Ms office suite

MS Office Suite — Document Generation Skill

Generate professional-grade documents: Excel workbooks, Word docs, PowerPoint decks, and PDFs. Optimized for CPA-quality financial reports.

Dependencies

Install per-project in pyproject.toml (or standalone venv):

[project.optional-dependencies]
reports = [
    "python-docx>=1.1.0",
    "openpyxl>=3.1.0",
    "python-pptx>=0.6.23",
    "reportlab>=4.0",
    "weasyprint>=62.0",
    "matplotlib>=3.8.0",
    "markdown>=3.6",
]

Standalone install: uv pip install python-docx openpyxl python-pptx reportlab weasyprint matplotlib markdown

Output Conventions

• Reports go to <project>/reports/ directory
• Filename: <project>-<report-type>-<YYYY-MM-DD>.{pdf,xlsx,docx,pptx}
• Always generate both source format AND PDF (e.g., XLSX + PDF)
• Include generation timestamp in document footer

Excel Workbooks (openpyxl)

Cost Comparison Matrix

from openpyxl import Workbook
from openpyxl.styles import Font, PatternFill, Alignment, Border, Side, numbers
from openpyxl.chart import BarChart, Reference
from openpyxl.utils import get_column_letter

def create_cost_workbook(project_name: str, data: dict) -> str:
    """Generate a multi-sheet cost comparison workbook.

    Args:
        project_name: e.g. "sample-app"
        data: {
            "tiers": ["Starter", "Growth", "Scale"],
            "providers": ["AWS", "Azure", "GCP", "OCI", "Hetzner"],
            "costs": {
                "Starter": {"AWS": {"compute": 25, "db": 15, "storage": 5, ...}, ...},
                ...
            },
            "revenue": {"conservative": 500, "base": 2000, "optimistic": 5000}
        }
    """
    wb = Workbook()

    # Style constants
    header_font = Font(name="Calibri", size=11, bold=True, color="FFFFFF")
    header_fill = PatternFill(start_color="2F5496", end_color="2F5496", fill_type="solid")
    money_fmt = '#,##0.00'
    thin_border = Border(
        left=Side(style='thin'), right=Side(style='thin'),
        top=Side(style='thin'), bottom=Side(style='thin')
    )

    # Sheet 1: Summary comparison
    ws = wb.active
    ws.title = "Summary"
    # ... build summary with totals per tier per provider

    # Sheet per tier: detailed line items
    for tier in data["tiers"]:
        ws = wb.create_sheet(title=tier)
        # Row headers: cost categories (compute, db, storage, network, CDN, DNS, SSL, monitoring, backup)
        # Columns: one per provider
        # Bottom rows: monthly total, annual total, break-even months

    # Sheet: ROI Analysis
    ws = wb.create_sheet(title="ROI Analysis")
    # Three scenarios: conservative, base, optimistic
    # Rows: monthly revenue, monthly cost (per tier), net margin, break-even month
    # Charts: bar chart comparing tiers, line chart for cumulative P&L

    # Sheet: Hidden Costs
    ws = wb.create_sheet(title="Hidden Costs")
    # Data transfer, DNS, SSL certs, log retention, backup storage, support plans

    path = f"reports/{project_name}-cloud-cost-{date.today()}.xlsx"
    wb.save(path)
    return path

Key openpyxl Patterns

# Conditional formatting — highlight cheapest provider per row
from openpyxl.formatting.rule import CellIsRule
ws.conditional_formatting.add(
    f"B2:{get_column_letter(col_count)}{row_count}",
    CellIsRule(operator="equal", formula=["MIN($B2:$F2)"],
              fill=PatternFill(bgColor="C6EFCE"))
)

# Auto-fit column widths (approximate)
for col in ws.columns:
    max_len = max(len(str(cell.value or "")) for cell in col)
    ws.column_dimensions[col[0].column_letter].width = min(max_len + 2, 30)

# Freeze panes (headers always visible)
ws.freeze_panes = "B2"

# Number formatting
for row in ws.iter_rows(min_row=2, min_col=2):
    for cell in row:
        cell.number_format = money_fmt

