel Formulas
```python
# Good: Let Excel calculate the sum
sheet['B10'] = '=SUM(B2:B9)'

# Good: Growth rate as Excel formula
sheet['C5'] = '=(C4-C2)/C2'

# Good: Average using Excel function
sheet['D20'] = '=AVERAGE(D2:D19)'
```

This applies to ALL calculations - totals, percentages, ratios, differences, etc. The spreadsheet should be able to recalculate when source data changes.

## Common Workflow

Two routes. Prefer the first when formulas matter.

### Route A - write formulas with officecli (evaluated as you write)

`officecli set` evaluates a formula at write time and caches the result, so
there is no separate recalculation pass:

```bash
officecli create output.xlsx
officecli set output.xlsx /Sheet1/A1 --prop value="Revenue" --prop bold=true
officecli set output.xlsx /Sheet1/B10 --prop formula="=SUM(B2:B9)"

# read the computed value back
officecli get output.xlsx /Sheet1/B10 --json
```

Then verify (see Verifying formula values below).

### Route B - build with openpyxl/pandas, then fix up the values

openpyxl writes formulas as STRINGS with no cached values, so a reader that
does not recalculate sees blanks.

1. **Choose tool**: pandas for data, openpyxl for formulas/formatting
2. **Create/Load**: Create new workbook or load existing file
3. **Modify**: Add/edit data, formulas, and formatting
4. **Save**: Write to file
5. **Populate formula values (MANDATORY IF USING FORMULAS)**: re-write each
   formula cell through `officecli set`, which evaluates it:
   ```bash
   officecli set output.xlsx /Sheet1/B10 --prop formula="=SUM(B2:B9)"
   ```
   For many cells, do it in one pass with `batch`:
   ```bash
   officecli batch output.xlsx --commands '[
     {"op":"set","path":"/Sheet1/B10","props":{"formula":"=SUM(B2:B9)"}},
     {"op":"set","path":"/Sheet1/C10","props":{"formula":"=AVERAGE(C2:C9)"}}
   ]' --json
   ```
6. **Verify and fix any errors**: see Verifying formula values below. Common
   errors to fix:
   - `#REF!`: Invalid cell references
   - `#DIV/0!`: Division by zero
   - `#VALUE!`: Wrong data type in formula
   - `#NAME?`: Unrecognized formula name

**Flush before a non-officecli program reads the file.** officecli keeps a
resident process, so openpyxl/pandas/an upload may otherwise read a stale file:

```bash
officecli save output.xlsx     # flush, keep resident warm
officecli close output.xlsx    # flush + release
```

### Creating new Excel files

```python
# Using openpyxl for formulas and formatting
from openpyxl import Workbook
from openpyxl.styles import Font, PatternFill, Alignment

wb = Workbook()
sheet = wb.active

# Add data
sheet['A1'] = 'Hello'
sheet['B1'] = 'World'
sheet.append(['Row', 'of', 'data'])

# Add formula
sheet['B2'] = '=SUM(A1:A10)'

# Formatting
sheet['A1'].font = Font(bold=True, color='FF0000')
sheet['A1'].fill = PatternFill('solid', start_color='FFFF00')
sheet['A1'].alignment = Alignment(horizontal='center')

# Column width
sheet.column_dimensions['A'].width = 20

wb.save('output.xlsx')
```

### Editing existing Excel files

```python
# Using openpyxl to preserve formulas and formatting
from openpyxl import load_workbook

# Load existing file
wb = load_workbook('existing.xlsx')
sheet = wb.active  # or wb['SheetName'] for specific sheet

# Working with multiple sheets
for sheet_name in wb.sheetnames:
    sheet = wb[sheet_name]
    print(f"Sheet: {sheet_name}")

# Modify cells
sheet['A1'] = 'New Value'
sheet.insert_rows(2)  # Insert row at position 2
sheet.delete_cols(3)  # Delete column 3

# Add new sheet
new_sheet = wb.create_sheet('NewSheet')
new_sheet['A1'] = 'Data'

wb.save('modified.xlsx')
```

## Verifying formula values

Excel files created or modified by openpyxl contain formulas as strings but no
calculated values. `officecli` evaluates a formula when you WRITE it (`set
... --prop formula=...`), and exposes three separate readback keys so you can
tell a real value from a stale one:

| key | meaning |
|---|---|
| `cachedValue` | the value stored in the file - what a non-recalculating reader sees |
| `computedValue` | what officecli's evaluator computes NOW |
| `uncalculated` | `true` if the formula cell has no cached value at all |

```bash
# one cell
officecli get output.xlsx /Sheet1/B10 --json

# just the cached value
officecli get output.xlsx /Sheet1/B10 --json | jq '.data.results[0].format.cachedValue'

# every formula cell
officecli query output.xlsx 'cell:has(formula)' --json

# formulas shown inline with their resolved values
officecli view output.xlsx annotated

# error scan across the workbook (#REF!, #DIV/0!, #VALUE!, #NAME?, ...)
officecli view output.xlsx issues --json
```

**`cachedValue` != `computedValue` means the file on disk is lying.** That is a
stale cache, reported by `view issues` as `formula_cache_stale`. Fix it by
re-writing the formula (see below), not by ignoring it.

