Agent skill
excel-automation
Create, parse, and control Excel files on macOS. Professional formatting with openpyxl, complex xlsm parsing with stdlib zipfile+xml for investment bank financial models, and Excel window control via AppleScript. Use when creating formatted Excel reports, parsing financial models that openpyxl cannot handle, or automating Excel on macOS.
Install this agent skill to your Project
npx add-skill https://github.com/daymade/claude-code-skills/tree/main/excel-automation
SKILL.md
Excel Automation
Create professional Excel files, parse complex financial models, and control Excel on macOS.
Quick Start
# Create a formatted Excel report
uv run --with openpyxl scripts/create_formatted_excel.py output.xlsx
# Parse a complex xlsm that openpyxl can't handle
uv run scripts/parse_complex_excel.py model.xlsm # List sheets
uv run scripts/parse_complex_excel.py model.xlsm "DCF" # Extract a sheet
uv run scripts/parse_complex_excel.py model.xlsm --fix # Fix corrupted names
# Control Excel via AppleScript (with timeout to prevent hangs)
timeout 5 osascript -e 'tell application "Microsoft Excel" to activate'
Overview
Three capabilities:
| Capability | Tool | When to Use |
|---|---|---|
| Create formatted Excel | openpyxl |
Reports, mockups, dashboards |
| Parse complex xlsm/xlsx | zipfile + xml.etree |
Financial models, VBA workbooks, >1MB files |
| Control Excel window | AppleScript (osascript) |
Zoom, scroll, select cells programmatically |
Tool Selection Decision Tree
Is the file simple (data export, no VBA, <1MB)?
├─ YES → openpyxl or pandas
└─ NO
├─ Is it .xlsm or from investment bank / >1MB?
│ └─ YES → zipfile + xml.etree.ElementTree (stdlib)
└─ Is it truly .xls (BIFF format)?
└─ YES → xlrd
Signals of "complex" Excel: file >1MB, .xlsm extension, from investment bank/broker, contains VBA macros.
IMPORTANT: Always run file <path> first — extensions lie. A .xls file may actually be a ZIP-based xlsx.
Creating Excel Files (openpyxl)
Professional Color Convention (Investment Banking Standard)
| Color | RGB Code | Meaning |
|---|---|---|
| Blue | 0000FF |
User input / assumption |
| Black | 000000 |
Calculated value |
| Green | 008000 |
Cross-sheet reference |
| White on dark blue | FFFFFF on 4472C4 |
Section headers |
| Dark blue text | 1F4E79 |
Title |
Core Formatting Patterns
from openpyxl.styles import Font, PatternFill, Border, Side, Alignment
# Fonts
BLUE_FONT = Font(color="0000FF", size=10, name="Calibri")
BLACK_FONT_BOLD = Font(color="000000", size=10, name="Calibri", bold=True)
GREEN_FONT = Font(color="008000", size=10, name="Calibri")
HEADER_FONT = Font(color="FFFFFF", size=12, name="Calibri", bold=True)
# Fills
DARK_BLUE_FILL = PatternFill("solid", fgColor="4472C4")
LIGHT_BLUE_FILL = PatternFill("solid", fgColor="D9E1F2")
INPUT_GREEN_FILL = PatternFill("solid", fgColor="E2EFDA")
LIGHT_GRAY_FILL = PatternFill("solid", fgColor="F2F2F2")
# Borders
THIN_BORDER = Border(bottom=Side(style="thin", color="B2B2B2"))
BOTTOM_DOUBLE = Border(bottom=Side(style="double", color="000000"))
Number Format Codes
| Format | Code | Example |
|---|---|---|
| Currency | '$#,##0' |
$1,234 |
| Currency with decimals | '$#,##0.00' |
$1,234.56 |
| Percentage | '0.0%' |
12.3% |
| Percentage (2 decimal) | '0.00%' |
12.34% |
| Number with commas | '#,##0' |
1,234 |
| Multiplier | '0.0x' |
1.5x |
Conditional Formatting (Sensitivity Tables)
Red-to-green gradient for sensitivity analysis:
from openpyxl.formatting.rule import ColorScaleRule
rule = ColorScaleRule(
start_type="min", start_color="F8696B", # Red (low)
mid_type="percentile", mid_value=50, mid_color="FFEB84", # Yellow (mid)
end_type="max", end_color="63BE7B" # Green (high)
)
ws.conditional_formatting.add(f"B2:F6", rule)
Execution
uv run --with openpyxl scripts/create_formatted_excel.py
Full template script: See scripts/create_formatted_excel.py
Parsing Complex Excel (zipfile + xml)
When openpyxl fails on complex xlsm files (corrupted DefinedNames, complex VBA), use stdlib directly.