Chart Generation

# Bar chart comparing providers
chart = BarChart()
chart.type = "col"
chart.title = f"{tier} Tier — Monthly Cost by Provider"
chart.y_axis.title = "USD/month"
chart.y_axis.numFmt = '$#,##0'
data_ref = Reference(ws, min_col=2, max_col=col_count, min_row=1, max_row=row_count)
cats_ref = Reference(ws, min_col=1, min_row=2, max_row=row_count)
chart.add_data(data_ref, titles_from_data=True)
chart.set_categories(cats_ref)
chart.shape = 4
ws.add_chart(chart, "H2")

Word Documents (python-docx)

Executive Summary / Memo

from docx import Document
from docx.shared import Inches, Pt, Cm, RGBColor
from docx.enum.text import WD_ALIGN_PARAGRAPH
from docx.enum.table import WD_TABLE_ALIGNMENT

def create_memo(title: str, sections: list[dict]) -> str:
    """Generate a professional memo/report.

    sections: [{"heading": "...", "body": "...", "table": [[...]], "bullet_points": [...]}]
    """
    doc = Document()

    # Title
    title_para = doc.add_heading(title, level=0)
    title_para.alignment = WD_ALIGN_PARAGRAPH.CENTER

    # Metadata line
    meta = doc.add_paragraph()
    meta.alignment = WD_ALIGN_PARAGRAPH.CENTER
    run = meta.add_run(f"Generated: {datetime.now().strftime('%Y-%m-%d %H:%M')} | CONFIDENTIAL")
    run.font.size = Pt(9)
    run.font.color.rgb = RGBColor(128, 128, 128)

    doc.add_paragraph()  # spacer

    for section in sections:
        doc.add_heading(section["heading"], level=1)

        if "body" in section:
            doc.add_paragraph(section["body"])

        if "bullet_points" in section:
            for point in section["bullet_points"]:
                doc.add_paragraph(point, style="List Bullet")

        if "table" in section:
            rows, cols = len(section["table"]), len(section["table"][0])
            table = doc.add_table(rows=rows, cols=cols, style="Light Shading Accent 1")
            for i, row_data in enumerate(section["table"]):
                for j, cell_val in enumerate(row_data):
                    table.rows[i].cells[j].text = str(cell_val)

    # Footer
    section = doc.sections[0]
    footer = section.footer
    footer_para = footer.paragraphs[0]
    footer_para.text = f"{title} — Page "
    footer_para.alignment = WD_ALIGN_PARAGRAPH.CENTER

    path = f"reports/{title.lower().replace(' ', '-')}-{date.today()}.docx"
    doc.save(path)
    return path

PDF Generation

Option 1: weasyprint (HTML/CSS → PDF) — best for styled reports

from weasyprint import HTML

def html_to_pdf(html_content: str, output_path: str, css: str = None):
    """Convert HTML string to PDF. Supports full CSS3."""
    stylesheets = []
    if css:
        from weasyprint import CSS
        stylesheets = [CSS(string=css)]
    HTML(string=html_content).write_pdf(output_path, stylesheets=stylesheets)

# CPA report CSS
REPORT_CSS = """
@page { size: letter; margin: 1in; }
body { font-family: 'Helvetica Neue', Arial, sans-serif; font-size: 10pt; color: #333; }
h1 { color: #2F5496; border-bottom: 2px solid #2F5496; padding-bottom: 8px; }
h2 { color: #2F5496; margin-top: 24px; }
table { width: 100%; border-collapse: collapse; margin: 16px 0; }
th { background: #2F5496; color: white; padding: 8px 12px; text-align: left; }
td { padding: 6px 12px; border-bottom: 1px solid #ddd; }
tr:nth-child(even) { background: #f8f9fa; }
.money { text-align: right; font-family: 'Courier New', monospace; }
.disclaimer { font-size: 8pt; color: #888; margin-top: 32px; border-top: 1px solid #ddd; padding-top: 8px; }
.conservative { color: #c0392b; font-weight: bold; }
.assumption { background: #fff3cd; padding: 8px; border-left: 4px solid #ffc107; margin: 12px 0; }
"""

Option 2: reportlab — best for programmatic layouts with charts

from reportlab.lib.pagesizes import letter
from reportlab.lib import colors
from reportlab.lib.units import inch
from reportlab.platypus import SimpleDocTemplate, Table, TableStyle, Paragraph, Spacer, Image
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle

def create_pdf_report(title: str, tables: list, charts: list, output_path: str):
    doc = SimpleDocTemplate(output_path, pagesize=letter,
                           topMargin=0.75*inch, bottomMargin=0.75*inch)
    styles = getSampleStyleSheet()

