用户需求 → 判断任务类型 → 选择工具 → 执行操作 → 公式重算 → 验证输出
User request → Task classification → Tool selection → Execute → Recalc formulas → Validate

python scripts/recalc.py output.xlsx
# Check JSON output: status should be "success"
# If "errors_found", fix and recalculate

total = df['Sales'].sum()
sheet['B10'] = total  # Hardcodes 5000

sheet['B10'] = '=SUM(B2:B9)'  # Dynamic formula

from openpyxl.styles import Font

# Blue text (0000FF): User inputs and assumptions
ws['A1'].font = Font(color='0000FF')

# Black text (000000): ALL formulas and calculations
ws['B1'].font = Font(color='000000')

# Green text (008000): Links within same workbook
ws['C1'].font = Font(color='008000')

# Red text (FF0000): External file links
ws['D1'].font = Font(color='FF0000')

# Yellow background (FFFF00): Key assumptions
from openpyxl.styles import PatternFill
ws['E1'].fill = PatternFill(start_color='FFFF00', fill_type='solid')

from openpyxl.styles import numbers

# Years: Format as text strings "2024" (not numbers)
ws['A1'].number_format = '@'  # Text format
ws['A1'] = "'2024"  # Force text

# Currency: $#,##0 + specify units in header
ws['B1'].number_format = '$#,##0'

# Zeros: Format as "-"
ws['C1'].number_format = '$#,##0;($#,##0);"-"'

# Percentages: 0.0% (one decimal)
ws['D1'].number_format = '0.0%'

# Multiples: 0.0x for valuation multiples
ws['E1'].number_format = '0.0"x"'

# Negative numbers: Use parentheses (123) not minus -123
ws['F1'].number_format = '#,##0;(#,##0)'

import pandas as pd

# Read Excel (single sheet or all)
df = pd.read_excel('file.xlsx')  # 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
df.groupby('Category')['Sales'].sum()  # Group by

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

from openpyxl import Workbook
from openpyxl.styles import Font

wb = Workbook()
ws = wb.active

# Data with formulas
ws['A1'] = 'Revenue'
ws['A2'] = 1000
ws['A3'] = 1200
ws['A4'] = 'Total'
ws['B4'] = '=SUM(A2:A3)'  # Use formula, not hardcoded value

# Color coding
ws['A2'].font = Font(color='0000FF')  # Blue: User input
ws['B4'].font = Font(color='000000')  # Black: Formula

wb.save('output.xlsx')

# After saving file with formulas
python scripts/recalc.py output.xlsx

# Output JSON shows:
# - status: "success" or "errors_found"
# - error_summary: Details of any errors
# - total_errors: Count of formula errors

from openpyxl import Workbook
from openpyxl.styles import Font, PatternFill

wb = Workbook()
ws = wb.active

# Headers
ws['A1'] = 'Year'
ws['B1'] = 'Revenue ($mm)'
ws['C1'] = 'Growth %'

# Data with color coding
ws['A2'] = "'2024"  # Force text for year
ws['B2'] = 100
ws['B2'].font = Font(color='0000FF')  # Blue: User input

ws['C2'] = 0.15  # Growth assumption
ws['C2'].font = Font(color='0000FF')
ws['C2'].fill = PatternFill(start_color='FFFF00', fill_type='solid')  # Yellow: Key assumption
ws['C2'].number_format = '0.0%'

# Formulas
ws['B3'] = '=B2*(1+C2)'  # Revenue calculation
ws['B3'].font = Font(color='000000')  # Black: Formula

wb.save('financial_model.xlsx')

# MUST recalculate
import subprocess
subprocess.run(['python', 'scripts/recalc.py', 'financial_model.xlsx'])

import subprocess
import json

# Run recalc and check results
result = subprocess.run(
    ['python', 'scripts/recalc.py', 'output.xlsx'],
    capture_output=True,
    text=True
)

data = json.loads(result.stdout)

if data['status'] == 'errors_found':
    print(f"Found {data['total_errors']} errors:")
    for error_type, count in data['error_summary'].items():
        print(f"  {error_type}: {count}")
    # Fix errors and recalculate
else:
    print("✅ Zero formula errors - file ready!")

pip install -r requirements.txt
# Includes: openpyxl, pandas, xlrd, openpyxl

# macOS
brew install --cask libreoffice

# Ubuntu/Debian
sudo apt-get install libreoffice

# Windows: Download from https://www.libreoffice.org/

import pandas as pd
from openpyxl import load_workbook
from openpyxl.styles import Font

# Step 1: Create data with pandas
data = {
    'Product': ['A', 'B', 'C'],
    'Q1': [1000, 800, 1200],
    'Q2': [1100, 850, 1300],
    'Q3': [1050, 900, 1250],
    'Q4': [1200, 950, 1400]
}
df = pd.DataFrame(data)
df.to_excel('sales.xlsx', index=False, startrow=1)

# Step 2: Add formulas with openpyxl
wb = load_workbook('sales.xlsx')
ws = wb.active

# Add title
ws['A1'] = '2024 Sales Report ($000s)'
ws['A1'].font = Font(bold=True, size=14)

# Add totals row
last_row = len(df) + 2
ws[f'A{last_row}'] = 'Total'
ws[f'A{last_row}'].font = Font(bold=True)

for col in ['B', 'C', 'D', 'E']:
    ws[f'{col}{last_row}'] = f'=SUM({col}2:{col}{last_row-1})'
    ws[f'{col}{last_row}'].font = Font(color='000000', bold=True)

wb.save('sales.xlsx')

# Step 3: Recalculate
import subprocess
subprocess.run(['python', 'scripts/recalc.py', 'sales.xlsx'])

from openpyxl import Workbook
from openpyxl.styles import Font, PatternFill, Alignment

wb = Workbook()
ws = wb.active

# Headers
headers = ['Metric', '2024', '2025', '2026']
for col, header in enumerate(headers, start=1):
    ws.cell(1, col, header)
    ws.cell(1, col).font = Font(bold=True)
    ws.cell(1, col).alignment = Alignment(horizontal='center')

# Assumptions (Blue + Yellow background)
ws['A2'] = 'Base Revenue'
ws['B2'] = 10000
ws['B2'].font = Font(color='0000FF')
ws['B2'].fill = PatternFill(start_color='FFFF00', fill_type='solid')

ws['A3'] = 'Growth Rate'
ws['B3'] = 0.15
ws['B3'].font = Font(color='0000FF')
ws['B3'].fill = PatternFill(start_color='FFFF00', fill_type='solid')
ws['B3'].number_format = '0.0%'

# Calculations (Black formulas)
ws['A5'] = 'Projected Revenue'
ws['B5'] = '=B2'
ws['B5'].font = Font(color='000000')

ws['C5'] = '=B5*(1+$B$3)'
ws['C5'].font = Font(color='000000')

ws['D5'] = '=C5*(1+$B$3)'
ws['D5'].font = Font(color='000000')

# Number formatting
for col in ['B', 'C', 'D']:
    ws[f'{col}5'].number_format = '$#,##0'

wb.save('projection.xlsx')

# MUST recalculate
import subprocess
subprocess.run(['python', 'scripts/recalc.py', 'projection.xlsx'])

import pandas as pd
import glob

all_data = []

for file in glob.glob('data/*.xlsx'):
    df = pd.read_excel(file)
    df['source_file'] = file
    all_data.append(df)

combined = pd.concat(all_data, ignore_index=True)
combined.to_excel('combined_report.xlsx', index=False)

print(f"Processed {len(all_data)} files, total {len(combined)} rows")