Python Code

Nebula Writer can use Python as a local analysis tool for spreadsheets, CSVs, charts, generated documents, and reproducible examples.

When to Use This Page#

• You want to analyze data inside a writing project
• You need charts, tables, or summaries for a report
• You want the assistant to produce Python that you can inspect
• You work with Excel, CSV, DOCX, PPTX, or PDF artifacts

How Python Fits Into Nebula Writer#

Python runs against files in your project folder. That means generated charts, cleaned data, and exported artifacts can live next to the draft, paper, spreadsheet, or slide deck that uses them.

Typical uses:

• Read a CSV or Excel workbook
• Summarize columns with pandas
• Generate a chart image
• Create a cleaned workbook
• Produce a table for Markdown or LaTeX
• Generate a DOCX or PPTX artifact from structured data

Example Workspace#

Keep scripts in a folder such as analysis/ and generated outputs in folders such as figures/, tables/, or exports/.

project/
  research/
    paper.tex
    research-brief.md
  office/
    research-dashboard.xlsx
  analysis/
    summarize_dashboard.py
  figures/
    dashboard-summary.png

Read Excel with pandas#

import pandas as pd

df = pd.read_excel("office/research-dashboard.xlsx", sheet_name="Data")
print(df.head())
print(df.describe(include="all"))

Use this when you want the assistant to inspect data before changing prose.

Read analysis/summarize_dashboard.py and office/research-dashboard.xlsx. Explain what the script does and suggest one chart for the Results section.

Create a Chart#

import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_excel("office/research-dashboard.xlsx", sheet_name="Data")

ax = df.groupby("Category")["Amount"].sum().sort_values().plot(kind="barh")
ax.set_title("Budget by category")
ax.set_xlabel("Amount")

plt.tight_layout()
plt.savefig("figures/budget-by-category.png", dpi=200)

Then reference the chart from Markdown or LaTeX:

![Budget by category](../figures/budget-by-category.png)

\begin{figure}
  \centering
  \includegraphics[width=.8\linewidth]{figures/budget-by-category.png}
  \caption{Budget by category.}
\end{figure}

Create an Excel Output#

Use openpyxl or xlsxwriter when the output should remain editable in Excel.

import pandas as pd

df = pd.read_excel("office/research-dashboard.xlsx", sheet_name="Data")
summary = df.groupby("Category", as_index=False)["Amount"].sum()

with pd.ExcelWriter("exports/category-summary.xlsx", engine="xlsxwriter") as writer:
    summary.to_excel(writer, sheet_name="Summary", index=False)

Open the generated workbook in Nebula Writer to inspect it alongside your draft.

Create a PowerPoint Draft#

Use python-pptx for a first pass when data needs to become a deck.

from pptx import Presentation
from pptx.util import Inches

prs = Presentation()
slide = prs.slides.add_slide(prs.slide_layouts[5])
slide.shapes.title.text = "Research dashboard summary"

box = slide.shapes.add_textbox(Inches(1), Inches(1.5), Inches(8), Inches(2))
box.text_frame.text = "Key finding: budget risk is concentrated in three categories."

prs.save("exports/dashboard-summary.pptx")

Then open the .pptx in Nebula Writer for slide review and editing.

Create a DOCX Report#

Use python-docx when you want a generated report that can still be edited as a Word document.

from docx import Document

doc = Document()
doc.add_heading("Research Dashboard Summary", level=1)
doc.add_paragraph("This draft was generated from the project workbook.")
doc.add_heading("Key risks", level=2)
doc.add_paragraph("Review categories with the largest variance before publishing.")
doc.save("exports/dashboard-summary.docx")

Useful Libraries#

Common Python libraries available for analysis workflows include:

Prompt Examples#

Write a Python script that reads @research-dashboard.xlsx and saves a chart to figures/.

Explain this Python script and tell me which output files it creates.

Create a cleaned CSV from this workbook, but do not overwrite the original.

Use Python to generate a one-slide PPTX summary from the spreadsheet.

Safety Tips#

• Keep generated files in exports/, figures/, or tables/
• Do not overwrite original data unless you asked for it explicitly
• Ask the assistant to show the script before running a destructive operation
• Review generated charts before inserting them into a paper or deck
• Keep scripts committed with the draft when reproducibility matters