XLSX Internal ZIP Structure
file.xlsx (ZIP archive)
├── [Content_Types].xml
├── xl/
│ ├── workbook.xml ← Sheet names + order
│ ├── sharedStrings.xml ← All text values (lookup table)
│ ├── worksheets/
│ │ ├── sheet1.xml ← Cell data for sheet 1
│ │ ├── sheet2.xml ← Cell data for sheet 2
│ │ └── ...
│ └── _rels/
│ └── workbook.xml.rels ← Maps rId → sheetN.xml
└── _rels/.rels
Sheet Name Resolution (Two-Step)
Sheet names in workbook.xml link to physical files via _rels/workbook.xml.rels:
import zipfile
import xml.etree.ElementTree as ET
MAIN_NS = 'http://schemas.openxmlformats.org/spreadsheetml/2006/main'
REL_NS = 'http://schemas.openxmlformats.org/officeDocument/2006/relationships'
RELS_NS = 'http://schemas.openxmlformats.org/package/2006/relationships'
def get_sheet_path(zf, sheet_name):
"""Resolve sheet name to physical XML file path inside ZIP."""
# Step 1: workbook.xml → find rId for the sheet name
wb_xml = ET.fromstring(zf.read('xl/workbook.xml'))
sheets = wb_xml.findall(f'.//{{{MAIN_NS}}}sheet')
rid = None
for s in sheets:
if s.get('name') == sheet_name:
rid = s.get(f'{{{REL_NS}}}id')
break
if not rid:
raise ValueError(f"Sheet '{sheet_name}' not found")
# Step 2: workbook.xml.rels → map rId to file path
rels_xml = ET.fromstring(zf.read('xl/_rels/workbook.xml.rels'))
for rel in rels_xml.findall(f'{{{RELS_NS}}}Relationship'):
if rel.get('Id') == rid:
return 'xl/' + rel.get('Target')
raise ValueError(f"No file mapping for {rid}")
Cell Data Extraction
def extract_cells(zf, sheet_path):
"""Extract all cell values from a sheet XML."""
# Build shared strings lookup
shared = []
try:
ss_xml = ET.fromstring(zf.read('xl/sharedStrings.xml'))
for si in ss_xml.findall(f'{{{MAIN_NS}}}si'):
texts = si.itertext()
shared.append(''.join(texts))
except KeyError:
pass # No shared strings
# Parse sheet cells
sheet_xml = ET.fromstring(zf.read(sheet_path))
rows = sheet_xml.findall(f'.//{{{MAIN_NS}}}row')
data = {}
for row in rows:
for cell in row.findall(f'{{{MAIN_NS}}}c'):
ref = cell.get('r') # e.g., "A1"
cell_type = cell.get('t') # "s" = shared string, None = number
val_el = cell.find(f'{{{MAIN_NS}}}v')
if val_el is not None and val_el.text:
if cell_type == 's':
data[ref] = shared[int(val_el.text)]
else:
try:
data[ref] = float(val_el.text)
except ValueError:
data[ref] = val_el.text
return data
Fixing Corrupted DefinedNames
Investment bank xlsm files often have corrupted <definedName> entries containing "Formula removed":
def fix_defined_names(zf_in_path, zf_out_path):
"""Remove corrupted DefinedNames and repackage."""
import shutil, tempfile
with tempfile.TemporaryDirectory() as tmp:
tmp = Path(tmp)
with zipfile.ZipFile(zf_in_path, 'r') as zf:
zf.extractall(tmp)
wb_xml_path = tmp / 'xl' / 'workbook.xml'
tree = ET.parse(wb_xml_path)
root = tree.getroot()
ns = {'main': MAIN_NS}
defined_names = root.find('.//main:definedNames', ns)
if defined_names is not None:
for name in list(defined_names):
if name.text and "Formula removed" in name.text:
defined_names.remove(name)
tree.write(wb_xml_path, encoding='utf-8', xml_declaration=True)
with zipfile.ZipFile(zf_out_path, 'w', zipfile.ZIP_DEFLATED) as zf:
for fp in tmp.rglob('*'):
if fp.is_file():
zf.write(fp, fp.relative_to(tmp))
Full template script: See scripts/parse_complex_excel.py
Controlling Excel on macOS (AppleScript)
All commands verified on macOS with Microsoft Excel.
