pdfs

name	pdfs
description	Read, create, and review PDFs
metadata	{"displayName":"PDF Master","icon":"doc.richtext.fill","color":"red","placeholder":"Upload a PDF or describe what you need"}

PDF reading, creation, and review guidance

Reading PDFs

• Use pdftoppm -png $OUTDIR/$BASENAME.pdf $OUTDIR/$BASENAME to convert PDFs to PNGs.
• Then open the PNGs and read the images.
• pdfplumber is also installed and can be used to read PDFs. It can be used as a complementary tool to pdftoppm but not replacing it.
• Only do python printing as a last resort because you will miss important details with text extraction (e.g. figures, tables, diagrams).

Primary tooling for creating PDFs

• Generate PDFs programmatically with reportlab as the primary tool. In most cases, you should use reportlab to create PDFs.
• If there are other packages you think are necessary for the task (eg. pypdf, pyMuPDF), you can use them but you may need to pip install them first.

Font support for non-ASCII characters

• Before creating PDFs with Chinese, Japanese, Korean (CJK) or other non-ASCII characters, install the required fonts:

  apt-get update && apt-get install -y fonts-noto-cjk fonts-noto-core fonts-noto-ui-core
  

• After installing fonts, rebuild the font cache: fc-cache -fv
• Verify fonts are available: fc-list :lang=zh (for Chinese)
• After each meaningful update—content additions, layout adjustments, or style changes—render the PDF to images to check layout fidelity:
  • pdftoppm -png $INPUT_PDF $OUTPUT_PREFIX
• Inspect every exported PNG before continuing work. If anything looks off, fix the source and re-run the render → inspect loop until the pages are clean.

Quality expectations

• Maintain a polished, intentional visual design: consistent typography, spacing, margins, color palette, and clear section breaks across all pages.
• Avoid major rendering issues—no clipped text, overlapping elements, black squares, broken tables, or unreadable glyphs. The rendered pages should look like a curated document, not raw template output.
• Charts, tables, diagrams, and images must be sharp, well-aligned, and properly labeled in the PNGs. Legends and axes should be readable without excessive zoom.
• Text must be readable at normal viewing size; avoid walls of filler text or dense, unstructured bullet lists. Use whitespace to separate ideas.
• Never use the U+2011 non-breaking hyphen or other unicode dashes as they will not be rendered correctly. Use ASCII hyphens instead.
• Citations, references, and footnotes must be human-readable and professional. No tool-internal tokens (e.g., [145036110387964†L158-L160]), malformed URLs, or placeholder text should be present in the document.
• You must convert all citations into a human-readable format in the document with standard scholarly citation format. No 【【turn1541736113682297662view0†L11-L19】 notations are allowed in the document as the reader cannot interpret them (such citations will be severely penalized).
• Correct every formatting defect you see in the PNGs, including but not limited to: overlapping text or shapes, clipped text or shapes that are cut off, black squares, broken tables, unreadable characters, etc.

Final checks

• Do not ship the PDF until the latest PNG inspection shows zero visual or formatting defects. Confirm page numbering, headers/footers, and section transitions look polished.
• You must resolve any issues you spot (typos, spacing, alignment, color contrast) and re-render for a final verification pass before submitting.
• Keep intermediate PNGs/PDFs clearly labeled or cleaned up so reviewers can navigate the outputs easily.