PDF Document Processing

Read PDFs with real structural understanding (Docling) and create professional PDFs (fpdf2), 100% locally. .pdf files are binary — never open one with a text/Read tool; use inspect_pdf.py / read_pdf.py to see inside, and write_pdf.py to author new ones.

flowchart LR
  subgraph read [Read]
    Pdf["some.pdf"] --> Insp["inspect_pdf.py<br/>structure summary"]
    Pdf --> Rd["read_pdf.py<br/>JSON / md / text / html"]
  end
  subgraph write [Write]
    Cfg["config.json<br/>(content blocks)"] --> Wr["write_pdf.py"] --> Out["report.pdf"]
  end

Prerequisites