### The staleness trap - read this before shipping a multi-formula model

A formula is evaluated at the moment it is written, using whatever its
precedents had cached AT THAT MOMENT. So a downstream formula written before its
upstream was computed caches a wrong value - often `0` - and that wrong value
survives into every reader that does not recalculate.

After any multi-formula build - especially `SUMPRODUCT`, `SUMIFS` with dynamic
criteria, `INDEX`/`MATCH`, or cross-sheet chains - **re-touch every downstream
cell** by running the same `set` again so it recomputes from the now-correct
upstream:

```bash
# second pass: re-write the dependent formulas in dependency order
officecli set model.xlsx /Summary/B2 --prop formula="=SUMPRODUCT(Data!B2:B50,Data!C2:C50)"
```

Do the re-touch pass WITHOUT a resident open (`officecli close model.xlsx`
first) - re-touching cross-sheet chains through a resident is unreliable.

Two more limits worth knowing:
- A cell the evaluator could not compute carries the sentinel
  `#OCLI_NOTEVAL!`. Remedy: close residents, then `set` the cell again.
- **Dynamic arrays**: only the anchor (top-left) cell is evaluated; the spilled
  cells are produced by Excel when it opens the file.

### If you have LibreOffice and want a full-workbook recalculation

`scripts/recalc.py` still exists and does a true `calculateAll()` across every
sheet, then scans all cells for Excel errors. It requires LibreOffice
(`soffice`), which is usually NOT installed - check first and do not make it
part of the default path:

```bash
# macOS / Linux
command -v soffice && python scripts/recalc.py output.xlsx 30
```

```powershell
# Windows PowerShell
if (Get-Command soffice -ErrorAction SilentlyContinue) { python scripts/recalc.py output.xlsx 30 }
```

Note the helper it uses (`scripts/office/soffice.py`) carries a Linux-only
sandbox shim (LD_PRELOAD + gcc), so on macOS/Windows this path works only if
`soffice` is already on PATH.

Prefer the officecli route above; reach for this only when you specifically need
a whole-workbook recalculation by a real spreadsheet engine.

## Formula Verification Checklist

Quick checks to ensure formulas work correctly:

### Essential Verification
- [ ] **Test 2-3 sample references**: Verify they pull correct values before building full model
- [ ] **Column mapping**: Confirm Excel columns match (e.g., column 64 = BL, not BK)
- [ ] **Row offset**: Remember Excel rows are 1-indexed (DataFrame row 5 = Excel row 6)

### Common Pitfalls
- [ ] **NaN handling**: Check for null values with `pd.notna()`
- [ ] **Far-right columns**: FY data often in columns 50+ 
- [ ] **Multiple matches**: Search all occurrences, not just first
- [ ] **Division by zero**: Check denominators before using `/` in formulas (#DIV/0!)
- [ ] **Wrong references**: Verify all cell references point to intended cells (#REF!)
- [ ] **Cross-sheet references**: Use correct format (Sheet1!A1) for linking sheets

### Formula Testing Strategy
- [ ] **Start small**: Test formulas on 2-3 cells before applying broadly
- [ ] **Verify dependencies**: Check all cells referenced in formulas exist
- [ ] **Test edge cases**: Include zero, negative, and very large values

### Checking for formula errors

Primary check - no LibreOffice needed:

```bash
officecli view output.xlsx issues --json
```

It reports `#REF!` / `#VALUE!` / `#NAME?` / `#DIV/0!` and the
`formula_cache_stale` subtype (a `cachedValue` that disagrees with the
evaluator). Fix what it lists, re-write the affected formulas, re-check.

### Interpreting scripts/recalc.py Output (LibreOffice route only)
When you deliberately used the `scripts/recalc.py` fallback, it returns JSON
with error details:
```json
{
  "status": "success",           // or "errors_found"
  "total_errors": 0,              // Total error count
  "total_formulas": 42,           // Number of formulas in file
  "error_summary": {              // Only present if errors found
    "#REF!": {
      "count": 2,
      "locations": ["Sheet1!B5", "Sheet1!C10"]
    }
  }
}
```

## Best Practices

### Library Selection
- **pandas**: Best for data analysis, bulk operations, and simple data export
- **openpyxl**: Best for complex formatting, formulas, and Excel-specific features

### Working with openpyxl
- Cell indices are 1-based (row=1, column=1 refers to cell A1)
- Use `data_only=True` to read calculated values: `load_workbook('file.xlsx', data_only=True)`
- **Warning**: If opened with `data_only=True` and saved, formulas are replaced with values and permanently lost
- For large files: Use `read_only=True` for reading or `write_only=True` for writing
- Formulas are preserved but not evaluated - re-write each formula cell through
  `officecli set ... --prop formula="..."` to populate its cached value (see
  Verifying formula values)

### Working with pandas
- Specify data types to avoid inference issues: `pd.read_excel('file.xlsx', dtype={'id': str})`
- For large files, read specific columns: `pd.read_excel('file.xlsx', usecols=['A', 'C', 'E'])`
- Handle dates properly: `pd.read_excel('file.xlsx', parse_dates=['date_column'])`

## Code Style Guidelines
**IMPORTANT**: When generating Python code for Excel operations:
- Write minimal, concise Python code without unnecessary comments
- Avoid verbose variable names and redundant operations
- Avoid unnecessary print statements

**For Excel files themselves**:
- Add comments to cells with complex formulas or important assumptions
- Document data sources for hardcoded values
- Include notes for key calculations and model sections