pdfx is a fast, privacy-friendly command-line tool for working with PDFs, images, and documents entirely on your machine.

No uploads.
No tracking.
No shady websites.

Just one binary and predictable commands.

Every time you fill a form or submit documents, you end up needing to:

• merge PDFs
• delete or reorder pages
• compress files under size limits
• convert images or documents to PDF

The usual solution?
Uploading sensitive documents (IDs, contracts, resumes) to random websites.

pdfx exists to fix this.

pdfx is a local alternative to tools like SmallPDF / iLovePDF —
but transparent, scriptable, and privacy-first.

• 🔐 Local-first — all processing happens on your machine
• 🚫 No cloud, no telemetry, no tracking
• ⚡ Fast & scriptable — designed for CLI workflows
• 📦 Single Go binary — easy to distribute
• 🧩 Explicit dependencies — no hidden magic

• Merge multiple PDFs
• Delete pages (true delete, inclusive ranges)
• Split PDFs into pages
• Extract selected pages
• Compress PDFs (3 quality levels)

• Convert images → PDF
• Compress images (good / better / best)
• Convert image formats (jpg ↔ png)

• .docx, .pptx, .xlsx → PDF (LibreOffice)
• .txt, .md, .html → PDF (Pandoc)

• Built-in doctor command to verify dependencies

First install the external dependency for sanity check use pdfx doctor then install the binary.

Linux

sudo apt install ghostscript pandoc libreoffice

MacOS

brew install ghostscript pandoc
brew install --cask libreoffice

Download the latest release from GitHub Releases and place it in your PATH.

chmod +x pdfx
mv pdfx /usr/local/bin/

git clone https://github.com/yourusername/pdfx
cd pdfx
go build

pdfx merge a.pdf b.pdf c.pdf -o merged.pdf

pdfx delete report.pdf --pages 1,3-5,n -o cleaned.pdf

Deletes page 1, pages 3-5, and the last page (n).

pdfx split report.pdf --pages 1-3

Creates: report_page_1.pdf, report_page_2.pdf, report_page_3.pdf

pdfx compress large.pdf --level best -o small.pdf

Compression levels:

• good — High quality, moderate compression
• better — Balanced (default)
• best — Smallest file size

pdfx img2pdf *.jpg -o images.pdf

Supported formats: .jpg, .jpeg, .png, .webp, .bmp, .tif, .tiff

pdfx convert resume.docx -o resume.pdf
pdfx convert notes.md -o notes.pdf

pdfx img-compress photo.jpg --level best -o photo_small.jpg

pdfx img-convert image.webp -o image.png

Used by delete, split, and extract commands:

⚠️ Note: 1-n (delete all pages) is not allowed in the delete command.

pdfx doctor

Verifies that external tools are installed and available in your PATH.

pdfx merge <pdf1> <pdf2> [pdf3 ...] -o <output.pdf>

Flags:

• -o, --output — Output PDF file (default: merged.pdf)

pdfx delete <input.pdf> --pages <ranges> -o <output.pdf>

Flags:

• --pages — Pages to delete (required)
• -o, --output — Output PDF

Examples:

pdfx delete file.pdf --pages 3 -o out.pdf
pdfx delete file.pdf --pages 1,4-6,n -o cleaned.pdf

pdfx split <input.pdf> --pages <ranges>

Extracts selected pages into individual PDF files.

Example:

pdfx split report.pdf --pages 1-3

pdfx extract <input.pdf> --pages <ranges> -o <output.pdf>

Example:

pdfx extract report.pdf --pages 2,4-6 -o excerpt.pdf

pdfx img2pdf <image1> <image2> [...] -o <output.pdf>

Examples:

pdfx img2pdf a.jpg b.png -o images.pdf
pdfx img2pdf *.jpg -o album.pdf

pdfx compress <input.pdf> --level <good|better|best> -o <output.pdf>

Flags:

• --level — Compression level (default: better)
• -o, --output — Output PDF (default: compressed.pdf)

pdfx convert <input.file> -o <output.pdf>

Supported formats:

• Text: .txt, .md, .html (via Pandoc)
• Office: .docx, .pptx, .xlsx, .odt (via LibreOffice)

pdfx img-compress <image> --level <good|better|best> -o <output>

Supported formats:

• Input: .jpg, .jpeg, .png, .webp
• Output: .jpg, .png

pdfx img-convert <image> -o <output>

Supported output: .jpg, .png

Every command has built-in help:

pdfx --help
pdfx delete --help
pdfx compress --help

• ✅ No internet access
• ✅ No uploads
• ✅ No data collection
• ✅ No telemetry
• ✅ No background services

You own your data. Always.

• Page reordering
• Batch operations
• Config file support
• Homebrew / Scoop packages
• Windows-friendly installers

MIT License — use it, fork it, improve it.

This project was built to be useful first, clever second.

If it saved you time or protected your privacy —
⭐ star the repo, or better, tell someone about it.

Happy hacking 🚀