Verified Commands
# Activate Excel (bring to front)
osascript -e 'tell application "Microsoft Excel" to activate'
# Open a file
osascript -e 'tell application "Microsoft Excel" to open POSIX file "/path/to/file.xlsx"'
# Set zoom level (percentage)
osascript -e 'tell application "Microsoft Excel"
set zoom of active window to 120
end tell'
# Scroll to specific row
osascript -e 'tell application "Microsoft Excel"
set scroll row of active window to 45
end tell'
# Scroll to specific column
osascript -e 'tell application "Microsoft Excel"
set scroll column of active window to 3
end tell'
# Select a cell range
osascript -e 'tell application "Microsoft Excel"
select range "A1" of active sheet
end tell'
# Select a specific sheet by name
osascript -e 'tell application "Microsoft Excel"
activate object sheet "DCF" of active workbook
end tell'
Timing and Timeout
Always add sleep 1 between AppleScript commands and subsequent operations (e.g., screenshot) to allow UI rendering.
IMPORTANT: osascript will hang indefinitely if Excel is not running or not responding. Always wrap with timeout:
# Safe pattern: 5-second timeout
timeout 5 osascript -e 'tell application "Microsoft Excel" to activate'
# Check exit code: 124 = timed out
if [ $? -eq 124 ]; then
echo "Excel not responding — is it running?"
fi
Common Mistakes
| Mistake | Correction |
|---|---|
| openpyxl fails on complex xlsm → try monkey-patching | Switch to zipfile + xml.etree immediately |
Count Chinese characters with wc -c |
Use wc -m (chars, not bytes; Chinese = 3 bytes/char) |
| Trust file extension | Run file <path> first to confirm actual format |
openpyxl load_workbook hangs on large xlsm |
Use zipfile for targeted extraction instead of loading entire workbook |
Important Notes
- Execute Python scripts with
uv run --with openpyxl(never use system Python) - LibreOffice (
soffice --headless) can convert formats and recalculate formulas - Detailed formatting reference: See
references/formatting-reference.md
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
claude-code-history-files-finder
Finds and recovers content from Claude Code session history files. This skill should be used when searching for deleted files, tracking changes across sessions, analyzing conversation history, or recovering code from previous Claude interactions. Triggers include mentions of "session history", "recover deleted", "find in history", "previous conversation", or ".claude/projects".
claude-skills-troubleshooting
Diagnose and resolve Claude Code plugin and skill issues. This skill should be used when plugins are installed but not showing in available skills list, skills are not activating as expected, or when troubleshooting enabledPlugins configuration in settings.json. Triggers include "plugin not working", "skill not showing", "installed but disabled", or "enabledPlugins" issues.
fixing-claude-export-conversations
Fixes broken line wrapping in Claude Code exported conversation files (.txt), reconstructing tables, paragraphs, paths, and tool calls that were hard-wrapped at fixed column widths. Includes an automated validation suite (generic, file-agnostic checks). Triggers when the user has a Claude Code export file with broken formatting, mentions "fix export", "fix conversation", "exported conversation", "make export readable", references a file matching YYYY-MM-DD-HHMMSS-*.txt, or has a .txt file with broken tables, split paths, or mangled tool output from Claude Code.
continue-claude-work
Recover actionable context from local `.claude` session artifacts and continue interrupted work without running `claude --resume`. This skill should be used when the user provides a Claude session ID, asks to continue prior work from local history, or wants to inspect `.claude` files before resuming implementation.
promptfoo-evaluation
Configures and runs LLM evaluation using Promptfoo framework. Use when setting up prompt testing, creating evaluation configs (promptfooconfig.yaml), writing Python custom assertions, implementing llm-rubric for LLM-as-judge, or managing few-shot examples in prompts. Triggers on keywords like "promptfoo", "eval", "LLM evaluation", "prompt testing", or "model comparison".
fact-checker
Verifies factual claims in documents using web search and official sources, then proposes corrections with user confirmation. Use when the user asks to fact-check, verify information, validate claims, check accuracy, or update outdated information in documents. Supports AI model specs, technical documentation, statistics, and general factual statements.
Didn't find tool you were looking for?