pdf-generate

---

name: pdf-generate description: "Print to PDF" — render a markdown/text report (headings, paragraphs, bullet lists) to a clean PDF document. version: "0.1.0"

pdf-generate

Turn a report into a PDF. Use this when a task asks for a PDF deliverable — a summary, a report, an invoice-like document — rather than leaving the output as plain text.

The helper at scripts/md_to_pdf.py uses reportlab (pip install reportlab) because it is pure-Python and needs no system libraries — it works in any runtime out of the box. It renders a practical subset of markdown: #/##/### headings, paragraphs, and -/* bullet lists.

If you need full HTML/CSS fidelity (complex layout, web pages), use weasyprint instead (pip install weasyprint) — note it requires system libraries (Cairo, Pango), so the runtime must have those; prefer the reportlab helper when your input is a report you control.

Using the helper

# Markdown file -> PDF
python scripts/md_to_pdf.py report.md report.pdf

# From stdin, with a document title
python scripts/md_to_pdf.py - report.pdf --title "Q3 Summary" < report.md

The input is markdown (or plain text — plain text just becomes paragraphs). The output is a paginated PDF on US Letter with sensible margins.

Procedure

1. Compose the content as markdown first. Get the structure right in text — headings for sections, bullets for lists — before rendering. It's far easier to iterate on markdown than on a PDF.
2. Render with the helper to your target path.
3. Confirm the file exists and is non-empty (ls -l report.pdf) — a 0-byte PDF means rendering failed.

Supported markdown subset

Markdown	Renders as
# Title	large heading
## Section / ### Sub	smaller headings
- item or * item	bullet list item
blank-line-separated text	paragraphs

Anything fancier (tables, images, inline emphasis) is rendered as plain text. If you need those, pre-render to HTML and use weasyprint.

Cautions

• Declare the dependency: the runner must pip install reportlab (or weasyprint) before this runs — it is not stdlib.
• Very long unbroken lines are wrapped by the PDF engine, but giant tables won't lay out well in the markdown path — use the HTML route for those.