paginated-report-pdflisted

Generates a client-facing, document-style paginated PDF report (cover page, numbered sections, one page per topic/site/item, tables, callout boxes, numbered step lists) from an HTML source, using headless Chrome — no external PDF library needed. Use this whenever the user asks for a "PDF report", "informe en PDF", "documento para el cliente", "PDF con nuestra marca/logo", or wants an existing HTML dashboard/artifact turned into a proper paginated document instead of a browser screenshot or a raw print of the dashboard itself. Also use when asked to add a persistent logo/header to every page of a PDF, or to regenerate/update a PDF that was built this way before. Works for ANY brand/client — always run the brand intake step first, never assume a specific client's colors or logo.
educlopez/mallard · ★ 3 · Data & Documents · score 72

Install: claude install-skill educlopez/mallard

# Paginated report PDF (HTML → Chrome headless → PDF) Produces a real, linear document (cover, numbered sections, per-item pages) rendered by headless Chrome from a self-contained HTML file — not a screenshot, not a raw print of an interactive dashboard. Brand-agnostic: the skill asks for the brand's visual identity before writing a single line of HTML, so the same instructions produce a correctly-branded PDF for whichever client/project you're in. ## When to use this vs. other options - **Use this** when the deliverable is a real *document*: cover page, sections, tables, a page per item, meant to be read start-to-finish or archived/emailed as a PDF. - **Don't use this** to just "export" an existing interactive dashboard (tabs, sidebar nav) as-is — that produces a broken, unreadable PDF (nav elements repeated, tab content hidden). Write a *separate*, purpose-built paginated HTML file instead. It's fine — even expected — for this file to reuse content/wording from the dashboard, restructured into linear document flow. - Never reach for a heavier toolchain (wkhtmltopdf, Puppeteer, a PDF library) unless headless Chrome is confirmed unavailable — it almost always already is (see prerequisite check below). If the task is PDF *manipulation* (merge/split/extract/forms/OCR) rather than *authoring a new document*, that's a different job — don't force this skill onto it. ## Step 0 — brand intake (always do this first) Before writing any HTML, establish the bran