    # Custom styles
    styles.add(ParagraphStyle(name='ReportTitle', parent=styles['Title'],
                              textColor=colors.HexColor('#2F5496'), fontSize=18))
    styles.add(ParagraphStyle(name='Disclaimer', parent=styles['Normal'],
                              textColor=colors.grey, fontSize=7))

    elements = [Paragraph(title, styles['ReportTitle']), Spacer(1, 12)]

    for table_data in tables:
        t = Table(table_data["rows"], colWidths=table_data.get("widths"))
        t.setStyle(TableStyle([
            ('BACKGROUND', (0, 0), (-1, 0), colors.HexColor('#2F5496')),
            ('TEXTCOLOR', (0, 0), (-1, 0), colors.white),
            ('FONTNAME', (0, 0), (-1, 0), 'Helvetica-Bold'),
            ('FONTSIZE', (0, 0), (-1, -1), 9),
            ('GRID', (0, 0), (-1, -1), 0.5, colors.grey),
            ('ROWBACKGROUNDS', (0, 1), (-1, -1), [colors.white, colors.HexColor('#f8f9fa')]),
            ('ALIGN', (1, 1), (-1, -1), 'RIGHT'),
        ]))
        elements.append(t)
        elements.append(Spacer(1, 12))

    # Embed matplotlib charts as images
    for chart_path in charts:
        elements.append(Image(chart_path, width=6*inch, height=3*inch))
        elements.append(Spacer(1, 12))

    doc.build(elements)

Matplotlib Charts for PDF Embedding

import matplotlib.pyplot as plt
import matplotlib.ticker as mticker

def cost_comparison_chart(providers: list, costs: dict, output_path: str):
    """Bar chart: providers on x-axis, cost categories stacked."""
    fig, ax = plt.subplots(figsize=(8, 4))
    categories = list(next(iter(costs.values())).keys())
    x = range(len(providers))
    bottom = [0] * len(providers)

    colors = ['#2F5496', '#4472C4', '#5B9BD5', '#A5C8E1', '#D6E4F0', '#ED7D31', '#FFC000']
    for i, cat in enumerate(categories):
        values = [costs[p][cat] for p in providers]
        ax.bar(x, values, bottom=bottom, label=cat, color=colors[i % len(colors)])
        bottom = [b + v for b, v in zip(bottom, values)]

    ax.set_xticks(x)
    ax.set_xticklabels(providers)
    ax.yaxis.set_major_formatter(mticker.StrMethodFormatter('${x:,.0f}'))
    ax.set_ylabel('Monthly Cost (USD)')
    ax.legend(bbox_to_anchor=(1.05, 1), loc='upper left', fontsize=8)
    plt.tight_layout()
    plt.savefig(output_path, dpi=150, bbox_inches='tight')
    plt.close()

def breakeven_chart(months: range, revenue: dict, cost: float, output_path: str):
    """Line chart: cumulative revenue vs cost over time."""
    fig, ax = plt.subplots(figsize=(8, 4))
    for scenario, monthly in revenue.items():
        cumulative = [monthly * m - cost * m for m in months]
        style = '--' if scenario == 'conservative' else '-' if scenario == 'base' else ':'
        ax.plot(months, cumulative, style, label=f'{scenario} (${monthly}/mo)')

    ax.axhline(y=0, color='red', linestyle='-', alpha=0.3)
    ax.yaxis.set_major_formatter(mticker.StrMethodFormatter('${x:,.0f}'))
    ax.set_xlabel('Months')
    ax.set_ylabel('Cumulative Profit/Loss')
    ax.legend()
    plt.tight_layout()
    plt.savefig(output_path, dpi=150, bbox_inches='tight')
    plt.close()

PowerPoint Decks (python-pptx)

from pptx import Presentation
from pptx.util import Inches, Pt, Emu
from pptx.dml.color import RGBColor
from pptx.enum.text import PP_ALIGN

def create_deck(title: str, slides_data: list[dict]) -> str:
    """Generate a presentation deck.

    slides_data: [
        {"layout": "title", "title": "...", "subtitle": "..."},
        {"layout": "content", "title": "...", "bullets": ["..."]},
        {"layout": "table", "title": "...", "table": [[...]]},
        {"layout": "chart", "title": "...", "chart_image": "path.png"},
    ]
    """
    prs = Presentation()
    prs.slide_width = Inches(13.333)  # 16:9
    prs.slide_height = Inches(7.5)

    for slide_data in slides_data:
        layout_idx = {"title": 0, "content": 1, "blank": 6}.get(slide_data["layout"], 1)
        slide = prs.slides.add_slide(prs.slide_layouts[layout_idx])

        if "title" in slide_data and slide.shapes.title:
            slide.shapes.title.text = slide_data["title"]

        if slide_data["layout"] == "chart" and "chart_image" in slide_data:
            slide.shapes.add_picture(
                slide_data["chart_image"],
                Inches(1), Inches(1.5), Inches(11), Inches(5.5)
            )

        if slide_data["layout"] == "table":
            table_data = slide_data.get("table")
            if not table_data or not table_data[0]:
                raise ValueError("table slides require a non-empty 'table' matrix")

            column_count = len(table_data[0])
            if any(len(row) != column_count for row in table_data):
                raise ValueError("table slides require rows with equal column counts")

            table = slide.shapes.add_table(
                len(table_data),
                column_count,
                Inches(0.75),
                Inches(1.5),
                Inches(11.8),
                Inches(5.25),
            ).table
            for row_index, row_data in enumerate(table_data):
                for column_index, value in enumerate(row_data):
                    cell = table.cell(row_index, column_index)
                    cell.text = str(value)
                    if row_index == 0:
                        for run in cell.text_frame.paragraphs[0].runs:
                            run.font.bold = True

        if "bullets" in slide_data:
            tf = slide.placeholders[1].text_frame
            for i, bullet in enumerate(slide_data["bullets"]):
                if i == 0:
                    tf.text = bullet
                else:
                    p = tf.add_paragraph()
                    p.text = bullet

    path = f"reports/{title.lower().replace(' ', '-')}-{date.today()}.pptx"
    prs.save(path)
    return path

CPA-Grade Financial Report Rules

These rules are NON-NEGOTIABLE for any financial document:

1. Three-scenario projections: Always show Conservative, Base, and Optimistic. Label which is which. Default to Conservative for headline numbers.

2. Source every cost figure: Include footnotes or inline citations. "AWS RDS db.t3.medium: $X/mo (us-east-1, on-demand, retrieved YYYY-MM-DD)"

3. No vanity metrics: No TAM/SAM/SOM unless backed by real data. No "if we capture just 1% of the market" reasoning.

4. Unit economics first: CAC (customer acquisition cost), LTV (lifetime value), gross margin, burn rate. If unknown, say "TBD — requires market testing."

5. Hidden costs: Always include data transfer, DNS, SSL, monitoring, logging, backups, support plans. These typically add 15-30% to base compute+DB costs.

6. Self-hosted baseline: For projects running on existing infrastructure, compare cloud cost to the marginal cost of self-hosting (compute + electricity costs at typical rates).

7. Break-even analysis: For each tier, show months to break-even at each revenue scenario. If break-even exceeds 24 months at base case, flag it.

8. Assumptions box: Every report gets a clearly labeled "Key Assumptions" section listing what was assumed and what could change.

9. Conservative by default: When in doubt, round costs UP and revenue DOWN. finance/CPA reviewers immediately spot optimistic bias.

10. Currency and precision: USD, two decimal places for per-unit costs, whole dollars for monthly/annual totals.

Markdown to PDF

import markdown
from weasyprint import HTML

def markdown_to_pdf(md_content: str, output_path: str):
    """Convert markdown to styled PDF via HTML intermediate."""
    html_body = markdown.markdown(md_content, extensions=['tables', 'fenced_code'])
    full_html = f"""<!DOCTYPE html>
    <html><head><style>{REPORT_CSS}</style></head>
    <body>{html_body}</body></html>"""
    HTML(string=full_html).write_pdf(output_path)

Quick Reference — When to Use What

Need	Tool	Best For
Multi-sheet data with formulas	openpyxl (Excel)	Cost comparisons, financial models
Styled narrative report	weasyprint (HTML→PDF)	Executive summaries, legal memos
Programmatic layout with charts	reportlab (PDF)	One-pagers with embedded charts
Client-facing document	python-docx (Word)	Editable deliverables
Presentation	python-pptx (PPT)	Pitch decks, board presentations
Quick report from markdown	weasyprint + markdown	Internal reports, research memos